16 Commits
Author SHA1 Message Date
vcoppe c0e6e3d2eb fix release step 2026-11-01 17:43:28 +01:00
vcoppe 9bd38cd5f8 fix tests 2026-11-01 17:24:49 +01:00
vcoppe a163e1b897 New Crowdin updates (#404)
* New translations en.json (Dutch)

[ci skip]

* New translations files-and-stats.mdx (Romanian)

[ci skip]

* New translations files-and-stats.mdx (French)

[ci skip]

* New translations files-and-stats.mdx (Spanish)

[ci skip]

* New translations files-and-stats.mdx (Belarusian)

[ci skip]

* New translations files-and-stats.mdx (Catalan)

[ci skip]

* New translations files-and-stats.mdx (Czech)

[ci skip]

* New translations files-and-stats.mdx (Danish)

[ci skip]

* New translations files-and-stats.mdx (German)

[ci skip]

* New translations files-and-stats.mdx (Greek)

[ci skip]

* New translations files-and-stats.mdx (Basque)

[ci skip]

* New translations files-and-stats.mdx (Finnish)

[ci skip]

* New translations files-and-stats.mdx (Hebrew)

[ci skip]

* New translations files-and-stats.mdx (Hungarian)

[ci skip]

* New translations files-and-stats.mdx (Italian)

[ci skip]

* New translations files-and-stats.mdx (Korean)

[ci skip]

* New translations files-and-stats.mdx (Lithuanian)

[ci skip]

* New translations files-and-stats.mdx (Dutch)

[ci skip]

* New translations files-and-stats.mdx (Norwegian)

[ci skip]

* New translations files-and-stats.mdx (Polish)

[ci skip]

* New translations files-and-stats.mdx (Portuguese)

[ci skip]

* New translations files-and-stats.mdx (Russian)

[ci skip]

* New translations files-and-stats.mdx (Swedish)

[ci skip]

* New translations files-and-stats.mdx (Turkish)

[ci skip]

* New translations files-and-stats.mdx (Ukrainian)

[ci skip]

* New translations files-and-stats.mdx (Chinese Simplified)

[ci skip]

* New translations files-and-stats.mdx (Vietnamese)

[ci skip]

* New translations files-and-stats.mdx (Portuguese, Brazilian)

[ci skip]

* New translations files-and-stats.mdx (Indonesian)

[ci skip]

* New translations files-and-stats.mdx (Thai)

[ci skip]

* New translations files-and-stats.mdx (Latvian)

[ci skip]

* New translations files-and-stats.mdx (Chinese Traditional, Hong Kong)

[ci skip]

* New translations files-and-stats.mdx (Serbian (Latin))

[ci skip]

* New translations gpx.mdx (Dutch)

[ci skip]

* New translations map-controls.mdx (Basque)

[ci skip]

* New translations edit.mdx (Dutch)

[ci skip]
2026-11-01 17:18:56 +01:00
vcoppe 33abec525a remove unused key in locales 2026-11-01 17:00:12 +01:00
vcoppe 77c542d9aa fix windows build continued 2026-11-01 16:52:36 +01:00
vcoppe 17b5c1d567 fix windows build 2026-11-01 16:07:49 +01:00
vcoppe 2048076fe6 tauri build test 2026-11-01 15:52:34 +01:00
vcoppe d7ff7d645b New Crowdin updates (#403)
* New translations edit.mdx (Lithuanian)

[ci skip]

* New translations edit.mdx (Dutch)

[ci skip]

* New translations edit.mdx (Norwegian)

[ci skip]

* New translations edit.mdx (Polish)

[ci skip]

* New translations edit.mdx (Portuguese)

[ci skip]

* New translations edit.mdx (Russian)

[ci skip]

* New translations edit.mdx (Swedish)

[ci skip]

* New translations edit.mdx (Turkish)

[ci skip]

* New translations edit.mdx (Ukrainian)

[ci skip]

* New translations edit.mdx (Chinese Simplified)

[ci skip]

* New translations edit.mdx (Vietnamese)

[ci skip]

* New translations edit.mdx (Portuguese, Brazilian)

[ci skip]

* New translations edit.mdx (Indonesian)

[ci skip]

* New translations edit.mdx (Thai)

[ci skip]

* New translations edit.mdx (Latvian)

[ci skip]

* New translations edit.mdx (Chinese Traditional, Hong Kong)

[ci skip]

* New translations edit.mdx (Serbian (Latin))

[ci skip]

* New translations file.mdx (Romanian)

[ci skip]

* New translations file.mdx (French)

[ci skip]

* New translations file.mdx (Spanish)

[ci skip]

* New translations file.mdx (Belarusian)

[ci skip]

* New translations file.mdx (Catalan)

[ci skip]

* New translations file.mdx (Czech)

[ci skip]

* New translations file.mdx (Danish)

[ci skip]

* New translations file.mdx (German)

[ci skip]

* New translations file.mdx (Greek)

[ci skip]

* New translations file.mdx (Basque)

[ci skip]

* New translations file.mdx (Finnish)

[ci skip]

* New translations file.mdx (Hebrew)

[ci skip]

* New translations file.mdx (Hungarian)

[ci skip]

* New translations file.mdx (Italian)

[ci skip]

* New translations file.mdx (Korean)

[ci skip]

* New translations file.mdx (Lithuanian)

[ci skip]

* New translations file.mdx (Dutch)

[ci skip]

* New translations file.mdx (Norwegian)

[ci skip]

* New translations file.mdx (Polish)

[ci skip]

* New translations file.mdx (Portuguese)

[ci skip]

* New translations file.mdx (Russian)

[ci skip]

* New translations file.mdx (Swedish)

[ci skip]

* New translations file.mdx (Turkish)

[ci skip]

* New translations file.mdx (Ukrainian)

[ci skip]

* New translations file.mdx (Chinese Simplified)

[ci skip]

* New translations file.mdx (Vietnamese)

[ci skip]

* New translations file.mdx (Portuguese, Brazilian)

[ci skip]

* New translations file.mdx (Indonesian)

[ci skip]

* New translations file.mdx (Thai)

[ci skip]

* New translations file.mdx (Latvian)

[ci skip]

* New translations file.mdx (Chinese Traditional, Hong Kong)

[ci skip]

* New translations file.mdx (Serbian (Latin))

[ci skip]

* New translations view.mdx (Romanian)

[ci skip]

* New translations view.mdx (French)

[ci skip]

* New translations view.mdx (Spanish)

[ci skip]

* New translations view.mdx (Belarusian)

[ci skip]

* New translations view.mdx (Catalan)

[ci skip]

* New translations view.mdx (Czech)

[ci skip]

* New translations view.mdx (Danish)

[ci skip]

* New translations view.mdx (German)

[ci skip]

* New translations view.mdx (Greek)

[ci skip]

* New translations view.mdx (Basque)

[ci skip]

* New translations view.mdx (Finnish)

[ci skip]

* New translations view.mdx (Hebrew)

[ci skip]

* New translations view.mdx (Hungarian)

[ci skip]

* New translations view.mdx (Italian)

[ci skip]

* New translations view.mdx (Korean)

[ci skip]

* New translations view.mdx (Lithuanian)

[ci skip]

* New translations view.mdx (Dutch)

[ci skip]

* New translations view.mdx (Norwegian)

[ci skip]

* New translations view.mdx (Polish)

[ci skip]

* New translations view.mdx (Portuguese)

[ci skip]

* New translations view.mdx (Russian)

[ci skip]

* New translations view.mdx (Swedish)

[ci skip]

* New translations view.mdx (Turkish)

[ci skip]

* New translations view.mdx (Ukrainian)

[ci skip]

* New translations view.mdx (Chinese Simplified)

[ci skip]

* New translations view.mdx (Vietnamese)

[ci skip]

* New translations view.mdx (Portuguese, Brazilian)

[ci skip]

* New translations view.mdx (Indonesian)

[ci skip]

* New translations view.mdx (Thai)

[ci skip]

* New translations view.mdx (Latvian)

[ci skip]

* New translations view.mdx (Chinese Traditional, Hong Kong)

[ci skip]

* New translations view.mdx (Serbian (Latin))

[ci skip]

* New translations extract.mdx (Romanian)

[ci skip]

* New translations extract.mdx (French)

[ci skip]

* New translations extract.mdx (Spanish)

[ci skip]

* New translations extract.mdx (Belarusian)

[ci skip]

* New translations extract.mdx (Catalan)

[ci skip]

* New translations extract.mdx (Czech)

[ci skip]

* New translations extract.mdx (Danish)

[ci skip]

* New translations extract.mdx (German)

[ci skip]

* New translations extract.mdx (Greek)

[ci skip]

* New translations extract.mdx (Basque)

[ci skip]

* New translations extract.mdx (Finnish)

[ci skip]

* New translations extract.mdx (Hebrew)

[ci skip]

* New translations extract.mdx (Hungarian)

[ci skip]

* New translations extract.mdx (Italian)

[ci skip]

* New translations extract.mdx (Korean)

[ci skip]

* New translations extract.mdx (Lithuanian)

[ci skip]

* New translations extract.mdx (Dutch)

[ci skip]

* New translations extract.mdx (Norwegian)

[ci skip]

* New translations extract.mdx (Polish)

[ci skip]

* New translations extract.mdx (Portuguese)

[ci skip]

* New translations extract.mdx (Russian)

[ci skip]

* New translations extract.mdx (Swedish)

[ci skip]

* New translations extract.mdx (Turkish)

[ci skip]

* New translations extract.mdx (Ukrainian)

[ci skip]

* New translations extract.mdx (Chinese Simplified)

[ci skip]

* New translations extract.mdx (Vietnamese)

[ci skip]

* New translations extract.mdx (Portuguese, Brazilian)

[ci skip]

* New translations extract.mdx (Indonesian)

[ci skip]

* New translations extract.mdx (Thai)

[ci skip]

* New translations extract.mdx (Latvian)

[ci skip]

* New translations extract.mdx (Chinese Traditional, Hong Kong)

[ci skip]

* New translations extract.mdx (Serbian (Latin))

[ci skip]

* New translations merge.mdx (Romanian)

[ci skip]

* New translations merge.mdx (French)

[ci skip]

* New translations merge.mdx (Spanish)

[ci skip]

* New translations merge.mdx (Belarusian)

[ci skip]

* New translations merge.mdx (Catalan)

[ci skip]

* New translations merge.mdx (Czech)

[ci skip]

* New translations merge.mdx (Danish)

[ci skip]

* New translations merge.mdx (German)

[ci skip]

* New translations merge.mdx (Greek)

[ci skip]

* New translations merge.mdx (Basque)

[ci skip]

* New translations merge.mdx (Finnish)

[ci skip]

* New translations merge.mdx (Hebrew)

[ci skip]

* New translations merge.mdx (Hungarian)

[ci skip]

* New translations merge.mdx (Italian)

[ci skip]

* New translations merge.mdx (Korean)

[ci skip]

* New translations merge.mdx (Lithuanian)

[ci skip]

* New translations merge.mdx (Dutch)

[ci skip]

* New translations merge.mdx (Norwegian)

[ci skip]

* New translations merge.mdx (Polish)

[ci skip]

* New translations merge.mdx (Portuguese)

[ci skip]

* New translations merge.mdx (Russian)

[ci skip]

* New translations merge.mdx (Swedish)

[ci skip]

* New translations merge.mdx (Turkish)

[ci skip]

* New translations merge.mdx (Ukrainian)

[ci skip]

* New translations merge.mdx (Chinese Simplified)

[ci skip]

* New translations merge.mdx (Vietnamese)

[ci skip]

* New translations merge.mdx (Portuguese, Brazilian)

[ci skip]

* New translations merge.mdx (Indonesian)

[ci skip]

* New translations merge.mdx (Thai)

[ci skip]

* New translations merge.mdx (Latvian)

[ci skip]

* New translations merge.mdx (Chinese Traditional, Hong Kong)

[ci skip]

* New translations merge.mdx (Serbian (Latin))

[ci skip]

* New translations scissors.mdx (Romanian)

[ci skip]

* New translations scissors.mdx (French)

[ci skip]

* New translations scissors.mdx (Spanish)

[ci skip]

* New translations scissors.mdx (Belarusian)

[ci skip]

* New translations scissors.mdx (Catalan)

[ci skip]

* New translations scissors.mdx (Czech)

[ci skip]

* New translations scissors.mdx (Danish)

[ci skip]

* New translations scissors.mdx (German)

[ci skip]

* New translations scissors.mdx (Greek)

[ci skip]

* New translations scissors.mdx (Basque)

[ci skip]

* New translations scissors.mdx (Finnish)

[ci skip]

* New translations scissors.mdx (Hebrew)

[ci skip]

* New translations scissors.mdx (Hungarian)

[ci skip]

* New translations scissors.mdx (Italian)

[ci skip]

* New translations scissors.mdx (Korean)

[ci skip]

* New translations scissors.mdx (Lithuanian)

[ci skip]

* New translations scissors.mdx (Dutch)

[ci skip]

* New translations scissors.mdx (Norwegian)

[ci skip]

* New translations scissors.mdx (Polish)

[ci skip]

* New translations scissors.mdx (Portuguese)

[ci skip]

* New translations scissors.mdx (Russian)

[ci skip]

* New translations scissors.mdx (Swedish)

[ci skip]

* New translations scissors.mdx (Turkish)

[ci skip]

* New translations scissors.mdx (Ukrainian)

[ci skip]

* New translations scissors.mdx (Chinese Simplified)

[ci skip]

* New translations scissors.mdx (Vietnamese)

[ci skip]

* New translations scissors.mdx (Portuguese, Brazilian)

[ci skip]

* New translations scissors.mdx (Indonesian)

[ci skip]

* New translations scissors.mdx (Thai)

[ci skip]

* New translations scissors.mdx (Latvian)

[ci skip]

* New translations scissors.mdx (Chinese Traditional, Hong Kong)

[ci skip]

* New translations scissors.mdx (Serbian (Latin))

[ci skip]

* New translations en.json (French)

[ci skip]

* New translations edit.mdx (Romanian)

[ci skip]

* New translations edit.mdx (French)

[ci skip]

* New translations edit.mdx (Spanish)

[ci skip]

* New translations edit.mdx (Belarusian)

[ci skip]

* New translations edit.mdx (Catalan)

[ci skip]

* New translations edit.mdx (Czech)

[ci skip]

* New translations edit.mdx (Danish)

[ci skip]

* New translations edit.mdx (German)

[ci skip]

* New translations edit.mdx (Greek)

[ci skip]

* New translations edit.mdx (Basque)

[ci skip]

* New translations edit.mdx (Finnish)

[ci skip]

* New translations edit.mdx (Hebrew)

[ci skip]

* New translations edit.mdx (Hungarian)

[ci skip]

* New translations edit.mdx (Italian)

[ci skip]

* New translations edit.mdx (Korean)

[ci skip]

* New translations edit.mdx (Lithuanian)

[ci skip]

* New translations edit.mdx (Dutch)

[ci skip]

* New translations edit.mdx (Norwegian)

[ci skip]

* New translations edit.mdx (Polish)

[ci skip]

* New translations edit.mdx (Portuguese)

[ci skip]

* New translations edit.mdx (Russian)

[ci skip]

* New translations edit.mdx (Swedish)

[ci skip]

* New translations edit.mdx (Turkish)

[ci skip]

* New translations edit.mdx (Ukrainian)

[ci skip]

* New translations edit.mdx (Chinese Simplified)

[ci skip]

* New translations edit.mdx (Vietnamese)

[ci skip]

* New translations edit.mdx (Portuguese, Brazilian)

[ci skip]

* New translations edit.mdx (Indonesian)

[ci skip]

* New translations edit.mdx (Thai)

[ci skip]

* New translations edit.mdx (Latvian)

[ci skip]

* New translations edit.mdx (Chinese Traditional, Hong Kong)

[ci skip]

* New translations edit.mdx (Serbian (Latin))

[ci skip]

* New translations file.mdx (French)

[ci skip]

* New translations view.mdx (French)

[ci skip]

* New translations extract.mdx (French)

[ci skip]

* New translations merge.mdx (French)

[ci skip]

* New translations scissors.mdx (French)

[ci skip]

* New translations files-and-stats.mdx (French)

[ci skip]

* New translations gpx.mdx (French)

[ci skip]

* New translations edit.mdx (Romanian)

[ci skip]

* New translations edit.mdx (French)

[ci skip]

* New translations edit.mdx (Spanish)

[ci skip]

* New translations edit.mdx (Belarusian)

[ci skip]

* New translations edit.mdx (Catalan)

[ci skip]

* New translations edit.mdx (Czech)

[ci skip]

* New translations edit.mdx (Danish)

[ci skip]

* New translations edit.mdx (German)

[ci skip]

* New translations edit.mdx (Greek)

[ci skip]

* New translations edit.mdx (Basque)

[ci skip]

* New translations edit.mdx (Finnish)

[ci skip]

* New translations edit.mdx (Hebrew)

[ci skip]

* New translations edit.mdx (Hungarian)

[ci skip]

* New translations edit.mdx (Italian)

[ci skip]

* New translations edit.mdx (Korean)

[ci skip]

* New translations edit.mdx (Lithuanian)

[ci skip]

* New translations edit.mdx (Dutch)

[ci skip]

* New translations edit.mdx (Norwegian)

[ci skip]

* New translations edit.mdx (Polish)

[ci skip]

* New translations edit.mdx (Portuguese)

[ci skip]

* New translations edit.mdx (Russian)

[ci skip]

* New translations edit.mdx (Swedish)

[ci skip]

* New translations edit.mdx (Turkish)

[ci skip]

* New translations edit.mdx (Ukrainian)

[ci skip]

* New translations edit.mdx (Chinese Simplified)

[ci skip]

* New translations edit.mdx (Vietnamese)

[ci skip]

* New translations edit.mdx (Portuguese, Brazilian)

[ci skip]

* New translations edit.mdx (Indonesian)

[ci skip]

* New translations edit.mdx (Thai)

[ci skip]

* New translations edit.mdx (Latvian)

[ci skip]

* New translations edit.mdx (Chinese Traditional, Hong Kong)

[ci skip]

* New translations edit.mdx (Serbian (Latin))

[ci skip]

* New translations en.json (French)

[ci skip]
2026-11-01 15:41:07 +01:00
vcoppe 262b590bd4 remove note 2026-11-01 15:04:24 +01:00
vcoppe 14bf9cddd2 Merge branch 'dev' of https://github.com/gpxstudio/gpx.studio into dev 2026-11-01 14:51:44 +01:00
vcoppe b9566160e5 rephrasing 2026-11-01 14:51:42 +01:00
vcoppe 0cc9d6f664 New Crowdin updates (#402)
* New translations en.json (Polish)

[ci skip]

* New translations en.json (Romanian)

[ci skip]

* New translations en.json (French)

[ci skip]

* New translations en.json (Spanish)

[ci skip]

* New translations en.json (Belarusian)

[ci skip]

* New translations en.json (Catalan)

[ci skip]

* New translations en.json (Czech)

[ci skip]

* New translations en.json (Danish)

[ci skip]

* New translations en.json (German)

[ci skip]

* New translations en.json (Greek)

[ci skip]

* New translations en.json (Basque)

[ci skip]

* New translations en.json (Finnish)

[ci skip]

* New translations en.json (Hebrew)

[ci skip]

* New translations en.json (Italian)

[ci skip]

* New translations en.json (Korean)

[ci skip]

* New translations en.json (Lithuanian)

[ci skip]

* New translations en.json (Dutch)

[ci skip]

* New translations en.json (Norwegian)

[ci skip]

* New translations en.json (Portuguese)

[ci skip]

* New translations en.json (Russian)

[ci skip]

* New translations en.json (Swedish)

[ci skip]

* New translations en.json (Turkish)

[ci skip]

* New translations en.json (Ukrainian)

[ci skip]

* New translations en.json (Chinese Simplified)

[ci skip]

* New translations en.json (Vietnamese)

[ci skip]

* New translations en.json (Portuguese, Brazilian)

[ci skip]

* New translations en.json (Indonesian)

[ci skip]

* New translations en.json (Thai)

[ci skip]

* New translations en.json (Latvian)

[ci skip]

* New translations en.json (Chinese Traditional, Hong Kong)

[ci skip]

* New translations en.json (Serbian (Latin))

[ci skip]

* New translations files-and-stats.mdx (Romanian)

[ci skip]

* New translations files-and-stats.mdx (French)

[ci skip]

* New translations files-and-stats.mdx (Spanish)

[ci skip]

* New translations files-and-stats.mdx (Belarusian)

[ci skip]

* New translations files-and-stats.mdx (Catalan)

[ci skip]

* New translations files-and-stats.mdx (Czech)

[ci skip]

* New translations files-and-stats.mdx (Danish)

[ci skip]

* New translations files-and-stats.mdx (German)

[ci skip]

* New translations files-and-stats.mdx (Greek)

[ci skip]

* New translations files-and-stats.mdx (Basque)

[ci skip]

* New translations files-and-stats.mdx (Finnish)

[ci skip]

* New translations files-and-stats.mdx (Hebrew)

[ci skip]

* New translations files-and-stats.mdx (Hungarian)

[ci skip]

* New translations files-and-stats.mdx (Italian)

[ci skip]

* New translations files-and-stats.mdx (Korean)

[ci skip]

* New translations files-and-stats.mdx (Lithuanian)

[ci skip]

* New translations files-and-stats.mdx (Dutch)

[ci skip]

* New translations files-and-stats.mdx (Norwegian)

[ci skip]

* New translations files-and-stats.mdx (Polish)

[ci skip]

* New translations files-and-stats.mdx (Portuguese)

[ci skip]

* New translations files-and-stats.mdx (Russian)

[ci skip]

* New translations files-and-stats.mdx (Swedish)

[ci skip]

* New translations files-and-stats.mdx (Turkish)

[ci skip]

* New translations files-and-stats.mdx (Ukrainian)

[ci skip]

* New translations files-and-stats.mdx (Chinese Simplified)

[ci skip]

* New translations files-and-stats.mdx (Vietnamese)

[ci skip]

* New translations files-and-stats.mdx (Portuguese, Brazilian)

[ci skip]

* New translations files-and-stats.mdx (Indonesian)

[ci skip]

* New translations files-and-stats.mdx (Thai)

[ci skip]

* New translations files-and-stats.mdx (Latvian)

[ci skip]

* New translations files-and-stats.mdx (Chinese Traditional, Hong Kong)

[ci skip]

* New translations files-and-stats.mdx (Serbian (Latin))

[ci skip]
2026-10-31 18:53:31 +01:00
vcoppe 1942e0b9e1 New Crowdin updates (#376)
* New translations en.json (Spanish)

[ci skip]

* New translations elevation.mdx (German)

[ci skip]

* New translations integration.mdx (Basque)

[ci skip]

* New translations map-controls.mdx (Basque)

[ci skip]

* New translations view.mdx (Basque)

[ci skip]

* New translations merge.mdx (Basque)

[ci skip]

* New translations elevation.mdx (Basque)

[ci skip]

* New translations en.json (Dutch)

[ci skip]

* New translations elevation.mdx (Dutch)

[ci skip]

* New translations routing.mdx (Russian)

[ci skip]

* New translations files-and-stats.mdx (Chinese Traditional, Hong Kong)

[ci skip]

* New translations settings.mdx (Chinese Traditional, Hong Kong)

[ci skip]

* New translations view.mdx (Chinese Traditional, Hong Kong)

[ci skip]

* New translations en.json (Portuguese)

[ci skip]

* New translations en.json (Portuguese)

[ci skip]

* New translations edit.mdx (Russian)

[ci skip]

* New translations en.json (Portuguese)

[ci skip]

* New translations files-and-stats.mdx (Portuguese)

[ci skip]

* New translations getting-started.mdx (Portuguese)

[ci skip]

* New translations integration.mdx (Portuguese)

[ci skip]

* New translations map-controls.mdx (Portuguese)

[ci skip]

* New translations file.mdx (Portuguese)

[ci skip]

* New translations view.mdx (Portuguese)

[ci skip]

* New translations clean.mdx (Portuguese)

[ci skip]

* New translations extract.mdx (Portuguese)

[ci skip]

* New translations merge.mdx (Portuguese)

[ci skip]

* New translations minify.mdx (Portuguese)

[ci skip]

* New translations poi.mdx (Portuguese)

[ci skip]

* New translations routing.mdx (Portuguese)

[ci skip]

* New translations scissors.mdx (Portuguese)

[ci skip]

* New translations time.mdx (Portuguese)

[ci skip]

* New translations faq.mdx (Portuguese)

[ci skip]

* New translations elevation.mdx (Portuguese)

[ci skip]

* New translations en.json (Portuguese)

[ci skip]

* New translations en.json (Portuguese)

[ci skip]

* New translations files-and-stats.mdx (Portuguese)

[ci skip]

* New translations getting-started.mdx (Portuguese)

[ci skip]

* New translations edit.mdx (Portuguese)

[ci skip]

* New translations file.mdx (Russian)

[ci skip]

* New translations settings.mdx (Russian)

[ci skip]

* New translations view.mdx (Russian)

[ci skip]

* New translations settings.mdx (Russian)

[ci skip]

* New translations routing.mdx (Russian)

[ci skip]

* New translations en.json (Hungarian)

[ci skip]

* New translations en.json (Hungarian)

[ci skip]

* New translations edit.mdx (Hungarian)

[ci skip]

* New translations file.mdx (Hungarian)

[ci skip]

* New translations view.mdx (Hungarian)

[ci skip]

* New translations view.mdx (Hungarian)

[ci skip]

* New translations clean.mdx (Hungarian)

[ci skip]

* New translations merge.mdx (Hungarian)

[ci skip]

* New translations elevation.mdx (Hungarian)

[ci skip]

* New translations en.json (Hungarian)

[ci skip]

* New translations merge.mdx (Hungarian)

[ci skip]

* New translations getting-started.mdx (Hungarian)

[ci skip]

* New translations settings.mdx (Hungarian)

[ci skip]

* New translations minify.mdx (Hungarian)

[ci skip]

* New translations routing.mdx (Hungarian)

[ci skip]

* New translations scissors.mdx (Hungarian)

[ci skip]

* New translations faq.mdx (Hungarian)

[ci skip]

* New translations en.json (Hungarian)

[ci skip]

* New translations view.mdx (Hungarian)

[ci skip]

* New translations faq.mdx (Hungarian)

[ci skip]

* New translations files-and-stats.mdx (Hungarian)

[ci skip]

* New translations gpx.mdx (Hungarian)

[ci skip]

* New translations integration.mdx (Hungarian)

[ci skip]

* New translations map-controls.mdx (Hungarian)

[ci skip]

* New translations en.json (Polish)

[ci skip]

* New translations settings.mdx (Polish)

[ci skip]

* New translations routing.mdx (Polish)

[ci skip]

* New translations files-and-stats.mdx (Chinese Simplified)

[ci skip]

* New translations getting-started.mdx (Chinese Simplified)

[ci skip]

* New translations gpx.mdx (Chinese Simplified)

[ci skip]

* New translations map-controls.mdx (Chinese Simplified)

[ci skip]

* New translations edit.mdx (Chinese Simplified)

[ci skip]

* New translations edit.mdx (Chinese Traditional, Hong Kong)

[ci skip]

* New translations settings.mdx (Chinese Simplified)

[ci skip]

* New translations view.mdx (Chinese Simplified)

[ci skip]

* New translations view.mdx (Chinese Traditional, Hong Kong)

[ci skip]

* New translations toolbar.mdx (Chinese Simplified)

[ci skip]

* New translations extract.mdx (Chinese Simplified)

[ci skip]

* New translations merge.mdx (Chinese Simplified)

[ci skip]

* New translations minify.mdx (Chinese Simplified)

[ci skip]

* New translations poi.mdx (Chinese Simplified)

[ci skip]

* New translations routing.mdx (Chinese Simplified)

[ci skip]

* New translations scissors.mdx (Chinese Simplified)

[ci skip]

* New translations time.mdx (Chinese Simplified)

[ci skip]

* New translations faq.mdx (Chinese Simplified)

[ci skip]

* New translations elevation.mdx (Chinese Simplified)

[ci skip]
2026-10-31 18:45:38 +01:00
vcoppe 70a24f53b3 silence warning 2026-10-31 18:33:12 +01:00
vcoppe 001246a49e fix tests 2026-10-31 18:25:48 +01:00
vcoppeandvcoppe 8fe7d8d64d Migration to new Rust-based engine (#401)
* wasm first steps

* parsing start

* parsing continued

* progress

* progress

* progress

* progress

* progress

* progress

* progress

* progress

* fixes

* reorganize

* change type

* fix

* progress

* refactoring

* progress

* progress

* progress

* keep style info

* progress

* progress

* remove (copy) in name

* progress

* merge methods into 1

* splice

* progress

* progress

* progress

* cleaning

* progress

* progress

* progress

* visibility

* progress

* progress

* progress

* progress

* progress

* fixes

* progress

* fix markers

* fix

* progress

* progress

* progress

* progress

* progress

* progress

* small fix

* use engine info

* fix popups

* small fix

* progress

* progress

* progress

* clean tool

* handle self closing elements

* reduce data exchanges across boundary

* copy on wasm side to avoid any data corruption

* speed fix

* routing

* refactoring

* reverse

* reverse

* crop

* split

* merge

* progress

* time tool

* fix bounds for waypoints

* reduce

* fix reactivity

* handle corrupt time data

* export

* progress

* remove gpx js package

* remove todo

* update readme

* cleaning

* progress

* fix link fields

* remove unused fields

* refactoring and tests

* import and export of routes, ref #192

* per-chunk categories storage

* cross-tab sync

* outdated state

* refactoring

* reorder methods

* refactoring

* fix

* improve flow

* renaming

* progress

* progress

* fix

* progress

* per file history

* keep original flag

* use history variant

* fix

* progress

* file explorer

* progress

* fix

* sticky folders

* improved sticky folders

* fix sortable

* fix sortable

* fix

* fix tab size

* no file delete on desktop

* fix routing on other file

* fix sorting

* remember language

* open links in browser

* commit missing changes

* avoid selecting on resize

* clarify filename

* explorer actions

* fix folder styling

* drop anywhere in folder

* cleaning

* fixes

* remove focus-ring

* full screen

* small fix

* consistent highlight color

* improved preview

* performance improvements

* sparser data transfers

* sparser data transfers

* show popup once data is loaded

* fix routing control size

* fix

* sparse desktop build

* remove custom ign styles

* use new endpoint for openhikingmap

* strip native menu

* reorder context menu items

* route full support

* route - track conversion

* create route file action

* route export, filter out anchors at high zoom level

* clarify name handling

* move hide & center to view

* routes continued

* update docs

* fix power format, closes #378

* lint fixes

* fix display of additional data with gaps, closes #361

* fix linting issues

* fix import modules

* fix multidrag

* change base to resolve

* fix type errors

* fix layer typings

* fix last type error

* each keys

* improve typings

* deploy to cloudflare

---------

Co-authored-by: vcoppe <vcoppe@hexaly.com>
2026-10-31 18:08:22 +01:00
518 changed files with 9695 additions and 3797 deletions

No files matched your search

+67
View File
@@ -0,0 +1,67 @@
name: Deploy to Cloudflare Pages
on:
push:
branches: 'dev'
workflow_dispatch:
concurrency:
group: cloudflare-${{ github.ref }}
cancel-in-progress: true
jobs:
deploy:
runs-on: ubuntu-latest
permissions:
contents: read
deployments: write
steps:
- name: Checkout
uses: actions/checkout@v5
- name: Install Node.js
uses: actions/setup-node@v6
with:
node-version: 24
cache: npm
cache-dependency-path: website/package-lock.json
- name: Install Rust
uses: dtolnay/rust-toolchain@stable
with:
targets: wasm32-unknown-unknown
- name: Install wasm-pack
uses: taiki-e/install-action@v2
with:
tool: wasm-pack
- name: Test gpx-rs
run: cargo test
- name: Build wasm
run: wasm-pack build --target bundler
working-directory: wasm
- name: Install dependencies for website
run: npm install --prefix website
- name: Test website
run: npm test --prefix website
- name: Create env file
run: echo PUBLIC_MAPTILER_KEY=${{ secrets.PUBLIC_MAPTILER_KEY }} >> website/.env
- name: Build website
env:
BASE_PATH: ''
run: npm run build --prefix website
- name: Deploy to Cloudflare Pages
uses: cloudflare/wrangler-action@v3
with:
apiToken: ${{ secrets.CLOUDFLARE_API_TOKEN }}
accountId: ${{ secrets.CLOUDFLARE_ACCOUNT_ID }}
command: pages deploy website/build --project-name=${{ vars.CLOUDFLARE_PROJECT_NAME || 'gpxstudio' }} --branch=${{ github.ref_name }} --commit-dirty=true
gitHubToken: ${{ secrets.GITHUB_TOKEN }}
+124
View File
@@ -0,0 +1,124 @@
name: Desktop beta
on:
push:
branches: 'dev'
workflow_dispatch:
concurrency:
group: desktop-beta-${{ github.ref }}
cancel-in-progress: true
jobs:
build:
strategy:
fail-fast: false
matrix:
include:
- platform: macos-latest
rust_targets: aarch64-apple-darwin,x86_64-apple-darwin,wasm32-unknown-unknown
args: --target universal-apple-darwin
bundles: target/universal-apple-darwin/release/bundle/dmg/*.dmg
# the oldest supported Ubuntu, so that the AppImage runs on older systems too
- platform: ubuntu-22.04
rust_targets: wasm32-unknown-unknown
args: ''
bundles: |
target/release/bundle/deb/*.deb
target/release/bundle/rpm/*.rpm
target/release/bundle/appimage/*.AppImage
- platform: windows-latest
rust_targets: wasm32-unknown-unknown
args: ''
bundles: |
target/release/bundle/msi/*.msi
target/release/bundle/nsis/*.exe
runs-on: ${{ matrix.platform }}
steps:
- name: Checkout
uses: actions/checkout@v5
- name: Install Linux dependencies
if: runner.os == 'Linux'
run: |
sudo apt-get update
sudo apt-get install -y libwebkit2gtk-4.1-dev libappindicator3-dev librsvg2-dev patchelf
- name: Install Node.js
uses: actions/setup-node@v6
with:
node-version: 24
cache: npm
cache-dependency-path: website/package-lock.json
- name: Install Rust
uses: dtolnay/rust-toolchain@stable
with:
targets: ${{ matrix.rust_targets }}
- name: Cache Rust build
uses: Swatinem/rust-cache@v2
- name: Install wasm-pack
uses: taiki-e/install-action@v2
with:
tool: wasm-pack
# the desktop app does not use it, but the website depends on its package
- name: Build wasm
run: wasm-pack build --target bundler
working-directory: wasm
- name: Install dependencies for website
run: npm install --prefix website
# the key would be readable in the app: it is left empty, so only the MapTiler layer fails
- name: Create env file
run: echo "PUBLIC_MAPTILER_KEY=" >> website/.env
- name: Build app
run: npx --yes @tauri-apps/cli@2 build ${{ matrix.args }}
working-directory: tauri
# after the build, which gives the app the website it embeds
- name: Test engine and app
run: cargo test -p gpx-engine -p gpx-studio-desktop
- name: Upload bundles
uses: actions/upload-artifact@v4
with:
name: bundles-${{ matrix.platform }}
path: ${{ matrix.bundles }}
if-no-files-found: error
# replaces the beta release only once every platform is built
release:
needs: build
runs-on: ubuntu-latest
permissions:
contents: write
env:
GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
GH_REPO: ${{ github.repository }}
steps:
- name: Download bundles
uses: actions/download-artifact@v5
with:
pattern: bundles-*
path: bundles
merge-multiple: true
- name: Publish beta release
# the bundles of a platform are in a folder per format (deb, msi…)
run: |
mapfile -t files < <(find bundles -type f)
gh release delete desktop-beta --cleanup-tag --yes || true
gh release create desktop-beta "${files[@]}" \
--prerelease \
--target ${{ github.sha }} \
--title "Desktop beta" \
--notes "Beta of the desktop app, built from ${{ github.sha }} on the dev branch. The builds are not signed yet: macOS and Windows will warn before opening them."
+505
View File
@@ -0,0 +1,505 @@
<?xml version="1.0" encoding="UTF-8"?>
<gpx xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xmlns="http://www.topografix.com/GPX/1/1"
xsi:schemaLocation="http://www.topografix.com/GPX/1/1 http://www.topografix.com/GPX/1/1/gpx.xsd
http://www.garmin.com/xmlschemas/GpxExtensions/v3 http://www.garmin.com/xmlschemas/GpxExtensionsv3.xsd
http://www.garmin.com/xmlschemas/TrackPointExtension/v1 http://www.garmin.com/xmlschemas/TrackPointExtensionv1.xsd
http://www.topografix.com/GPX/gpx_style/0/2 http://www.topografix.com/GPX/gpx_style/0/2/gpx_style.xsd"
xmlns:gpxtpx="http://www.garmin.com/xmlschemas/TrackPointExtension/v1"
xmlns:gpxpx="http://www.garmin.com/xmlschemas/PowerExtension/v1"
xmlns:gpx_style="http://www.topografix.com/GPX/gpx_style/0/2" version="1.1"
creator="https://gpx.studio">
<metadata>
<name>with_power</name>
<author>
<name>gpx.studio</name>
<link href="https://gpx.studio"></link>
</author>
</metadata>
<trk>
<name>with_power</name>
<type>Cycling</type>
<trkseg>
<trkpt lat="50.790867" lon="4.404968">
<ele>109.0</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.790714" lon="4.405036">
<ele>110.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.790336" lon="4.405259">
<ele>110.3</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.790165" lon="4.405331">
<ele>110.0</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.790008" lon="4.405359">
<ele>110.3</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.789818" lon="4.405359">
<ele>109.3</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.789409" lon="4.40534">
<ele>107.0</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.789105" lon="4.405411">
<ele>106.0</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.788799" lon="4.405527">
<ele>108.5</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.788645" lon="4.405606">
<ele>109.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.7885" lon="4.405711">
<ele>110.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.78822" lon="4.405959">
<ele>112.0</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.787956" lon="4.406092">
<ele>112.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.787814" lon="4.406143">
<ele>113.5</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.787674" lon="4.406177">
<ele>114.3</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.787451" lon="4.406199">
<ele>115.3</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.787297" lon="4.406177">
<ele>114.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.78716" lon="4.406098">
<ele>114.3</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.787045" lon="4.405984">
<ele>114.3</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.786683" lon="4.405653">
<ele>114.5</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.786538" lon="4.405543">
<ele>115.0</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.78635" lon="4.405441">
<ele>115.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.786275" lon="4.40542">
<ele>115.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.786182" lon="4.405435">
<ele>116.0</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.786121" lon="4.405475">
<ele>115.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.786042" lon="4.405558">
<ele>115.5</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.785821" lon="4.405925">
<ele>114.5</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.785672" lon="4.406119">
<ele>112.5</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.785516" lon="4.406256">
<ele>110.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.785384" lon="4.406364">
<ele>109.0</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.785126" lon="4.406475">
<ele>106.3</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.784697" lon="4.406537">
<ele>104.3</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.784591" lon="4.40657">
<ele>104.0</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.784507" lon="4.406612">
<ele>103.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.784435" lon="4.40669">
<ele>103.3</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.784209" lon="4.407148">
<ele>103.5</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.784162" lon="4.407257">
<ele>103.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.784077" lon="4.407372">
<ele>104.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.784006" lon="4.407435">
<ele>105.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.783924" lon="4.407471">
<ele>106.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.783837" lon="4.407486">
<ele>107.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.783771" lon="4.407472">
<ele>108.5</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.783697" lon="4.407428">
<ele>109.3</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.783626" lon="4.407363">
<ele>110.0</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.783548" lon="4.407274">
<ele>110.5</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.783458" lon="4.407134">
<ele>110.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.783123" lon="4.406435">
<ele>111.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.782982" lon="4.406168">
<ele>112.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.782871" lon="4.406044">
<ele>113.3</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.78279" lon="4.406021">
<ele>113.5</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.782714" lon="4.406018">
<ele>113.5</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.782607" lon="4.406047">
<ele>113.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.782405" lon="4.406194">
<ele>114.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.782175" lon="4.406413">
<ele>115.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.781749" lon="4.407018">
<ele>118.5</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.781654" lon="4.407316">
<ele>119.5</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.781563" lon="4.407764">
<ele>121.3</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.781487" lon="4.407984">
<ele>122.0</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.781422" lon="4.408216">
<ele>122.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.781395" lon="4.408508">
<ele>123.5</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.781399" lon="4.409114">
<ele>126.3</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.781367" lon="4.409428">
<ele>128.0</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.781286" lon="4.409607">
<ele>129.0</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.78116" lon="4.409789">
<ele>130.0</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.780804" lon="4.409993">
<ele>130.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.780389" lon="4.410334">
<ele>131.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.780232" lon="4.410563">
<ele>132.3</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.780094" lon="4.410827">
<ele>132.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.779723" lon="4.411582">
<ele>135.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.779591" lon="4.411791">
<ele>135.5</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.779125" lon="4.412435">
<ele>132.5</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.778676" lon="4.412979">
<ele>134.0</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.778194" lon="4.413466">
<ele>136.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.777427" lon="4.414302">
<ele>137.5</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.777165" lon="4.414736">
<ele>137.3</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.776927" lon="4.415201">
<ele>137.5</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.776778" lon="4.415613">
<ele>137.3</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.776553" lon="4.416425">
<ele>134.8</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.776326" lon="4.417304">
<ele>132.3</ele>
<extensions>
<gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>
</extensions>
</trkpt>
<trkpt lat="50.776129" lon="4.418383">
<ele>129.5</ele>
<extensions>
<gpxpx:PowerInWatts>210</gpxpx:PowerInWatts>
</extensions>
</trkpt>
</trkseg>
</trk>
</gpx>
+39 -8
View File
@@ -49,7 +49,7 @@ pub fn select_elements<H: History>(
}
/// Selects all the elements of the same kind as the selected ones, in the same place: all the
/// files, the tracks of the file, the segments of the track, the waypoints of the file. All the
/// files, the routes or the tracks of the file, the segments of the track, the waypoints of the file. All the
/// files when nothing is selected.
pub fn select_all<H: History>(engine: &mut Engine<H>) -> Outcome {
execute(engine, Action::SelectAll)
@@ -62,6 +62,19 @@ pub fn arrow_select<H: History>(engine: &mut Engine<H>, down: bool, add: bool) -
execute(engine, Action::ArrowSelect { down, add })
}
pub fn select_routes<H: History>(
engine: &mut Engine<H>,
file_id: Uuid,
route_ids: &[Uuid],
mode: engine::SelectMode,
) -> Outcome {
let selection = engine::Selection::Route {
file_id: engine::FileId(file_id),
rte_ids: route_ids.iter().copied().map(engine::RouteId).collect(),
};
select_elements(engine, selection, mode)
}
pub fn select_tracks<H: History>(
engine: &mut Engine<H>,
file_id: Uuid,
@@ -135,21 +148,30 @@ pub fn redo<H: History>(engine: &mut Engine<H>) -> Outcome {
// Files
/// Creates a file and selects it. With `lng` and `lat`, it starts with a track and a segment that
/// hold that trackpoint (at `ele`, 0 by default).
/// Creates a file and selects it. With `route`, it starts with a route. With `lng` and `lat`, it
/// starts with a segment that holds that trackpoint (at `ele`, 0 by default): the one of its route
/// with `route`, of a track otherwise.
pub fn new_file<H: History>(
engine: &mut Engine<H>,
name: &str,
lng: Option<f64>,
lat: Option<f64>,
ele: Option<f64>,
route: bool,
) -> Outcome {
let trackpoint = lng.zip(lat).map(|(lng, lat)| engine::NewTrackpoint {
lng,
lat,
ele: ele.unwrap_or_default(),
});
edit(engine, Command::New(engine::New { name, trackpoint }))
edit(
engine,
Command::New(engine::New {
name,
trackpoint,
route,
}),
)
}
/// Loads files as a single command (one undo step). The files are in `data`, one after the other,
@@ -183,8 +205,8 @@ pub fn load_files<H: History>(
}
/// Brings changes of the files from the outside in (see [`engine::SyncFiles`]): each file is
/// `(id, data, name)`, and replaces the file of that id, or is added if there is none; it is called
/// `name`, whatever its data says. The files to `remove` are deleted. The selection is left alone.
/// `(id, data, name)`, and replaces the file of that id, or is added if there is none; `name` is
/// its name when its data has none. The files to `remove` are deleted. The selection is left alone.
pub fn sync_files<H: History>(
engine: &mut Engine<H>,
files: &[(Uuid, &[u8], &str)],
@@ -304,6 +326,15 @@ pub fn style<H: History>(
)
}
/// Converts the selected routes to tracks, or the selected tracks or segments to routes.
pub fn convert<H: History>(engine: &mut Engine<H>) -> Outcome {
edit(engine, Command::Convert(engine::Convert))
}
pub fn new_route<H: History>(engine: &mut Engine<H>) -> Outcome {
edit(engine, Command::NewRoute(engine::NewRoute))
}
pub fn new_track<H: History>(engine: &mut Engine<H>) -> Outcome {
edit(engine, Command::NewTrack(engine::NewTrack))
}
@@ -575,7 +606,7 @@ pub fn route<H: History>(
at_revision(engine, revision, |engine| {
edit(
engine,
Command::Route(engine::Route {
Command::Reroute(engine::Reroute {
start,
end,
lng,
@@ -618,7 +649,7 @@ pub fn change_loop_start<H: History>(engine: &mut Engine<H>, revision: u32, inde
})
}
/// Cuts the file, the track or the segment (see `SplitType`) of the trackpoint `at` of the
/// Cuts the file, the route, the track or the segment (see `SplitType`) of the trackpoint `at` of the
/// selection in two, there: the trackpoint ends the first part and starts the second one. A file that
/// is cut is kept (and its two parts are new files) if the history needs it, see
/// [`History::KEEPS_ORIGINALS`].
+74 -16
View File
@@ -131,6 +131,10 @@ pub enum Selection {
File {
file_ids: Vec<Uuid>,
},
Route {
file_id: Uuid,
route_ids: Vec<Uuid>,
},
Track {
file_id: Uuid,
track_ids: Vec<Uuid>,
@@ -157,6 +161,10 @@ impl From<&engine::Selection> for Selection {
S::File { file_ids } => Self::File {
file_ids: file_ids.iter().map(|id| id.0).collect(),
},
S::Route { file_id, rte_ids } => Self::Route {
file_id: file_id.0,
route_ids: rte_ids.iter().map(|id| id.0).collect(),
},
S::Track { file_id, trk_ids } => Self::Track {
file_id: file_id.0,
track_ids: trk_ids.iter().map(|id| id.0).collect(),
@@ -187,6 +195,10 @@ impl From<Selection> for engine::Selection {
Selection::File { file_ids } => S::File {
file_ids: file_ids.into_iter().map(FileId).collect(),
},
Selection::Route { file_id, route_ids } => S::Route {
file_id: FileId(file_id),
rte_ids: route_ids.into_iter().map(engine::RouteId).collect(),
},
Selection::Track { file_id, track_ids } => S::Track {
file_id: FileId(file_id),
trk_ids: track_ids.into_iter().map(engine::TrackId).collect(),
@@ -229,6 +241,10 @@ pub enum MoveTarget {
Files {
index: f64,
},
Routes {
file_id: Uuid,
index: f64,
},
Tracks {
file_id: Uuid,
index: f64,
@@ -250,6 +266,10 @@ impl From<MoveTarget> for engine::MoveTarget {
let index = |index: f64| index.max(0.0) as usize;
match target {
MoveTarget::Files { index: i } => T::Files { index: index(i) },
MoveTarget::Routes { file_id, index: i } => T::Routes {
file_id: FileId(file_id),
index: index(i),
},
MoveTarget::Tracks { file_id, index: i } => T::Tracks {
file_id: FileId(file_id),
index: index(i),
@@ -286,6 +306,9 @@ impl Clipboard {
engine::ClipboardIds::Files(ids) => {
("files", strings(ids.into_iter().map(|id| id.0).collect()))
}
engine::ClipboardIds::Routes(ids) => {
("routes", strings(ids.into_iter().map(|id| id.0).collect()))
}
engine::ClipboardIds::Tracks(ids) => {
("tracks", strings(ids.into_iter().map(|id| id.0).collect()))
}
@@ -315,14 +338,15 @@ pub struct FileStructure {
name: String,
#[serde(skip_serializing_if = "Option::is_none")]
desc: Option<String>,
routes: Vec<RouteNode>,
tracks: Vec<TrackNode>,
waypoints: Vec<WaypointNode>,
waypoints_rev: String,
}
/// The name, description and style of a route or a track.
#[derive(Serialize)]
struct TrackNode {
id: String,
struct PathInfo {
#[serde(skip_serializing_if = "Option::is_none")]
name: Option<String>,
#[serde(skip_serializing_if = "Option::is_none")]
@@ -333,6 +357,33 @@ struct TrackNode {
opacity: Option<f64>,
#[serde(skip_serializing_if = "Option::is_none")]
width: Option<f64>,
}
impl From<&engine::PathInfo> for PathInfo {
fn from(info: &engine::PathInfo) -> Self {
Self {
name: info.name.clone(),
desc: info.desc.clone(),
color: info.color.clone(),
opacity: info.opacity,
width: info.width,
}
}
}
#[derive(Serialize)]
struct RouteNode {
id: String,
#[serde(flatten)]
info: PathInfo,
segment: SegmentNode,
}
#[derive(Serialize)]
struct TrackNode {
id: String,
#[serde(flatten)]
info: PathInfo,
segments: Vec<SegmentNode>,
}
@@ -343,6 +394,16 @@ struct SegmentNode {
length: f64,
}
impl From<&engine::SegmentNode> for SegmentNode {
fn from(seg: &engine::SegmentNode) -> Self {
Self {
id: seg.id.0.to_string(),
rev: seg.rev_id.0.to_string(),
length: seg.len as f64,
}
}
}
#[derive(Serialize)]
struct WaypointNode {
id: String,
@@ -358,25 +419,22 @@ impl FileStructure {
id: file.id.0.to_string(),
name: file.name.clone(),
desc: file.desc.clone(),
routes: file
.routes
.iter()
.map(|rte| RouteNode {
id: rte.id.0.to_string(),
info: (&rte.info).into(),
segment: (&rte.segment).into(),
})
.collect(),
tracks: file
.tracks
.iter()
.map(|trk| TrackNode {
id: trk.id.0.to_string(),
name: trk.name.clone(),
desc: trk.desc.clone(),
color: trk.color.clone(),
opacity: trk.opacity,
width: trk.width,
segments: trk
.segments
.iter()
.map(|seg| SegmentNode {
id: seg.id.0.to_string(),
rev: seg.rev_id.0.to_string(),
length: seg.len as f64,
})
.collect(),
info: (&trk.info).into(),
segments: trk.segments.iter().map(Into::into).collect(),
})
.collect(),
waypoints: file
+1 -3
View File
@@ -76,9 +76,7 @@ pub fn trackpoint<H: History>(
}
/// A file as GPX (UTF-8 bytes), `None` if it does not exist. The options say which data of the
/// trackpoints to keep, and with `as_route` that the segments are written as routes (`rte`)
/// instead of tracks: the anchors are the route points, and the trackpoints between them the
/// points of their path.
/// trackpoints to keep.
pub fn export_file<H: History>(
engine: &Engine<H>,
file_id: Uuid,
+14 -1
View File
@@ -6,7 +6,7 @@
//! between them are kept: every part comes after the previous one in the new order, as late as the
//! original one came after the part that followed it.
use crate::{Track, TrackSegment, Trackpoint};
use crate::{Route, Track, TrackSegment, Trackpoint};
/// The time of the first trackpoint of a segment.
pub fn start_time(segment: &TrackSegment) -> Option<i64> {
@@ -94,6 +94,19 @@ pub fn reverse_segments(
}
}
/// Reverses the routes of a file, and their trackpoints, like the segments of a track (see
/// [`reverse_segments`]).
pub fn reverse_routes(routes: &mut [Route], original_next: Option<i64>, new_previous: Option<i64>) {
routes.reverse();
let (mut next, mut previous) = (original_next, new_previous);
for route in routes.iter_mut() {
let original_start = start_time(&route.seg);
reverse_segment(&mut route.seg, next, previous);
next = original_start;
previous = end_time(&route.seg);
}
}
/// Reverses the tracks of a file, and their segments and trackpoints.
pub fn reverse_tracks(tracks: &mut [Track], original_next: Option<i64>, new_previous: Option<i64>) {
tracks.reverse();
+51 -1
View File
@@ -1,4 +1,4 @@
use crate::{Link, Track, Waypoints};
use crate::{Link, Route, Track, TrackSegment, Waypoints};
use super::common::uuid_id;
@@ -8,6 +8,9 @@ uuid_id!(FileId);
pub struct File {
pub id: FileId,
pub info: FileInfo,
/// The routes come before the tracks, as in GPX: everything that goes through the segments
/// of a file (statistics, the elevation profile...) takes the ones of the routes first.
pub rte: Vec<Route>,
pub trk: Vec<Track>,
pub wpt: Waypoints,
}
@@ -29,3 +32,50 @@ pub struct Author {
/// The author has a single link, unlike the other elements.
pub link: Option<Link>,
}
/// Where a segment is in its file: the one of a route, or one of a track.
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
pub enum SegmentIndex {
Route(usize),
Track { trk: usize, seg: usize },
}
impl File {
/// The segments of the routes, then the ones of the tracks.
pub fn segments(&self) -> impl Iterator<Item = &TrackSegment> {
self.rte
.iter()
.map(|rte| &rte.seg)
.chain(self.trk.iter().flat_map(|trk| &trk.trkseg))
}
/// Where the segments are, in the order of [`File::segments`].
pub fn segment_indices(&self) -> impl Iterator<Item = SegmentIndex> + '_ {
(0..self.rte.len())
.map(SegmentIndex::Route)
.chain(self.trk.iter().enumerate().flat_map(|(trk, track)| {
(0..track.trkseg.len()).map(move |seg| SegmentIndex::Track { trk, seg })
}))
}
/// The segment at `index`. Panics if there is none.
pub fn segment(&self, index: SegmentIndex) -> &TrackSegment {
match index {
SegmentIndex::Route(rte) => &self.rte[rte].seg,
SegmentIndex::Track { trk, seg } => &self.trk[trk].trkseg[seg],
}
}
/// The segment at `index`. Panics if there is none.
pub fn segment_mut(&mut self, index: SegmentIndex) -> &mut TrackSegment {
match index {
SegmentIndex::Route(rte) => &mut self.rte[rte].seg,
SegmentIndex::Track { trk, seg } => &mut self.trk[trk].trkseg[seg],
}
}
/// Whether the file has no route nor track.
pub fn has_no_path(&self) -> bool {
self.rte.is_empty() && self.trk.is_empty()
}
}
+2
View File
@@ -3,6 +3,7 @@ mod chunk;
mod chunked;
mod common;
mod file;
mod route;
mod segment;
mod track;
mod trackpoint;
@@ -14,6 +15,7 @@ pub use chunk::*;
pub use chunked::*;
pub use common::*;
pub use file::*;
pub use route::*;
pub use segment::*;
pub use track::*;
pub use trackpoint::*;
+15
View File
@@ -0,0 +1,15 @@
use crate::{TrackInfo, TrackSegment};
use super::common::uuid_id;
uuid_id!(RouteId);
/// A route of the file (`rte`): like a track, but with a single segment, its points.
///
/// It has the information of a track: GPX gives the same to both.
#[derive(Debug, Default, Clone)]
pub struct Route {
pub id: RouteId,
pub info: TrackInfo,
pub seg: TrackSegment,
}
+56 -44
View File
@@ -1,6 +1,6 @@
use crate::{
Author, Chunk, File, Link, LngLat, Track, TrackSegment, Trackpoint, TrackpointCategories,
TrackpointChunk, Waypoint, WaypointChunk, distance,
Author, Chunk, File, Link, LngLat, Route, Track, TrackSegment, Trackpoint,
TrackpointCategories, TrackpointChunk, Waypoint, WaypointChunk, distance,
};
use chrono::DateTime;
use quick_xml::Error;
@@ -18,6 +18,7 @@ enum GPXElement {
Author(Author),
Link(Link),
Text,
/// A track, or a route, whose information is read the same way.
Track(Track),
Segment(TrackSegment),
Trackpoint(Trackpoint),
@@ -396,7 +397,7 @@ pub fn parse(data: &[u8], categories: &mut TrackpointCategories) -> Result<File,
let mut stack: Vec<GPXElement> = vec![];
let mut trkpt_chunk = TrackpointChunk::default();
let mut wpt_chunk = WaypointChunk::default();
// Routes are read as tracks of a single segment, see `RouteReader`. The points of the detailed
// The segment of a route is read by `RouteReader`. The points of the detailed
// path that a route point comes with (`gpxx:rpt`) are collected until the route point is read.
let mut route = RouteReader::default();
let mut rpt_points: Vec<Trackpoint> = vec![];
@@ -530,12 +531,12 @@ pub fn parse(data: &[u8], categories: &mut TrackpointCategories) -> Result<File,
}
}
"rte" => {
if let Some(GPXElement::Track(mut trk)) = stack.pop() {
let trkseg = std::mem::take(&mut route).finish();
if !trkseg.is_empty() {
trk.trkseg.push(trkseg);
}
gpx.trk.push(trk);
if let Some(GPXElement::Track(trk)) = stack.pop() {
gpx.rte.push(Route {
info: trk.info,
seg: std::mem::take(&mut route).finish(),
..Default::default()
});
}
}
"rtept" => {
@@ -740,38 +741,37 @@ mod tests {
}
#[test]
fn test_parse_routes_as_tracks() {
fn test_parse_routes() {
let gpx = parse_data("with_routes");
assert_eq!(gpx.trk.len(), 2);
let trk = &gpx.trk[0];
assert_eq!(trk.info.name.as_deref(), Some("route 1"));
assert_eq!(trk.info.type_.as_deref(), Some("Cycling"));
assert_eq!(trk.trkseg.len(), 1);
let trkseg = &trk.trkseg[0];
assert_eq!(trkseg.len(), 49);
let trkpt = &trkseg[0];
assert!(gpx.trk.is_empty());
assert_eq!(gpx.rte.len(), 2);
let rte = &gpx.rte[0];
assert_eq!(rte.info.name.as_deref(), Some("route 1"));
assert_eq!(rte.info.type_.as_deref(), Some("Cycling"));
assert_eq!(rte.seg.len(), 49);
let trkpt = &rte.seg[0];
assert_eq!(trkpt.coordinates.lat, 50.790867);
assert_eq!(trkpt.coordinates.lng, 4.404968);
assert_eq!(trkpt.ele, 109.0);
let trk = &gpx.trk[1];
assert_eq!(trk.info.name.as_deref(), Some("route 2"));
assert_eq!(trk.trkseg[0].len(), 28);
let rte = &gpx.rte[1];
assert_eq!(rte.info.name.as_deref(), Some("route 2"));
assert_eq!(rte.seg.len(), 28);
assert_ne!(gpx.rte[0].id, rte.id);
}
#[test]
fn test_parse_route_point_extensions() {
let gpx = parse_data("with_route_extensions");
// a route without points has no segment
assert_eq!(gpx.trk.len(), 2);
assert!(gpx.trk[1].trkseg.is_empty());
// a route without points has an empty segment
assert_eq!(gpx.rte.len(), 2);
assert!(gpx.rte[1].seg.is_empty());
let trk = &gpx.trk[0];
assert_eq!(trk.info.name.as_deref(), Some("with extensions"));
assert_eq!(trk.trkseg.len(), 1);
let points: Vec<_> = trk.trkseg[0].iter().collect();
let rte = &gpx.rte[0];
assert_eq!(rte.info.name.as_deref(), Some("with extensions"));
let points: Vec<_> = rte.seg.iter().collect();
// the route points are kept, the points of the detailed path follow the one they come with
let coordinates: Vec<_> = points
.iter()
@@ -788,14 +788,13 @@ mod tests {
#[test]
fn test_route_points_are_the_anchors_and_nothing_else_is() {
let gpx = parse_data("with_route_extensions");
let anchors: Vec<_> = gpx.trk[0].trkseg[0].iter().map(|p| p.anchor).collect();
let anchors: Vec<_> = gpx.rte[0].seg.iter().map(|p| p.anchor).collect();
assert_eq!(anchors, [Some(0), None, None, Some(0), Some(0)]);
// no anchor is computed for the points of a route that has none in between
let gpx = parse_data("with_routes");
for trk in &gpx.trk {
let segment = &trk.trkseg[0];
assert!(segment.iter().all(|p| p.anchor == Some(0)));
for rte in &gpx.rte {
assert!(rte.seg.iter().all(|p| p.anchor == Some(0)));
}
// while a track gets the anchors that its shape needs
let track = parse_data("with_tracks");
@@ -807,14 +806,14 @@ mod tests {
#[test]
fn test_parse_the_points_of_the_detailed_path_of_a_route() {
let gpx = parse_data("with_route_points");
assert_eq!(gpx.trk.len(), 1);
let trk = &gpx.trk[0];
assert_eq!(trk.info.name.as_deref(), Some("Lausanne"));
assert_eq!(trk.info.cmt.as_deref(), Some("comment"));
assert_eq!(trk.info.desc.as_deref(), Some("description"));
assert_eq!(trk.info.type_.as_deref(), Some("Cycling"));
assert_eq!(gpx.rte.len(), 1);
let rte = &gpx.rte[0];
assert_eq!(rte.info.name.as_deref(), Some("Lausanne"));
assert_eq!(rte.info.cmt.as_deref(), Some("comment"));
assert_eq!(rte.info.desc.as_deref(), Some("description"));
assert_eq!(rte.info.type_.as_deref(), Some("Cycling"));
let points: Vec<_> = trk.trkseg[0].iter().cloned().collect();
let points: Vec<_> = rte.seg.iter().cloned().collect();
// 3 route points and 4 points between them, the repeated one being kept
assert_eq!(points.len(), 7);
let kinds: Vec<_> = points.iter().map(|p| p.anchor.is_some()).collect();
@@ -846,7 +845,7 @@ mod tests {
</rtept>
</rte></gpx>"#;
let gpx = parse(data.as_bytes(), &mut Default::default()).unwrap();
let points: Vec<_> = gpx.trk[0].trkseg[0].iter().cloned().collect();
let points: Vec<_> = gpx.rte[0].seg.iter().cloned().collect();
let coordinates: Vec<_> = points.iter().map(|p| p.coordinates.lat).collect();
assert_eq!(coordinates, [1.0, 2.0, 2.1, 2.2]);
// the route point is an anchor, and so is the end of the path
@@ -861,7 +860,7 @@ mod tests {
<gpxx:RoutePointExtension><gpxx:rpt lat="1.5" lon="1.5"/></gpxx:RoutePointExtension>
</extensions></rtept></rte></gpx>"#;
let gpx = parse(data.as_bytes(), &mut Default::default()).unwrap();
let anchors: Vec<_> = gpx.trk[0].trkseg[0].iter().map(|p| p.anchor).collect();
let anchors: Vec<_> = gpx.rte[0].seg.iter().map(|p| p.anchor).collect();
assert_eq!(anchors, [Some(0), Some(0)]);
}
@@ -930,14 +929,14 @@ mod tests {
assert_eq!(waypoints[0].links[2].text.as_deref(), Some("three"));
assert!(waypoints[1].links.is_empty());
// tracks, and routes which are read as tracks
assert_eq!(gpx.trk.len(), 2);
// tracks, and routes
assert_eq!((gpx.trk.len(), gpx.rte.len()), (1, 1));
assert_eq!(
hrefs(&gpx.trk[0].info.links),
["https://example.com/trk-1", "https://example.com/trk-2"]
);
assert_eq!(
hrefs(&gpx.trk[1].info.links),
hrefs(&gpx.rte[0].info.links),
["https://example.com/rte-1", "https://example.com/rte-2"]
);
}
@@ -1223,6 +1222,19 @@ mod tests {
assert!(trkpt.power.is_some_and(|p| p == 200));
}
#[test]
fn test_parse_trackpoint_power_3() {
let gpx = parse_data("with_power_3");
assert_eq!(gpx.trk.len(), 1);
let trk = &gpx.trk[0];
assert_eq!(trk.trkseg.len(), 1);
let trkseg = &trk.trkseg[0];
assert!(!trkseg.is_empty());
let trkpt = &trkseg[0];
assert!(trkpt.power.is_some_and(|p| p == 200));
}
#[test]
fn test_parse_trackpoint_atemp() {
let gpx = parse_data("with_temp");
+161 -124
View File
@@ -8,9 +8,6 @@ use crate::{File, Link, Track, Trackpoint, TrackpointCategories, Waypoint};
/// What the written file contains: the data of the trackpoints that are `false` are left out.
/// The same type tells which data a file has, see [`File::exportable_data`].
///
/// With `as_route`, the file is written with routes instead of tracks, see [`write`]. It is not
/// data: it is off in `ALL` and `NONE`, and in what [`File::exportable_data`] gives.
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
pub struct ExportOptions {
pub time: bool,
@@ -20,8 +17,6 @@ pub struct ExportOptions {
pub power: bool,
/// The OpenStreetMap data: surface, highway, SAC scale and MTB scale.
pub osm: bool,
/// Writes every segment as a route (`rte`) instead of a track.
pub as_route: bool,
}
impl ExportOptions {
@@ -36,7 +31,6 @@ impl ExportOptions {
atemp: value,
power: value,
osm: value,
as_route: false,
}
}
@@ -49,7 +43,6 @@ impl ExportOptions {
atemp: self.atemp || other.atemp,
power: self.power || other.power,
osm: self.osm || other.osm,
as_route: self.as_route || other.as_route,
}
}
}
@@ -59,12 +52,7 @@ impl File {
/// options of [`write`] can leave out.
pub fn exportable_data(&self) -> ExportOptions {
let mut data = ExportOptions::NONE;
for point in self
.trk
.iter()
.flat_map(|trk| &trk.trkseg)
.flat_map(|seg| seg.iter())
{
for point in self.segments().flat_map(|seg| seg.iter()) {
data.time |= point.time.is_some();
data.hr |= point.hr.is_some();
data.cad |= point.cad.is_some();
@@ -82,6 +70,10 @@ impl File {
}
}
/// The highest zoom level from which an anchor is shown that makes it a route point, see [`write`]:
/// the anchors that only appear when zooming in further are details of the path.
pub const ROUTE_POINT_MAX_ZOOM: u8 = 12;
const SCHEMA_LOCATION: &str = "http://www.topografix.com/GPX/1/1 http://www.topografix.com/GPX/1/1/gpx.xsd http://www.garmin.com/xmlschemas/GpxExtensions/v3 http://www.garmin.com/xmlschemas/GpxExtensionsv3.xsd http://www.garmin.com/xmlschemas/TrackPointExtension/v1 http://www.garmin.com/xmlschemas/TrackPointExtensionv1.xsd http://www.garmin.com/xmlschemas/PowerExtension/v1 http://www.garmin.com/xmlschemas/PowerExtensionv1.xsd http://www.topografix.com/GPX/gpx_style/0/2 http://www.topografix.com/GPX/gpx_style/0/2/gpx_style.xsd";
type XmlWriter = Writer<Vec<u8>>;
@@ -89,10 +81,10 @@ type XmlWriter = Writer<Vec<u8>>;
/// Writes a file as GPX 1.1, in UTF-8. The surface, highway, SAC scale and MTB scale of the trackpoints are
/// the names of their codes in `categories`.
///
/// The segments are written as tracks, or with `options.as_route` as routes: a segment becomes a
/// `rte` that has the information of its track, its anchors are the `rtept` and the trackpoints
/// between two anchors are the `gpxx:rpt` of the `rtept` before them (which only have a
/// position, unlike the `rtept`). The first and last trackpoints always are `rtept`.
/// The routes are written as routes (`rte`): the anchors of their segment that are shown from the
/// zoom level [`ROUTE_POINT_MAX_ZOOM`] or a lower one are the `rtept`, and the trackpoints between
/// two of them are the `gpxx:rpt` of the `rtept` before them (which only have a position, unlike
/// the `rtept`). The first and last trackpoints always are `rtept`.
pub fn write(file: &File, categories: &TrackpointCategories, options: ExportOptions) -> Vec<u8> {
let mut writer = Writer::new_with_indent(Vec::new(), b' ', 4);
// writing to memory does not fail
@@ -100,9 +92,9 @@ pub fn write(file: &File, categories: &TrackpointCategories, options: ExportOpti
writer.into_inner()
}
fn write_file(
fn write_file<'a>(
w: &mut XmlWriter,
file: &File,
file: &'a File,
categories: &TrackpointCategories,
options: ExportOptions,
) -> io::Result<()> {
@@ -136,20 +128,18 @@ fn write_file(
for wpt in file.wpt.iter() {
write_waypoint(w, wpt, options)?;
}
// a single unnamed route or track is named like the file
let alone = file.rte.len() + file.trk.len() == 1;
let name = |info: &'a crate::TrackInfo| match &info.name {
Some(name) if !name.is_empty() => Some(name.as_str()),
_ if alone => Some(file.info.name.as_str()),
_ => None,
};
for rte in &file.rte {
write_route(w, &rte.info, &rte.seg, name(&rte.info), categories, options)?;
}
for trk in &file.trk {
// a single unnamed track is named like the file
let name = match (&trk.info.name, file.trk.len()) {
(Some(name), _) if !name.is_empty() => Some(name.as_str()),
(_, 1) => Some(file.info.name.as_str()),
_ => None,
};
if options.as_route {
for segment in trk.trkseg.iter().filter(|segment| !segment.is_empty()) {
write_route(w, trk, segment, name, categories, options)?;
}
} else {
write_track(w, trk, name, categories, options)?;
}
write_track(w, trk, name(&trk.info), categories, options)?;
}
Ok(())
})?;
@@ -311,22 +301,28 @@ fn write_track(
fn write_route(
w: &mut XmlWriter,
trk: &Track,
info: &crate::TrackInfo,
segment: &crate::TrackSegment,
name: Option<&str>,
categories: &TrackpointCategories,
options: ExportOptions,
) -> io::Result<()> {
w.create_element("rte").write_inner_content(|w| {
write_track_info(w, name, &trk.info)?;
write_track_info(w, name, info)?;
let points: Vec<&Trackpoint> = segment.iter().collect();
let last = points.len() - 1;
let is_anchor = |i: usize| i == 0 || i == last || points[i].anchor.is_some();
let last = points.len().saturating_sub(1);
let is_route_point = |i: usize| {
i == 0
|| i == last
|| points[i]
.anchor
.is_some_and(|zoom| zoom <= ROUTE_POINT_MAX_ZOOM)
};
let mut i = 0;
while i < points.len() {
// the trackpoints up to the next anchor
// the trackpoints up to the next route point
let mut next = i + 1;
while next < points.len() && !is_anchor(next) {
while next < points.len() && !is_route_point(next) {
next += 1;
}
write_point(
@@ -406,9 +402,10 @@ fn write_point(
Ok(())
})?;
}
// `PowerInWatts` is itself the root element of Garmin's power extension, without a
// wrapper
if let Some(power) = power {
w.create_element("gpxpx:PowerExtension")
.write_inner_content(|w| number(w, "gpxpx:PowerInWatts", power))?;
number(w, "gpxpx:PowerInWatts", power)?;
}
if !path.is_empty() {
w.create_element("gpxx:RoutePointExtension")
@@ -455,9 +452,7 @@ mod tests {
}
fn points(file: &File) -> Vec<Trackpoint> {
file.trk
.iter()
.flat_map(|trk| &trk.trkseg)
file.segments()
.flat_map(|seg| seg.iter().cloned())
.collect()
}
@@ -517,6 +512,7 @@ mod tests {
"with_temp",
"with_power_1",
"with_power_2",
"with_power_3",
"with_surface",
"with_highway",
"with_style",
@@ -584,6 +580,16 @@ mod tests {
assert_eq!(again.trk[0].info.name.as_deref(), Some("a <b> & c"));
}
#[test]
fn test_power_is_written_without_a_wrapper() {
let (file, categories) = read("with_power_2");
let text: String = written_text(&file, &categories, ExportOptions::ALL)
.split_whitespace()
.collect();
assert!(text.contains("<extensions><gpxpx:PowerInWatts>200</gpxpx:PowerInWatts>"));
assert!(!text.contains("<gpxpx:PowerExtension>"));
}
#[test]
fn test_exportable_data() {
let (file, _) = read("with_time");
@@ -595,11 +601,20 @@ mod tests {
assert!(file.exportable_data().power);
}
fn as_route() -> ExportOptions {
ExportOptions {
as_route: true,
..ExportOptions::ALL
}
/// The file with each segment of its tracks as a route, which has the information of its
/// track.
fn as_routes(mut file: File) -> File {
file.rte = std::mem::take(&mut file.trk)
.into_iter()
.flat_map(|trk| {
trk.trkseg.into_iter().map(move |seg| crate::Route {
info: trk.info.clone(),
seg,
..Default::default()
})
})
.collect();
file
}
fn count(text: &str, pattern: &str) -> usize {
@@ -607,19 +622,17 @@ mod tests {
}
#[test]
fn test_export_as_route_writes_routes_instead_of_tracks() {
fn test_routes_are_written_with_their_route_points_and_their_path() {
let (file, categories) = read("with_route_points");
let text = written_text(&file, &categories, as_route());
let text = written_text(&file, &categories, ExportOptions::ALL);
assert_eq!(count(&text, "<rte>"), 1);
assert_eq!(count(&text, "<trk>"), 0);
assert_eq!(count(&text, "<trkseg>"), 0);
assert_eq!(count(&text, "<trkpt"), 0);
// the anchors are the route points, the trackpoints between them the points of their path
assert_eq!(count(&text, "<rtept"), 3);
assert_eq!(count(&text, "<gpxx:rpt"), 4);
assert_eq!(count(&text, "<gpxx:RoutePointExtension>"), 2);
// a route has the information of its track
assert!(text.contains("<name>Lausanne</name>"));
assert!(text.contains("<cmt>comment</cmt>"));
assert!(text.contains("<desc>description</desc>"));
@@ -634,14 +647,14 @@ mod tests {
}
#[test]
fn test_a_route_survives_being_exported_as_a_route() {
fn test_a_route_survives_being_written() {
let (file, mut categories) = read("with_route_points");
let written = write(&file, &categories, as_route());
let written = write(&file, &categories, ExportOptions::ALL);
let again = parse(&written, &mut categories).unwrap();
assert_eq!(again.trk.len(), 1);
assert_eq!(again.trk[0].info, file.trk[0].info);
let (a, b) = (&file.trk[0].trkseg[0], &again.trk[0].trkseg[0]);
assert_eq!(again.rte.len(), 1);
assert_eq!(again.rte[0].info, file.rte[0].info);
let (a, b) = (&file.rte[0].seg, &again.rte[0].seg);
assert_eq!(a.len(), b.len());
for (a, b) in a.iter().zip(b.iter()) {
assert_eq!(a.coordinates.lat, b.coordinates.lat);
@@ -659,85 +672,121 @@ mod tests {
}
#[test]
fn test_tracks_become_one_route_per_segment_with_the_information_of_the_track() {
fn test_the_anchors_shown_up_to_zoom_12_are_the_route_points() {
let (file, mut categories) = read("with_tracks_and_segments");
let segments: usize = file.trk.iter().map(|t| t.trkseg.len()).sum();
assert!(segments > 2);
let written = write(&file, &categories, as_route());
let text = String::from_utf8(written.clone()).unwrap();
assert_eq!(count(&text, "<rte>"), segments);
assert_eq!(count(&text, "<trk>"), 0);
// every route starts and ends with a route point
assert_eq!(count(&text, "</rte>"), segments);
let file = as_routes(file);
assert!(file.rte.len() > 2);
let written = write(&file, &categories, ExportOptions::ALL);
let again = parse(&written, &mut categories).unwrap();
assert_eq!(again.trk.len(), segments);
let mut routes = again.trk.iter();
for track in &file.trk {
for segment in &track.trkseg {
let route = routes.next().unwrap();
// the information of the track, repeated for each of its segments
assert_eq!(route.info.name, track.info.name);
assert_eq!(route.info.type_, track.info.type_);
let (a, b) = (segment, &route.trkseg[0]);
assert_eq!(a.len(), b.len());
for (a, b) in a.iter().zip(b.iter()) {
assert_eq!(
(a.coordinates.lat, a.coordinates.lng),
(b.coordinates.lat, b.coordinates.lng)
);
}
// the anchors of the segment are the ones of its route
let anchors = |s: &crate::TrackSegment| {
s.iter()
.enumerate()
.filter(|(_, p)| p.anchor.is_some())
.map(|(i, _)| i)
.collect::<Vec<_>>()
};
assert_eq!(anchors(a), anchors(b));
}
assert_eq!(again.rte.len(), file.rte.len());
let anchors = |s: &crate::TrackSegment, max_zoom: u8| {
s.iter()
.enumerate()
.filter(|(_, p)| p.anchor.is_some_and(|zoom| zoom <= max_zoom))
.map(|(i, _)| i)
.collect::<Vec<_>>()
};
for (a, b) in file.rte.iter().zip(&again.rte) {
assert_eq!(a.info.name, b.info.name);
assert_eq!(a.seg.len(), b.seg.len());
// the route points come back as anchors, the other anchors as points of the path
assert_eq!(
anchors(&a.seg, ROUTE_POINT_MAX_ZOOM),
anchors(&b.seg, u8::MAX)
);
}
}
#[test]
fn test_anchors_shown_from_a_higher_zoom_are_points_of_the_path() {
let (file, categories) = read("simple");
let mut file = as_routes(file);
let segment = &mut file.rte[0].seg;
segment.update_all(|_, point| point.anchor = None);
let len = segment.len();
segment.set_anchor(5, ROUTE_POINT_MAX_ZOOM);
segment.set_anchor(10, ROUTE_POINT_MAX_ZOOM + 1);
let text = written_text(&file, &categories, ExportOptions::ALL);
// the ends, and the anchor shown from zoom 12
assert_eq!(count(&text, "<rtept"), 3);
assert_eq!(count(&text, "<gpxx:rpt"), len - 3);
}
#[test]
fn test_the_ends_of_a_route_are_route_points_even_without_anchors() {
let (mut file, categories) = read("simple");
file.trk[0].trkseg[0].update_all(|_, point| point.anchor = None);
let len = file.trk[0].trkseg[0].len();
let text = written_text(&file, &categories, as_route());
let (file, categories) = read("simple");
let mut file = as_routes(file);
file.rte[0].seg.update_all(|_, point| point.anchor = None);
let len = file.rte[0].seg.len();
let text = written_text(&file, &categories, ExportOptions::ALL);
assert_eq!(count(&text, "<rtept"), 2);
assert_eq!(count(&text, "<gpxx:rpt"), len - 2);
// and a single trackpoint is a route of one point
file.trk[0].trkseg[0].splice(1, len, vec![]);
let text = written_text(&file, &categories, as_route());
file.rte[0].seg.splice(1, len, vec![]);
let text = written_text(&file, &categories, ExportOptions::ALL);
assert_eq!(count(&text, "<rtept"), 1);
assert_eq!(count(&text, "<gpxx:rpt"), 0);
}
#[test]
fn test_empty_segments_are_not_routes() {
let (mut file, categories) = read("simple");
file.trk[0].trkseg.push(crate::TrackSegment::default());
let text = written_text(&file, &categories, as_route());
assert_eq!(count(&text, "<rte>"), 1);
}
#[test]
fn test_routes_have_the_style_of_their_track() {
fn test_routes_keep_their_style() {
let (file, categories) = read("with_style");
let text = written_text(&file, &categories, as_route());
let file = as_routes(file);
let text = written_text(&file, &categories, ExportOptions::ALL);
assert!(count(&text, "<gpx_style:line>") >= 1);
assert_eq!(count(&text, "<gpx_style:line>"), count(&text, "<rte>"));
let again = parse(text.as_bytes(), &mut Default::default()).unwrap();
assert_eq!(again.trk[0].info.color, file.trk[0].info.color);
assert_eq!(again.trk[0].info.width, file.trk[0].info.width);
assert_eq!(again.rte[0].info.color, file.rte[0].info.color);
assert_eq!(again.rte[0].info.width, file.rte[0].info.width);
}
#[test]
fn test_routes_are_written_as_routes_before_the_tracks() {
let (mut file, categories) = read("with_tracks");
let (routes, _) = read("with_routes");
file.rte = routes.rte.clone();
let text = written_text(&file, &categories, ExportOptions::ALL);
assert_eq!(count(&text, "<rte>"), routes.rte.len());
assert_eq!(count(&text, "<trk>"), file.trk.len());
assert!(text.rfind("</rte>").unwrap() < text.find("<trk>").unwrap());
let again = parse(text.as_bytes(), &mut categories.clone()).unwrap();
assert_eq!(again.rte.len(), file.rte.len());
assert_eq!(again.trk.len(), file.trk.len());
for (a, b) in file.rte.iter().zip(&again.rte) {
assert_eq!(a.info, b.info);
same_points(
&a.seg.iter().cloned().collect::<Vec<_>>(),
&b.seg.iter().cloned().collect::<Vec<_>>(),
);
}
same_points(&points(&file), &points(&again));
}
#[test]
fn test_an_empty_route_is_written_and_a_single_route_is_named_like_the_file() {
let mut file = File::default();
file.info.name = "file".into();
file.rte.push(crate::Route::default());
let text = written_text(&file, &Default::default(), ExportOptions::ALL);
assert_eq!(count(&text, "<rte>"), 1);
assert_eq!(count(&text, "<rtept"), 0);
let again = parse(text.as_bytes(), &mut Default::default()).unwrap();
assert_eq!(again.rte.len(), 1);
assert_eq!(again.rte[0].info.name.as_deref(), Some("file"));
// not when there are others
file.trk.push(Track::default());
let text = written_text(&file, &Default::default(), ExportOptions::ALL);
// the metadata only
assert_eq!(count(&text, "<name>file</name>"), 1);
}
#[test]
fn test_route_points_keep_their_data_and_the_options_apply_to_them() {
let (file, categories) = read("with_hr");
let with_all = written_text(&file, &categories, as_route());
let file = as_routes(file);
let with_all = written_text(&file, &categories, ExportOptions::ALL);
assert!(with_all.contains("<gpxtpx:hr>"));
let without = written_text(
&file,
@@ -745,22 +794,10 @@ mod tests {
ExportOptions {
hr: false,
time: false,
as_route: true,
..ExportOptions::ALL
},
);
assert!(!without.contains("<gpxtpx:hr>"));
assert!(!without.contains("<time>"));
}
#[test]
fn test_the_route_option_is_not_data() {
let (file, _) = read("with_route_points");
assert!(!file.exportable_data().as_route);
// it is a way to write the data, never part of them
let all = std::hint::black_box(ExportOptions::ALL);
assert!(!all.as_route && !ExportOptions::NONE.as_route);
assert!(as_route().union(ExportOptions::NONE).as_route);
assert!(!ExportOptions::ALL.union(ExportOptions::NONE).as_route);
}
}
+163
View File
@@ -0,0 +1,163 @@
use std::{collections::HashSet, rc::Rc};
use crate::{Apply, CommandError, FileId, Route, Selection, State, Track};
/// Converts the selected routes to tracks, or the selected tracks or segments to routes, in their
/// file:
///
/// - a route becomes a track with its segment. The new tracks go before the other tracks, right
/// after the routes;
/// - a track becomes one route per segment, as a route has a single segment, and goes away. A
/// selected segment becomes a route and leaves its track, which stays even if it has no segment
/// left. The routes have the information of their track, and go after the other routes;
/// - a file with routes only has all of them converted to tracks, and a file with tracks only
/// has all of them converted to routes. The files with both, or neither, stay as they are.
///
/// The segments keep their ids, the new routes and tracks are selected (the files stay selected).
/// There is nothing to do for the other selections, or for tracks without segments.
#[derive(Debug)]
pub struct Convert;
impl Apply for Convert {
fn apply(self, state: &mut State) -> Result<(), CommandError> {
let selection = state.selection.clone();
let next = match &selection {
Selection::Route { file_id, rte_ids } => {
let file = Rc::make_mut(
state
.files
.get_mut(file_id)
.ok_or(CommandError::NothingToDo)?,
);
let (converted, kept) = std::mem::take(&mut file.rte)
.into_iter()
.partition::<Vec<_>, _>(|rte| rte_ids.contains(&rte.id));
file.rte = kept;
let tracks = to_tracks(converted);
let trk_ids: HashSet<_> = tracks.iter().map(|trk| trk.id).collect();
file.trk.splice(0..0, tracks);
Selection::Track {
file_id: *file_id,
trk_ids,
}
}
Selection::Track { file_id, trk_ids } => {
let file = Rc::make_mut(
state
.files
.get_mut(file_id)
.ok_or(CommandError::NothingToDo)?,
);
let (converted, kept) = std::mem::take(&mut file.trk)
.into_iter()
.partition::<Vec<_>, _>(|trk| trk_ids.contains(&trk.id));
file.trk = kept;
add_routes(&mut file.rte, *file_id, to_routes(converted))
}
Selection::TrackSegment {
file_id,
trk_id,
trkseg_ids,
} => {
let file = Rc::make_mut(
state
.files
.get_mut(file_id)
.ok_or(CommandError::NothingToDo)?,
);
let trk = file
.trk
.iter_mut()
.find(|trk| trk.id == *trk_id)
.ok_or(CommandError::NothingToDo)?;
let (converted, kept) = std::mem::take(&mut trk.trkseg)
.into_iter()
.partition::<Vec<_>, _>(|seg| trkseg_ids.contains(&seg.id));
trk.trkseg = kept;
let info = trk.info.clone();
let routes = converted.into_iter().map(|seg| Route {
id: Default::default(),
info: info.clone(),
seg,
});
add_routes(&mut file.rte, *file_id, routes)
}
Selection::File { file_ids } => {
let mut converted = false;
for file_id in file_ids {
let Some(file) = state.files.get_mut(file_id) else {
continue;
};
if !file.rte.is_empty() && file.trk.is_empty() {
let file = Rc::make_mut(file);
file.trk = to_tracks(std::mem::take(&mut file.rte));
} else if file.rte.is_empty()
&& file.trk.iter().any(|trk| !trk.trkseg.is_empty())
{
let file = Rc::make_mut(file);
file.rte = to_routes(std::mem::take(&mut file.trk)).collect();
} else {
continue;
}
converted = true;
}
return if converted {
Ok(())
} else {
Err(CommandError::NothingToDo)
};
}
Selection::Empty | Selection::Waypoints { .. } | Selection::Waypoint { .. } => {
return Err(CommandError::NothingToDo);
}
};
let empty = match &next {
Selection::Route { rte_ids, .. } => rte_ids.is_empty(),
Selection::Track { trk_ids, .. } => trk_ids.is_empty(),
_ => true,
};
if empty {
// nothing was converted (the changed copy of the file is dropped)
return Err(CommandError::NothingToDo);
}
*state.selection = next;
Ok(())
}
}
/// The tracks of the routes, each with the segment of its route.
fn to_tracks(routes: Vec<Route>) -> Vec<Track> {
routes
.into_iter()
.map(|rte| Track {
id: Default::default(),
info: rte.info,
trkseg: vec![rte.seg],
})
.collect()
}
/// The routes of the tracks, one per segment, each with the information of its track.
fn to_routes(tracks: Vec<Track>) -> impl Iterator<Item = Route> {
tracks.into_iter().flat_map(|trk| {
trk.trkseg.into_iter().map(move |seg| Route {
id: Default::default(),
info: trk.info.clone(),
seg,
})
})
}
/// Adds the routes after the other routes, and gives the selection of them.
fn add_routes(
rte: &mut Vec<Route>,
file_id: FileId,
routes: impl Iterator<Item = Route>,
) -> Selection {
let start = rte.len();
rte.extend(routes);
Selection::Route {
file_id,
rte_ids: rte[start..].iter().map(|rte| rte.id).collect(),
}
}
+24 -7
View File
@@ -1,4 +1,6 @@
use crate::{Apply, CommandError, Edit, Editor, File, State, Track, update_selected};
use crate::{
Apply, CommandError, Edit, Editor, File, Route, State, Track, TrackInfo, update_selected,
};
#[derive(Debug)]
pub struct Metadata<'a> {
@@ -13,21 +15,33 @@ impl Apply for Metadata<'_> {
}
}
impl Metadata<'_> {
fn set(&self, info: &mut TrackInfo) -> Edit {
info.name = Some(self.name.to_owned());
info.desc = Some(self.desc.to_owned());
Edit::Changed
}
}
impl Editor for Metadata<'_> {
fn file(&mut self, file: &mut File) -> Edit {
file.info.name = self.name.to_owned();
file.info.desc = Some(self.desc.to_owned());
// a single track is named like the file
if let [track] = file.trk.as_mut_slice() {
track.info.name = Some(self.name.to_owned());
// a single route or track is named like the file
match (file.rte.as_mut_slice(), file.trk.as_mut_slice()) {
([route], []) => route.info.name = Some(self.name.to_owned()),
([], [track]) => track.info.name = Some(self.name.to_owned()),
_ => {}
}
Edit::Changed
}
fn route(&mut self, route: &mut Route) -> Edit {
self.set(&mut route.info)
}
fn track(&mut self, track: &mut Track) -> Edit {
track.info.name = Some(self.name.to_owned());
track.info.desc = Some(self.desc.to_owned());
Edit::Changed
self.set(&mut track.info)
}
}
@@ -45,12 +59,14 @@ mod tests {
New {
name: "first",
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
New {
name: "second",
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap(); // now selected
@@ -108,6 +124,7 @@ mod tests {
New {
name: "file",
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
+4
View File
@@ -1,9 +1,13 @@
mod convert;
mod metadata;
mod new_route;
mod new_track;
mod new_track_segment;
mod style;
pub use convert::*;
pub use metadata::*;
pub use new_route::*;
pub use new_track::*;
pub use new_track_segment::*;
pub use style::*;
@@ -0,0 +1,55 @@
use crate::{Apply, CommandError, Edit, Editor, File, Route, State, update_selected};
use super::new_track::common_style;
/// Adds an empty route to the selected files, with the style that their routes and tracks agree
/// on (see [`crate::NewTrack`]).
#[derive(Debug)]
pub struct NewRoute;
impl Apply for NewRoute {
fn apply(mut self, state: &mut State) -> Result<(), CommandError> {
update_selected(state, &mut self);
Ok(())
}
}
impl Editor for NewRoute {
fn file(&mut self, file: &mut File) -> Edit {
file.rte.push(Route {
info: common_style(file),
..Default::default()
});
Edit::Changed
}
}
#[cfg(test)]
mod tests {
use crate::{New, engine::command::fixture::Fixture};
use super::*;
#[test]
fn test_new_route_has_the_style_of_the_tracks_and_an_empty_segment() {
let mut fx = Fixture::default();
New {
name: "file",
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
crate::NewTrack.apply(&mut fx.state()).unwrap();
let file = std::rc::Rc::make_mut(fx.files.values_mut().next().unwrap());
file.trk[0].info.color = Some("ff0000".into());
NewRoute.apply(&mut fx.state()).unwrap();
NewRoute.apply(&mut fx.state()).unwrap();
let file = fx.files.values().next().unwrap();
assert_eq!(file.rte.len(), 2);
assert_ne!(file.rte[0].id, file.rte[1].id);
assert!(file.rte[0].seg.is_empty());
assert_eq!(file.rte[1].info.color.as_deref(), Some("ff0000"));
}
}
+34 -11
View File
@@ -1,4 +1,4 @@
use crate::{Apply, CommandError, Edit, Editor, File, State, Track, update_selected};
use crate::{Apply, CommandError, Edit, Editor, File, State, Track, TrackInfo, update_selected};
#[derive(Debug)]
pub struct NewTrack;
@@ -12,20 +12,41 @@ impl Apply for NewTrack {
impl Editor for NewTrack {
fn file(&mut self, file: &mut File) -> Edit {
let mut track = Track::default();
track.info.color = common(file, |t| &t.info.color);
track.info.opacity = common(file, |t| &t.info.opacity);
track.info.width = common(file, |t| &t.info.width);
file.trk.push(track);
file.trk.push(Track {
info: common_style(file),
..Default::default()
});
Edit::Changed
}
}
/// Returns the value of a style attribute if all existing tracks agree on it.
fn common<T: Clone + PartialEq>(file: &File, get: impl Fn(&Track) -> &Option<T>) -> Option<T> {
let mut tracks = file.trk.iter();
let first = get(tracks.next()?);
tracks.all(|t| get(t) == first).then(|| first.clone())?
/// The style that all the routes and tracks of the file agree on: each attribute is set if they
/// all have the same value for it.
pub(super) fn common_style(file: &File) -> TrackInfo {
let infos: Vec<&TrackInfo> = file
.rte
.iter()
.map(|rte| &rte.info)
.chain(file.trk.iter().map(|trk| &trk.info))
.collect();
TrackInfo {
color: common(&infos, |info| &info.color),
opacity: common(&infos, |info| &info.opacity),
width: common(&infos, |info| &info.width),
..Default::default()
}
}
/// Returns the value of a style attribute if all the infos agree on it.
fn common<T: Clone + PartialEq>(
infos: &[&TrackInfo],
get: impl Fn(&TrackInfo) -> &Option<T>,
) -> Option<T> {
let mut infos = infos.iter();
let first = get(infos.next()?);
infos
.all(|info| get(info) == first)
.then(|| first.clone())?
}
#[cfg(test)]
@@ -40,6 +61,7 @@ mod tests {
New {
name: "file",
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
@@ -56,6 +78,7 @@ mod tests {
New {
name: "file",
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
@@ -29,6 +29,7 @@ mod tests {
New {
name: "file",
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
+16 -6
View File
@@ -1,4 +1,4 @@
use crate::{Apply, CommandError, Edit, Editor, State, Track, update_selected};
use crate::{Apply, CommandError, Edit, Editor, Route, State, Track, TrackInfo, update_selected};
#[derive(Debug)]
pub struct Style<'a> {
@@ -14,21 +14,31 @@ impl Apply for Style<'_> {
}
}
impl Editor for Style<'_> {
fn track(&mut self, track: &mut Track) -> Edit {
impl Style<'_> {
fn set(&self, info: &mut TrackInfo) -> Edit {
if let Some(color) = self.color {
track.info.color = Some(color.to_owned());
info.color = Some(color.to_owned());
}
if self.opacity.is_some() {
track.info.opacity = self.opacity;
info.opacity = self.opacity;
}
if self.width.is_some() {
track.info.width = self.width;
info.width = self.width;
}
Edit::Changed
}
}
impl Editor for Style<'_> {
fn route(&mut self, route: &mut Route) -> Edit {
self.set(&mut route.info)
}
fn track(&mut self, track: &mut Track) -> Edit {
self.set(&mut track.info)
}
}
#[cfg(test)]
mod tests {
use std::collections::HashSet;
+18 -1
View File
@@ -3,7 +3,7 @@ use std::rc::Rc;
use crate::{Apply, CommandError, FileId, Files, Selection, State, Waypoint};
/// Deletes the selected elements. With `whole_files`, the files holding the selected elements
/// are deleted instead, even if only a track or a waypoint is selected.
/// are deleted instead, even if only a route, a track or a waypoint is selected.
#[derive(Debug)]
pub struct Delete {
pub whole_files: bool,
@@ -23,6 +23,21 @@ impl Apply for Delete {
state.order.0.retain(|id| !file_ids.contains(id));
Selection::Empty
}
Selection::Route { file_id, rte_ids } => {
let file = state
.files
.get_mut(file_id)
.ok_or(CommandError::NothingToDo)?;
let file = Rc::make_mut(file);
let len = file.rte.len();
file.rte.retain(|rte| !rte_ids.contains(&rte.id));
if file.rte.len() == len {
return Err(CommandError::NothingToDo);
}
Selection::File {
file_ids: [*file_id].into(),
}
}
Selection::Track { file_id, trk_ids } => {
let file = state
.files
@@ -131,6 +146,7 @@ mod tests {
crate::New {
name: "keep",
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
@@ -265,6 +281,7 @@ mod tests {
crate::New {
name: "keep",
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
@@ -31,12 +31,14 @@ mod tests {
crate::New {
name: "a",
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
crate::New {
name: "b",
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
+27 -2
View File
@@ -1,8 +1,8 @@
use std::{collections::HashSet, rc::Rc};
use crate::{
Apply, CommandError, FileId, Files, Selection, State, Waypoint, copy_file, copy_segment,
copy_track, copy_waypoint,
Apply, CommandError, FileId, Files, Selection, State, Waypoint, copy_file, copy_route,
copy_segment, copy_track, copy_waypoint,
};
#[derive(Debug)]
@@ -35,6 +35,30 @@ impl Apply for Duplicate {
state.order.0.splice(at..at, order);
Selection::File { file_ids: copies }
}
Selection::Route { file_id, rte_ids } => {
let file = state
.files
.get_mut(file_id)
.ok_or(CommandError::NothingToDo)?;
let file = Rc::make_mut(file);
let mut copies = HashSet::new();
append_copies(
&mut file.rte,
|rte| rte_ids.contains(&rte.id),
|rte| {
let copy = copy_route(rte);
copies.insert(copy.id);
copy
},
);
if copies.is_empty() {
return Err(CommandError::NothingToDo);
}
Selection::Route {
file_id: *file_id,
rte_ids: copies,
}
}
Selection::Track { file_id, trk_ids } => {
let file = state
.files
@@ -193,6 +217,7 @@ mod tests {
crate::New {
name,
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
+3
View File
@@ -85,6 +85,7 @@ mod tests {
crate::New {
name: "keep",
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
@@ -140,6 +141,7 @@ mod tests {
crate::New {
name: "before",
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
@@ -189,6 +191,7 @@ mod tests {
crate::New {
name: "keep",
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
@@ -9,6 +9,8 @@ use crate::{
pub enum MoveTarget {
/// The list of files.
Files { index: usize },
/// The routes of a file.
Routes { file_id: FileId, index: usize },
/// The tracks of a file.
Tracks { file_id: FileId, index: usize },
/// The segments of a track.
@@ -26,9 +28,10 @@ pub enum MoveTarget {
///
/// What can go where:
/// - files: among the files;
/// - tracks: among the files (each track becomes a file), or among the tracks of a file;
/// - segments: among the files (each segment becomes a file), among the tracks of a file (each
/// segment becomes a track), or among the segments of a track;
/// - routes and tracks: among the files (each one becomes a file), or among the routes or the
/// tracks of a file, where they keep their kind (see [`Destination`]);
/// - segments: among the files (each segment becomes a file), among the routes or the tracks of a
/// file (each segment becomes a route or a track), or among the segments of a track;
/// - waypoints, or all the waypoints of a file (the waypoints node): among the waypoints of a
/// file.
#[derive(Debug)]
@@ -41,6 +44,9 @@ impl From<MoveTarget> for Destination {
fn from(target: MoveTarget) -> Self {
match target {
MoveTarget::Files { index } => Destination::Files(Place::Index(index)),
MoveTarget::Routes { file_id, index } => {
Destination::Routes(file_id, Place::Index(index))
}
MoveTarget::Tracks { file_id, index } => {
Destination::Tracks(file_id, Place::Index(index))
}
+53 -6
View File
@@ -1,12 +1,16 @@
use crate::{Apply, CommandError, File, LngLat, State, Track, TrackSegment, Trackpoint, produce};
use crate::{
Apply, CommandError, File, LngLat, Route, State, Track, TrackSegment, Trackpoint, produce,
};
/// Creates an empty file and selects it. With a `trackpoint`, the file starts with a track and a
/// segment that hold it: it is how the routing tool starts a route on an empty map, in a single
/// undo step.
/// Creates an empty file and selects it. With `route`, the file starts with a route, which is
/// empty. With a `trackpoint`, the file starts with a segment that holds it, the one of its route
/// with `route` or of a track otherwise: it is how the routing tool starts a route on an empty map,
/// in a single undo step.
#[derive(Debug)]
pub struct New<'a> {
pub name: &'a str,
pub trackpoint: Option<NewTrackpoint>,
pub route: bool,
}
#[derive(Debug, Clone, Copy)]
@@ -21,8 +25,8 @@ impl Apply for New<'_> {
produce(state, |_| {
let mut file = File::default();
file.info.name = self.name.to_owned();
let mut segment = TrackSegment::default();
if let Some(point) = self.trackpoint {
let mut segment = TrackSegment::default();
// the ends of a segment are anchors of the routing tool
segment.splice(
0,
@@ -36,6 +40,13 @@ impl Apply for New<'_> {
..Default::default()
}],
);
}
if self.route {
file.rte.push(Route {
seg: segment,
..Default::default()
});
} else if self.trackpoint.is_some() {
file.trk.push(Track {
trkseg: vec![segment],
..Default::default()
@@ -59,7 +70,8 @@ mod tests {
assert!(
New {
name: "new",
trackpoint: None
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.is_ok()
@@ -83,6 +95,7 @@ mod tests {
New {
name: "new",
trackpoint: Some(point),
route: false,
}
.apply(&mut fx.state())
.unwrap();
@@ -98,4 +111,38 @@ mod tests {
assert_eq!(segment[0].anchor, Some(0));
assert_eq!(fx.selected_files(), [file.id].into());
}
#[test]
fn test_new_with_a_route() {
let mut fx = Fixture::default();
New {
name: "new",
trackpoint: None,
route: true,
}
.apply(&mut fx.state())
.unwrap();
let file = fx.files.values().next().unwrap();
assert_eq!((file.rte.len(), file.trk.len()), (1, 0));
assert!(file.rte[0].seg.is_empty());
assert_eq!(fx.selected_files(), [file.id].into());
// the trackpoint goes to the route
let mut fx = Fixture::default();
New {
name: "new",
trackpoint: Some(NewTrackpoint {
lng: 1.0,
lat: 2.0,
ele: 3.0,
}),
route: true,
}
.apply(&mut fx.state())
.unwrap();
let file = fx.files.values().next().unwrap();
assert_eq!((file.rte.len(), file.trk.len()), (1, 0));
assert_eq!(file.rte[0].seg.len(), 1);
assert_eq!(file.rte[0].seg[0].anchor, Some(0));
}
}
+41 -10
View File
@@ -1,18 +1,21 @@
use std::{collections::HashSet, hash::Hash};
use crate::{
Apply, ClipboardContent, CommandError, Destination, FileId, Place, Selection, State, TrackId,
TrackSegmentId, WaypointId, transfer, waypoint_ids,
Apply, ClipboardContent, CommandError, Destination, FileId, Place, RouteId, Selection, State,
TrackId, TrackSegmentId, WaypointId, transfer, waypoint_ids,
};
/// Pastes what was copied or cut (see [`crate::Clipboard`]).
///
/// Where it goes depends on the selection, see [`crate::Clipboard::can_paste`]:
/// - nothing selected: the files are copied, and tracks and segments become new files (named
/// like their track, or like the file they come from);
/// - a file: the tracks are added to it, segments become new tracks in it, and the waypoints are
/// added to its own;
/// - a track: the segments are added to it, tracks go after it in its file;
/// - nothing selected: the files are copied, and routes, tracks and segments become new files
/// (named like their route or track, or like the file they come from);
/// - a file: the routes and tracks are added to it, segments become new tracks in it, and the
/// waypoints are added to its own;
/// - a route: routes, and segments as new routes, go after it in its file, tracks go before the
/// other tracks (right after the routes);
/// - a track: the segments are added to it, tracks go after it in its file, and routes after the
/// other routes;
/// - a segment: the segments go after it in its track;
/// - the waypoints or a waypoint: the waypoints are added to their file, after the waypoint.
///
@@ -46,6 +49,14 @@ impl Apply for Paste {
}
}
fn route_ids_of(state: &State, file_id: FileId) -> Vec<RouteId> {
state
.files
.get(&file_id)
.map(|file| file.rte.iter().map(|rte| rte.id).collect())
.unwrap_or_default()
}
fn track_ids_of(state: &State, file_id: FileId) -> Vec<TrackId> {
state
.files
@@ -79,14 +90,34 @@ fn destination(state: &State, content: &ClipboardContent) -> Result<Destination,
};
Ok(match (content, &*state.selection) {
(ClipboardContent::Files(_), _) => Destination::Files(Place::End),
(ClipboardContent::Tracks(_) | ClipboardContent::Segments(_), Selection::Empty) => {
Destination::Files(Place::End)
(
ClipboardContent::Routes(_)
| ClipboardContent::Tracks(_)
| ClipboardContent::Segments(_),
Selection::Empty,
) => Destination::Files(Place::End),
(ClipboardContent::Routes(_), Selection::File { file_ids }) => {
Destination::Routes(last_file(file_ids)?, Place::End)
}
// routes and tracks keep their kind (see `Destination`)
(
ClipboardContent::Routes(_)
| ClipboardContent::Tracks(_)
| ClipboardContent::Segments(_),
Selection::Route { file_id, rte_ids },
) => {
let last =
last_selected(route_ids_of(state, *file_id).into_iter(), rte_ids).ok_or(nothing)?;
Destination::Routes(*file_id, Place::After(last))
}
(
ClipboardContent::Tracks(_) | ClipboardContent::Segments(_),
Selection::File { file_ids },
) => Destination::Tracks(last_file(file_ids)?, Place::End),
(ClipboardContent::Tracks(_), Selection::Track { file_id, trk_ids }) => {
(
ClipboardContent::Routes(_) | ClipboardContent::Tracks(_),
Selection::Track { file_id, trk_ids },
) => {
let last =
last_selected(track_ids_of(state, *file_id).into_iter(), trk_ids).ok_or(nothing)?;
Destination::Tracks(*file_id, Place::After(last))
+6 -9
View File
@@ -7,8 +7,8 @@ use crate::{Apply, CommandError, FileId, Load, State};
pub struct ExternalFile<'a> {
pub id: FileId,
pub data: &'a [u8],
/// Name of the file outside (the name of the file on the disk, without its extension), which
/// is its name here whatever the data says: a file has a single name.
/// Name of the file when the data has none (the name of the file on the disk, without its
/// extension).
pub name: &'a str,
}
@@ -16,8 +16,7 @@ pub struct ExternalFile<'a> {
/// that are given replace the ones of the same id (which stay where they are in the order, and keep
/// their id), or are added at the end if there are none; the ones to `remove` are deleted.
///
/// Unlike [`crate::LoadFiles`], the files are named as they are outside (see [`ExternalFile`]), and
/// the selection is left alone (what it holds of a file that was
/// Unlike [`crate::LoadFiles`], the selection is left alone (what it holds of a file that was
/// replaced or deleted is dropped by the engine afterwards, as with any edit).
///
/// The files that cannot be read are skipped, and the files that are here stay as they are. It is an
@@ -40,9 +39,6 @@ impl Apply for SyncFiles<'_> {
match load.parse(state.categories) {
Ok(mut file) => {
file.id = external.id;
if !external.name.trim().is_empty() {
file.info.name = external.name.to_owned();
}
if state.files.insert(external.id, Rc::new(file)).is_none() {
state.order.0.push(external.id);
}
@@ -80,6 +76,7 @@ mod tests {
New {
name,
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
@@ -106,8 +103,8 @@ mod tests {
.apply(&mut fx.state())
.unwrap();
assert_eq!(fx.order.0, [first, second]);
// named as it is outside, not as the data says
assert_eq!(fx.files[&first].info.name, "name");
// named as the data says, the name outside is only for a file that has none
assert_eq!(fx.files[&first].info.name, "from disk");
assert_eq!(fx.files[&first].id, first);
// the selection is not changed
assert_eq!(fx.selected_files(), [second].into());
+131 -30
View File
@@ -5,9 +5,9 @@
use std::{collections::HashSet, rc::Rc};
use crate::{
ClipboardContent, ClipboardSegment, ClipboardTrack, CommandError, File, FileId, Selection,
State, Track, TrackId, TrackSegment, TrackSegmentId, Waypoint, WaypointId, copy_file,
copy_segment, copy_track, copy_waypoint,
ClipboardContent, ClipboardRoute, ClipboardSegment, ClipboardTrack, CommandError, File, FileId,
Route, RouteId, Selection, State, Track, TrackId, TrackSegment, TrackSegmentId, Waypoint,
WaypointId, copy_file, copy_route, copy_segment, copy_track, copy_waypoint,
};
/// Where elements are put, among the others of the same list.
@@ -37,15 +37,20 @@ impl<T: Copy + PartialEq> Place<T> {
/// Where elements go. What can go where:
/// - files: among the files;
/// - routes: among the files (each route becomes a file, named like the route, or like the file
/// it comes from), or among the routes of a file. Among the tracks of a file, they stay routes,
/// which come before the tracks: they go after the other routes;
/// - tracks: among the files (each track becomes a file, named like the track, or like the file
/// it comes from), or among the tracks of a file;
/// it comes from), or among the tracks of a file. Among the routes of a file, they stay tracks,
/// which come after the routes: they go before the other tracks;
/// - segments: among the files (each segment becomes a file with a track, named like its track,
/// or like its file), among the tracks of a file (each segment becomes a track), or among the
/// segments of a track;
/// or like its file), among the routes or the tracks of a file (each segment becomes a route or
/// a track), or among the segments of a track;
/// - waypoints: among the waypoints of a file.
#[derive(Debug, Clone, Copy)]
pub enum Destination {
Files(Place<FileId>),
Routes(FileId, Place<RouteId>),
Tracks(FileId, Place<TrackId>),
Segments(FileId, TrackId, Place<TrackSegmentId>),
Waypoints(FileId, Place<WaypointId>),
@@ -93,6 +98,14 @@ pub fn remove_elements(state: &mut State, content: &ClipboardContent) {
state.files.retain(|id, _| !ids.contains(id));
state.order.0.retain(|id| !ids.contains(id));
}
ClipboardContent::Routes(routes) => {
let ids: HashSet<RouteId> = routes.iter().map(|rte| rte.route.id).collect();
for file in state.files.values_mut() {
if file.rte.iter().any(|rte| ids.contains(&rte.id)) {
Rc::make_mut(file).rte.retain(|rte| !ids.contains(&rte.id));
}
}
}
ClipboardContent::Tracks(tracks) => {
let ids: HashSet<TrackId> = tracks.iter().map(|trk| trk.track.id).collect();
for file in state.files.values_mut() {
@@ -141,7 +154,9 @@ pub fn transfer(
// the destination has to be there, before anything is changed
let exists = match to {
Destination::Files(_) => true,
Destination::Tracks(file_id, _) => state.files.contains_key(&file_id),
Destination::Routes(file_id, _) | Destination::Tracks(file_id, _) => {
state.files.contains_key(&file_id)
}
Destination::Segments(file_id, trk_id, _) => state
.files
.get(&file_id)
@@ -152,17 +167,21 @@ pub fn transfer(
(content, to),
(ClipboardContent::Files(_), Destination::Files(_))
| (
ClipboardContent::Tracks(_),
Destination::Files(_) | Destination::Tracks(..)
ClipboardContent::Routes(_) | ClipboardContent::Tracks(_),
Destination::Files(_) | Destination::Routes(..) | Destination::Tracks(..)
)
| (
ClipboardContent::Segments(_),
Destination::Files(_) | Destination::Tracks(..) | Destination::Segments(..),
Destination::Files(_)
| Destination::Routes(..)
| Destination::Tracks(..)
| Destination::Segments(..),
)
| (ClipboardContent::Waypoints(_), Destination::Waypoints(..))
);
let empty = match content {
ClipboardContent::Files(files) => files.is_empty(),
ClipboardContent::Routes(routes) => routes.is_empty(),
ClipboardContent::Tracks(tracks) => tracks.is_empty(),
ClipboardContent::Segments(segments) => segments.is_empty(),
ClipboardContent::Waypoints(waypoints) => waypoints.is_empty(),
@@ -189,6 +208,7 @@ pub fn transfer(
.collect();
Ok(add_files(state, files, place))
}
(ClipboardContent::Routes(routes), to) => transfer_routes(state, routes, copy, to),
(ClipboardContent::Tracks(tracks), to) => transfer_tracks(state, tracks, copy, to),
(ClipboardContent::Segments(segments), to) => transfer_segments(state, segments, copy, to),
(ClipboardContent::Waypoints(waypoints), Destination::Waypoints(file_id, place)) => {
@@ -198,6 +218,89 @@ pub fn transfer(
}
}
/// Puts the routes among the routes of a file, at `place`, and gives the selection of them.
fn insert_routes(
state: &mut State,
file_id: FileId,
place: Place<RouteId>,
routes: Vec<Route>,
) -> Result<Selection, CommandError> {
let ids: HashSet<RouteId> = routes.iter().map(|rte| rte.id).collect();
let file = file_mut(state, file_id)?;
let existing: Vec<RouteId> = file.rte.iter().map(|rte| rte.id).collect();
let at = place.position(&existing);
file.rte.splice(at..at, routes);
Ok(Selection::Route {
file_id,
rte_ids: ids,
})
}
fn transfer_routes(
state: &mut State,
routes: &[ClipboardRoute],
copy: bool,
to: Destination,
) -> Result<Selection, CommandError> {
let routes: Vec<(Route, &str)> = routes
.iter()
.map(|clip| {
let route = if copy {
copy_route(&clip.route)
} else {
clip.route.clone()
};
(route, clip.file_name.as_str())
})
.collect();
match to {
Destination::Files(place) => {
let files = routes
.into_iter()
.map(|(route, file_name)| {
let name = route
.info
.name
.clone()
.unwrap_or_else(|| file_name.to_owned());
Rc::new(File {
rte: vec![route],
..new_file(&name, vec![])
})
})
.collect();
Ok(add_files(state, files, place))
}
Destination::Routes(file_id, place) => {
let routes = routes.into_iter().map(|(route, _)| route).collect();
insert_routes(state, file_id, place, routes)
}
Destination::Tracks(file_id, _) => {
let routes = routes.into_iter().map(|(route, _)| route).collect();
insert_routes(state, file_id, Place::End, routes)
}
_ => Err(CommandError::NothingToDo),
}
}
/// Puts the tracks among the tracks of a file, at `place`, and gives the selection of them.
fn insert_tracks(
state: &mut State,
file_id: FileId,
place: Place<TrackId>,
tracks: Vec<Track>,
) -> Result<Selection, CommandError> {
let ids: HashSet<TrackId> = tracks.iter().map(|trk| trk.id).collect();
let file = file_mut(state, file_id)?;
let existing: Vec<TrackId> = file.trk.iter().map(|trk| trk.id).collect();
let at = place.position(&existing);
file.trk.splice(at..at, tracks);
Ok(Selection::Track {
file_id,
trk_ids: ids,
})
}
fn transfer_tracks(
state: &mut State,
tracks: &[ClipboardTrack],
@@ -231,16 +334,12 @@ fn transfer_tracks(
Ok(add_files(state, files, place))
}
Destination::Tracks(file_id, place) => {
let tracks: Vec<Track> = tracks.into_iter().map(|(track, _)| track).collect();
let ids: HashSet<TrackId> = tracks.iter().map(|trk| trk.id).collect();
let file = file_mut(state, file_id)?;
let existing: Vec<TrackId> = file.trk.iter().map(|trk| trk.id).collect();
let at = place.position(&existing);
file.trk.splice(at..at, tracks);
Ok(Selection::Track {
file_id,
trk_ids: ids,
})
let tracks = tracks.into_iter().map(|(track, _)| track).collect();
insert_tracks(state, file_id, place, tracks)
}
Destination::Routes(file_id, _) => {
let tracks = tracks.into_iter().map(|(track, _)| track).collect();
insert_tracks(state, file_id, Place::Index(0), tracks)
}
_ => Err(CommandError::NothingToDo),
}
@@ -276,20 +375,22 @@ fn transfer_segments(
.collect();
Ok(add_files(state, files, place))
}
Destination::Routes(file_id, place) => {
let routes = segments
.into_iter()
.map(|(seg, _)| Route {
seg,
..Default::default()
})
.collect();
insert_routes(state, file_id, place, routes)
}
Destination::Tracks(file_id, place) => {
let tracks: Vec<Track> = segments
let tracks = segments
.into_iter()
.map(|(segment, _)| own_track(segment))
.collect();
let ids = tracks.iter().map(|trk| trk.id).collect();
let file = file_mut(state, file_id)?;
let existing: Vec<TrackId> = file.trk.iter().map(|trk| trk.id).collect();
let at = place.position(&existing);
file.trk.splice(at..at, tracks);
Ok(Selection::Track {
file_id,
trk_ids: ids,
})
insert_tracks(state, file_id, place, tracks)
}
Destination::Segments(file_id, trk_id, place) => {
let segments: Vec<TrackSegment> =
+8 -2
View File
@@ -4,6 +4,8 @@ mod file;
#[cfg(test)]
pub(crate) mod fixture;
mod pattern;
#[cfg(test)]
mod routes_tests;
mod tools;
pub use apply::*;
@@ -31,13 +33,15 @@ pub enum Command<'a> {
Move(Move),
DeleteAll(DeleteAll),
DeleteWaypoint(DeleteWaypoint),
Convert(Convert),
Metadata(Metadata<'a>),
Style(Style<'a>),
NewRoute(NewRoute),
NewTrack(NewTrack),
NewTrackSegment(NewTrackSegment),
Reverse(Reverse),
RoundTrip(RoundTrip),
Route(Route<'a>),
Reroute(Reroute<'a>),
NewWaypoint(NewWaypoint<'a>),
EditWaypoint(EditWaypoint<'a>),
MoveWaypoint(MoveWaypoint),
@@ -66,13 +70,15 @@ impl Apply for Command<'_> {
Command::Move(c) => c.apply(state),
Command::DeleteAll(c) => c.apply(state),
Command::DeleteWaypoint(c) => c.apply(state),
Command::Convert(c) => c.apply(state),
Command::Metadata(c) => c.apply(state),
Command::Style(c) => c.apply(state),
Command::NewRoute(c) => c.apply(state),
Command::NewTrack(c) => c.apply(state),
Command::NewTrackSegment(c) => c.apply(state),
Command::Reverse(c) => c.apply(state),
Command::RoundTrip(c) => c.apply(state),
Command::Route(c) => c.apply(state),
Command::Reroute(c) => c.apply(state),
Command::NewWaypoint(c) => c.apply(state),
Command::EditWaypoint(c) => c.apply(state),
Command::MoveWaypoint(c) => c.apply(state),
+12 -2
View File
@@ -1,4 +1,4 @@
use crate::{File, Track, TrackSegment, Waypoint};
use crate::{File, Route, Track, TrackSegment, Waypoint};
// The trackpoints of a segment are shared chunks, so copying elements is cheap.
@@ -18,10 +18,20 @@ pub fn copy_track(track: &Track) -> Track {
}
}
/// A copy of the file and of its tracks and segments, with new ids.
/// A copy of the route and of its segment, with new ids.
pub fn copy_route(route: &Route) -> Route {
Route {
id: Default::default(),
seg: copy_segment(&route.seg),
..route.clone()
}
}
/// A copy of the file and of its routes, tracks and segments, with new ids.
pub fn copy_file(file: &File) -> File {
File {
id: Default::default(),
rte: file.rte.iter().map(copy_route).collect(),
trk: file.trk.iter().map(copy_track).collect(),
..file.clone()
}
@@ -18,7 +18,7 @@ pub fn update_segments(
let file = files
.entry(location.file_id)
.or_insert_with(|| (*state.files[&location.file_id]).clone());
f(i, &mut file.trk[location.trk].trkseg[location.seg]);
f(i, file.segment_mut(location.index));
}
for (file_id, file) in files {
state.files.insert(file_id, Rc::new(file));
@@ -31,6 +31,8 @@ mod tests {
use crate::engine::command::fixture::Fixture;
use crate::SegmentIndex;
use super::*;
fn fixture() -> (Fixture, FileId) {
@@ -49,13 +51,11 @@ mod tests {
let locations = [
SegmentLocation {
file_id: id,
trk: 1,
seg: 0,
index: SegmentIndex::Track { trk: 1, seg: 0 },
},
SegmentLocation {
file_id: id,
trk: 0,
seg: 0,
index: SegmentIndex::Track { trk: 0, seg: 0 },
},
];
let mut calls = vec![];
@@ -1,6 +1,6 @@
use std::rc::Rc;
use crate::{File, FileId, Files, Selection, State, Track, TrackSegment, Waypoint};
use crate::{File, FileId, Files, Route, Selection, State, Track, TrackSegment, Waypoint};
/// What an [`Editor`] hook did to the element it was given.
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
@@ -21,13 +21,19 @@ impl Edit {
/// Independent edit of each selected element, without needing the rest of the state.
///
/// A selected file, track or segment is passed to the hook of its own level. By default a
/// file is forwarded to its tracks and a track to its segments, so an editor only overrides
/// the level(s) it cares about (metadata: file and track, style: track, reverse: segment).
/// Waypoints are only edited when selected, never through their file.
/// A selected file, route, track or segment is passed to the hook of its own level. By default
/// a file is forwarded to its routes and tracks, and a route or a track to its segments, so an
/// editor only overrides the level(s) it cares about (metadata: file, route and track, style:
/// route and track, reverse: segment). Waypoints are only edited when selected, never through
/// their file.
pub trait Editor {
fn file(&mut self, file: &mut File) -> Edit {
edit_each(&mut file.trk, |trk| self.track(trk))
let routes = edit_each(&mut file.rte, |rte| self.route(rte));
routes.merge(edit_each(&mut file.trk, |trk| self.track(trk)))
}
fn route(&mut self, route: &mut Route) -> Edit {
edit_segment(self, &mut route.seg)
}
fn track(&mut self, track: &mut Track) -> Edit {
@@ -95,7 +101,7 @@ fn edit_waypoints<E: Editor + ?Sized>(
}
}
/// Applies the editor to every selected file, track or segment.
/// Applies the editor to every selected file, route, track or segment.
pub fn update_selected<E: Editor>(state: &mut State, editor: &mut E) {
match &*state.selection {
Selection::File { file_ids } => {
@@ -103,6 +109,15 @@ pub fn update_selected<E: Editor>(state: &mut State, editor: &mut E) {
update_file(state.files, *id, |file| editor.file(file));
}
}
Selection::Route { file_id, rte_ids } => {
update_file(state.files, *file_id, |file| {
edit_where(
&mut file.rte,
|rte| rte_ids.contains(&rte.id),
|rte| editor.route(rte),
)
});
}
Selection::Track { file_id, trk_ids } => {
update_file(state.files, *file_id, |file| {
edit_where(
+724
View File
@@ -0,0 +1,724 @@
//! The commands on routes: they work as on tracks, a route being a track of a single segment.
use std::rc::Rc;
use crate::{
Apply, Clean, CleanType, Clipboard, ClipboardContent, CommandError, Convert, Crop, Delete,
Duplicate, Extract, File, FileId, LngLat, LngLatBounds, Merge, MergeType, Metadata, Move,
MoveTarget, Paste, Reverse, RouteId, SegmentIndex, Selection, Split, SplitType, Style,
engine::command::fixture::Fixture, parse,
};
/// A file with the 2 routes of `with_routes` and the 2 tracks (of 2 segments) of
/// `with_tracks_and_segments`, which is the only file and is in the order.
fn fixture() -> (Fixture, FileId) {
let read = |name: &str| {
let data = std::fs::read(format!("data/{name}.gpx")).unwrap();
parse(&data, &mut Default::default()).unwrap()
};
let mut file = read("with_tracks_and_segments");
file.rte = read("with_routes").rte;
assert_eq!((file.rte.len(), file.trk.len()), (2, 2));
let mut fx = Fixture::default();
let id = file.id;
fx.files.insert(id, Rc::new(file));
fx.order.0.push(id);
(fx, id)
}
fn file(fx: &Fixture, id: FileId) -> &File {
&fx.files[&id]
}
fn routes(fx: &Fixture, id: FileId, indices: &[usize]) -> Selection {
Selection::Route {
file_id: id,
rte_ids: indices.iter().map(|i| file(fx, id).rte[*i].id).collect(),
}
}
fn whole(id: FileId) -> Selection {
Selection::File {
file_ids: [id].into(),
}
}
fn lengths(file: &File) -> Vec<usize> {
file.segments().map(|segment| segment.len()).collect()
}
#[test]
fn test_the_segments_of_a_file_are_the_ones_of_its_routes_then_its_tracks() {
let (fx, id) = fixture();
let file = file(&fx, id);
let locations = whole(id).segment_locations(&fx.files, &fx.order.0);
let indices: Vec<SegmentIndex> = locations.iter().map(|l| l.index).collect();
assert_eq!(
indices,
[
SegmentIndex::Route(0),
SegmentIndex::Route(1),
SegmentIndex::Track { trk: 0, seg: 0 },
SegmentIndex::Track { trk: 0, seg: 1 },
SegmentIndex::Track { trk: 1, seg: 0 },
SegmentIndex::Track { trk: 1, seg: 1 },
]
);
// a selection of routes covers their segments only
let selected = routes(&fx, id, &[1]).segment_locations(&fx.files, &fx.order.0);
assert_eq!(selected.len(), 1);
assert_eq!(selected[0].segment(&fx.files).id, file.rte[1].seg.id);
}
#[test]
fn test_routes_are_selected_like_tracks() {
let (fx, id) = fixture();
let ids: Vec<RouteId> = file(&fx, id).rte.iter().map(|r| r.id).collect();
let mut selection = routes(&fx, id, &[0]);
selection.extend(routes(&fx, id, &[1]));
assert_eq!(selection, routes(&fx, id, &[0, 1]));
selection.toggle(routes(&fx, id, &[0]));
assert_eq!(selection, routes(&fx, id, &[1]));
// routes and tracks are not selected together
let track = Selection::Track {
file_id: id,
trk_ids: [file(&fx, id).trk[0].id].into(),
};
selection.extend(track.clone());
assert_eq!(selection, track);
assert_eq!(
routes(&fx, id, &[0]).all_at_level(&fx.files, &fx.order.0),
Some(routes(&fx, id, &[0, 1]))
);
assert_eq!(
routes(&fx, id, &[0]).neighbour(&fx.files, &fx.order.0, true),
Some(routes(&fx, id, &[1]))
);
assert_eq!(
routes(&fx, id, &[1]).neighbour(&fx.files, &fx.order.0, true),
None
);
assert_eq!(routes(&fx, id, &[1]).to_files(), whole(id));
let mut gone = Selection::Route {
file_id: id,
rte_ids: [ids[0], RouteId::default()].into(),
};
gone.retain_existing(&fx.files);
assert_eq!(gone, routes(&fx, id, &[0]));
}
#[test]
fn test_delete_and_duplicate_routes() {
let (mut fx, id) = fixture();
let kept = file(&fx, id).rte[1].id;
fx.selection = routes(&fx, id, &[0]);
Delete { whole_files: false }
.apply(&mut fx.state())
.unwrap();
assert_eq!(file(&fx, id).rte.len(), 1);
assert_eq!(file(&fx, id).rte[0].id, kept);
assert_eq!(file(&fx, id).trk.len(), 2);
assert_eq!(fx.selection, whole(id));
fx.selection = routes(&fx, id, &[0]);
Duplicate.apply(&mut fx.state()).unwrap();
let file = file(&fx, id);
assert_eq!(file.rte.len(), 2);
assert_ne!(file.rte[1].id, kept);
assert_ne!(file.rte[1].seg.id, file.rte[0].seg.id);
assert_eq!(file.rte[1].info, file.rte[0].info);
assert_eq!(
fx.selection,
Selection::Route {
file_id: id,
rte_ids: [file.rte[1].id].into()
}
);
}
#[test]
fn test_metadata_and_style_of_routes() {
let (mut fx, id) = fixture();
fx.selection = routes(&fx, id, &[1]);
Metadata {
name: "renamed",
desc: "about",
}
.apply(&mut fx.state())
.unwrap();
Style {
color: Some("00ff00"),
opacity: None,
width: Some(3.0),
}
.apply(&mut fx.state())
.unwrap();
let file = file(&fx, id);
let info = &file.rte[1].info;
assert_eq!(info.name.as_deref(), Some("renamed"));
assert_eq!(info.desc.as_deref(), Some("about"));
assert_eq!(info.color.as_deref(), Some("00ff00"));
assert_eq!(info.width, Some(3.0));
assert_ne!(file.rte[0].info.name.as_deref(), Some("renamed"));
// the style of a file is the one of its routes and tracks
fx.selection = whole(id);
Style {
color: Some("ff0000"),
opacity: None,
width: None,
}
.apply(&mut fx.state())
.unwrap();
let file = &fx.files[&id];
assert!(
file.rte
.iter()
.map(|r| &r.info)
.chain(file.trk.iter().map(|t| &t.info))
.all(|info| info.color.as_deref() == Some("ff0000"))
);
}
#[test]
fn test_copy_and_paste_routes() {
let (mut fx, id) = fixture();
let copied = file(&fx, id).rte[0].clone();
fx.clipboard = Clipboard::new(&routes(&fx, id, &[0]), &fx.files, &fx.order.0, false);
assert!(matches!(
fx.clipboard.as_ref().unwrap().content,
ClipboardContent::Routes(_)
));
// after the selected route
fx.selection = routes(&fx, id, &[1]);
Paste.apply(&mut fx.state()).unwrap();
let file = file(&fx, id);
assert_eq!(file.rte.len(), 3);
assert_ne!(file.rte[2].id, copied.id);
assert_eq!(file.rte[2].seg.len(), copied.seg.len());
assert!(
matches!(&fx.selection, Selection::Route { rte_ids, .. } if rte_ids.contains(&file.rte[2].id))
);
// nothing selected: a new file, named like the route
fx.selection = Selection::Empty;
Paste.apply(&mut fx.state()).unwrap();
assert_eq!(fx.order.0.len(), 2);
let new = &fx.files[&fx.order.0[1]];
assert_eq!(new.rte.len(), 1);
assert!(new.trk.is_empty());
assert_eq!(Some(&new.info.name), copied.info.name.as_ref());
// a track selected: among the routes of its file
fx.selection = Selection::Track {
file_id: id,
trk_ids: [fx.files[&id].trk[0].id].into(),
};
Paste.apply(&mut fx.state()).unwrap();
assert_eq!(fx.files[&id].rte.len(), 4);
// segments do not go in a route, but become routes next to it
let file = &fx.files[&id];
let track = &file.trk[0];
fx.clipboard = Clipboard::new(
&Selection::TrackSegment {
file_id: id,
trk_id: track.id,
trkseg_ids: [track.trkseg[1].id].into(),
},
&fx.files,
&fx.order.0,
false,
);
fx.selection = routes(&fx, id, &[0]);
Paste.apply(&mut fx.state()).unwrap();
let file = &fx.files[&id];
assert_eq!(file.rte.len(), 5);
assert_eq!(file.rte[1].seg.len(), file.trk[0].trkseg[1].len());
}
#[test]
fn test_cut_tracks_pasted_on_a_route_go_first_among_the_tracks() {
let (mut fx, id) = fixture();
let moved = file(&fx, id).trk[1].id;
fx.clipboard = Clipboard::new(
&Selection::Track {
file_id: id,
trk_ids: [moved].into(),
},
&fx.files,
&fx.order.0,
true,
);
fx.selection = routes(&fx, id, &[0]);
Paste.apply(&mut fx.state()).unwrap();
let file = file(&fx, id);
assert_eq!(file.trk.len(), 2);
assert_eq!(file.trk[0].id, moved);
assert_eq!(file.rte.len(), 2);
}
#[test]
fn test_move_routes() {
let (mut fx, id) = fixture();
let ids: Vec<RouteId> = file(&fx, id).rte.iter().map(|r| r.id).collect();
Move {
what: routes(&fx, id, &[1]),
to: MoveTarget::Routes {
file_id: id,
index: 0,
},
}
.apply(&mut fx.state())
.unwrap();
let order: Vec<RouteId> = file(&fx, id).rte.iter().map(|r| r.id).collect();
assert_eq!(order, [ids[1], ids[0]]);
// to the files: a file of its own
Move {
what: routes(&fx, id, &[0]),
to: MoveTarget::Files { index: 1 },
}
.apply(&mut fx.state())
.unwrap();
assert_eq!(file(&fx, id).rte.len(), 1);
let new = &fx.files[&fx.order.0[1]];
assert_eq!(new.rte[0].id, ids[1]);
// a segment moved among the routes becomes one
let track = &file(&fx, id).trk[0];
let segment = track.trkseg[0].id;
Move {
what: Selection::TrackSegment {
file_id: id,
trk_id: track.id,
trkseg_ids: [segment].into(),
},
to: MoveTarget::Routes {
file_id: id,
index: 1,
},
}
.apply(&mut fx.state())
.unwrap();
let file = file(&fx, id);
assert_eq!(file.rte[1].seg.id, segment);
assert_eq!(file.trk[0].trkseg.len(), 1);
}
#[test]
fn test_routes_and_tracks_moved_among_each_other_keep_their_kind() {
let (mut fx, id) = fixture();
let route = file(&fx, id).rte[0].id;
let track = file(&fx, id).trk[1].id;
// a route among the tracks: after the other routes, which come first
Move {
what: routes(&fx, id, &[0]),
to: MoveTarget::Tracks {
file_id: id,
index: 1,
},
}
.apply(&mut fx.state())
.unwrap();
let moved = file(&fx, id);
assert_eq!((moved.rte.len(), moved.trk.len()), (2, 2));
assert_eq!(moved.rte[1].id, route);
assert_eq!(
fx.selection,
Selection::Route {
file_id: id,
rte_ids: [route].into()
}
);
// a track among the routes: before the other tracks, which come after the routes
Move {
what: Selection::Track {
file_id: id,
trk_ids: [track].into(),
},
to: MoveTarget::Routes {
file_id: id,
index: 0,
},
}
.apply(&mut fx.state())
.unwrap();
let moved = file(&fx, id);
assert_eq!((moved.rte.len(), moved.trk.len()), (2, 2));
assert_eq!(moved.trk[0].id, track);
// and copied routes pasted on a track stay routes
fx.clipboard = Clipboard::new(&routes(&fx, id, &[0]), &fx.files, &fx.order.0, false);
fx.selection = Selection::Track {
file_id: id,
trk_ids: [track].into(),
};
Paste.apply(&mut fx.state()).unwrap();
let pasted = file(&fx, id);
assert_eq!((pasted.rte.len(), pasted.trk.len()), (3, 2));
}
#[test]
fn test_merge_routes() {
let (mut fx, id) = fixture();
let before = lengths(file(&fx, id));
let first = file(&fx, id).rte[0].id;
fx.selection = routes(&fx, id, &[0, 1]);
let merge = |type_| Merge {
type_,
remove_gaps: false,
keep_originals: false,
};
assert_eq!(
merge(MergeType::Group).apply(&mut fx.state()),
Err(CommandError::NothingToDo)
);
merge(MergeType::Connect).apply(&mut fx.state()).unwrap();
let file = file(&fx, id);
assert_eq!(file.rte.len(), 1);
assert_eq!(file.rte[0].id, first);
assert_eq!(file.rte[0].seg.len(), before[0] + before[1]);
assert_eq!(file.trk.len(), 2);
}
#[test]
fn test_merge_files_with_routes() {
let (mut fx, id) = fixture();
let other = File {
rte: file(&fx, id).rte.iter().map(crate::copy_route).collect(),
..File::default()
};
let other_id = other.id;
fx.files.insert(other_id, Rc::new(other));
fx.order.0.push(other_id);
let total: usize = lengths(file(&fx, id)).iter().sum::<usize>()
+ lengths(&fx.files[&other_id]).iter().sum::<usize>();
let mut grouped = Fixture {
files: fx.files.clone(),
order: fx.order.clone(),
selection: Selection::File {
file_ids: [id, other_id].into(),
},
..Default::default()
};
Merge {
type_: MergeType::Group,
remove_gaps: false,
keep_originals: false,
}
.apply(&mut grouped.state())
.unwrap();
let merged = &grouped.files[&id];
assert_eq!((merged.rte.len(), merged.trk.len()), (4, 2));
assert!(!grouped.files.contains_key(&other_id));
// connected in the first route, which comes first
fx.selection = Selection::File {
file_ids: [id, other_id].into(),
};
Merge {
type_: MergeType::Connect,
remove_gaps: false,
keep_originals: false,
}
.apply(&mut fx.state())
.unwrap();
let merged = &fx.files[&id];
assert_eq!((merged.rte.len(), merged.trk.len()), (1, 0));
assert_eq!(merged.rte[0].seg.len(), total);
}
#[test]
fn test_extract_a_file_with_routes() {
let (mut fx, id) = fixture();
let names: Vec<_> = file(&fx, id)
.rte
.iter()
.map(|r| r.info.name.clone().unwrap())
.collect();
fx.selection = whole(id);
Extract {
keep_originals: false,
}
.apply(&mut fx.state())
.unwrap();
// one file per route and one per track
assert_eq!(fx.order.0.len(), 4);
let files: Vec<&File> = fx.order.0.iter().map(|id| &*fx.files[id]).collect();
for (file, name) in files.iter().zip(&names) {
assert_eq!((file.rte.len(), file.trk.len()), (1, 0));
assert_eq!(&file.info.name, name);
}
for file in &files[2..] {
assert_eq!((file.rte.len(), file.trk.len()), (0, 2));
}
// a route has nothing to extract
let (mut fx, id) = fixture();
fx.selection = routes(&fx, id, &[0]);
assert_eq!(
Extract {
keep_originals: false
}
.apply(&mut fx.state()),
Err(CommandError::NothingToDo)
);
}
#[test]
fn test_split_a_route() {
let (mut fx, id) = fixture();
let before = file(&fx, id).rte[1].clone();
fx.selection = routes(&fx, id, &[1]);
Split {
at: 10,
split_type: SplitType::Routes,
keep_originals: false,
}
.apply(&mut fx.state())
.unwrap();
// two routes, which share the trackpoint
let split = file(&fx, id);
assert_eq!(split.rte.len(), 3);
assert_eq!(split.rte[1].id, before.id);
assert_ne!(split.rte[2].id, before.id);
assert_eq!(split.rte[1].seg.len(), 11);
assert_eq!(split.rte[2].seg.len(), before.seg.len() - 10);
assert_eq!(split.rte[2].info, before.info);
// a route is not a track nor a segment, and the other way around
let (mut fx, id) = fixture();
fx.selection = whole(id);
let segment_start = lengths(file(&fx, id))[..2].iter().sum::<usize>() as u32;
for (at, split_type) in [
(10, SplitType::Tracks),
(10, SplitType::Segments),
(segment_start + 10, SplitType::Routes),
] {
assert_eq!(
Split {
at,
split_type,
keep_originals: false,
}
.apply(&mut fx.state()),
Err(CommandError::NothingToDo)
);
}
assert_eq!(file(&fx, id).rte.len(), 2);
// into two files: the following routes and the tracks go to the second one
let (mut fx, id) = fixture();
let first_len = file(&fx, id).rte[0].seg.len();
fx.selection = whole(id);
Split {
at: 5,
split_type: SplitType::Files,
keep_originals: false,
}
.apply(&mut fx.state())
.unwrap();
assert_eq!(fx.order.0.len(), 2);
let (a, b) = (&fx.files[&fx.order.0[0]], &fx.files[&fx.order.0[1]]);
assert_eq!(a.id, id);
assert_eq!((a.rte.len(), a.trk.len()), (1, 0));
assert_eq!(a.rte[0].seg.len(), 6);
assert_eq!((b.rte.len(), b.trk.len()), (2, 2));
assert_eq!(b.rte[0].seg.len(), first_len - 5);
}
#[test]
fn test_crop_removes_the_routes_left_without_trackpoints() {
let (mut fx, id) = fixture();
let len = lengths(file(&fx, id));
fx.selection = whole(id);
// the second route only
let start = len[0] as u32;
Crop {
start,
end: start + len[1] as u32 - 1,
}
.apply(&mut fx.state())
.unwrap();
let file = file(&fx, id);
assert_eq!((file.rte.len(), file.trk.len()), (1, 0));
assert_eq!(file.rte[0].seg.len(), len[1]);
}
#[test]
fn test_clean_removes_the_routes_left_without_trackpoints() {
let (mut fx, id) = fixture();
fx.selection = routes(&fx, id, &[0]);
let everywhere = LngLatBounds {
sw: LngLat {
lng: -180.0,
lat: -90.0,
},
ne: LngLat {
lng: 180.0,
lat: 90.0,
},
};
Clean {
bounds: everywhere,
type_: CleanType::Inside,
trkpt: true,
wpt: false,
}
.apply(&mut fx.state())
.unwrap();
assert_eq!(file(&fx, id).rte.len(), 1);
assert_eq!(fx.selection, whole(id));
}
#[test]
fn test_reverse_a_route_and_the_routes_of_a_file() {
let (mut fx, id) = fixture();
let before = file(&fx, id).clone();
fx.selection = routes(&fx, id, &[0]);
Reverse.apply(&mut fx.state()).unwrap();
let route = &file(&fx, id).rte[0].seg;
let original = &before.rte[0].seg;
assert_eq!(
route[0].coordinates.lat,
original[original.len() - 1].coordinates.lat
);
assert_eq!(file(&fx, id).rte[1].seg.rev_id, before.rte[1].seg.rev_id);
let (mut fx, id) = fixture();
let before = file(&fx, id).clone();
fx.selection = whole(id);
Reverse.apply(&mut fx.state()).unwrap();
let file = file(&fx, id);
// the routes are in the opposite order, and still before the tracks
assert_eq!(file.rte[0].id, before.rte[1].id);
assert_eq!(file.trk[0].id, before.trk[1].id);
}
#[test]
fn test_convert_routes_to_tracks_and_back() {
let (mut fx, id) = fixture();
let before = file(&fx, id).clone();
// a route becomes a track, first among the tracks, with its segment and its information
fx.selection = routes(&fx, id, &[1]);
Convert.apply(&mut fx.state()).unwrap();
let converted = file(&fx, id);
assert_eq!((converted.rte.len(), converted.trk.len()), (1, 3));
assert_eq!(converted.trk[0].info, before.rte[1].info);
assert_eq!(converted.trk[0].trkseg.len(), 1);
assert_eq!(converted.trk[0].trkseg[0].id, before.rte[1].seg.id);
assert_eq!(
fx.selection,
Selection::Track {
file_id: id,
trk_ids: [converted.trk[0].id].into()
}
);
// and back: a route again, after the other routes
Convert.apply(&mut fx.state()).unwrap();
let back = file(&fx, id);
assert_eq!((back.rte.len(), back.trk.len()), (2, 2));
assert_eq!(back.rte[1].seg.id, before.rte[1].seg.id);
assert_eq!(back.rte[1].info, before.rte[1].info);
assert!(
matches!(&fx.selection, Selection::Route { rte_ids, .. } if rte_ids.contains(&back.rte[1].id))
);
}
#[test]
fn test_convert_a_track_to_one_route_per_segment() {
let (mut fx, id) = fixture();
let track = file(&fx, id).trk[0].clone();
assert_eq!(track.trkseg.len(), 2);
fx.selection = Selection::Track {
file_id: id,
trk_ids: [track.id].into(),
};
Convert.apply(&mut fx.state()).unwrap();
let converted = file(&fx, id);
assert_eq!((converted.rte.len(), converted.trk.len()), (4, 1));
for (route, segment) in converted.rte[2..].iter().zip(&track.trkseg) {
assert_eq!(route.seg.id, segment.id);
assert_eq!(route.info, track.info);
}
assert!(matches!(&fx.selection, Selection::Route { rte_ids, .. } if rte_ids.len() == 2));
}
#[test]
fn test_convert_segments_to_routes_keeps_their_track() {
let (mut fx, id) = fixture();
let track = file(&fx, id).trk[1].clone();
fx.selection = Selection::TrackSegment {
file_id: id,
trk_id: track.id,
trkseg_ids: [track.trkseg[0].id, track.trkseg[1].id].into(),
};
Convert.apply(&mut fx.state()).unwrap();
let converted = file(&fx, id);
assert_eq!((converted.rte.len(), converted.trk.len()), (4, 2));
assert!(converted.trk[1].trkseg.is_empty());
assert_eq!(converted.rte[2].seg.id, track.trkseg[0].id);
// nothing to convert otherwise
for selection in [
whole(id),
Selection::Empty,
Selection::Waypoints { file_id: id },
] {
fx.selection = selection;
assert_eq!(
Convert.apply(&mut fx.state()),
Err(CommandError::NothingToDo)
);
}
}
#[test]
fn test_convert_a_file_with_routes_or_tracks_only() {
let (mut fx, id) = fixture();
let before = file(&fx, id).clone();
let only_tracks = File {
trk: before.trk.clone(),
..File::default()
};
let only_tracks_id = only_tracks.id;
fx.files.insert(only_tracks_id, Rc::new(only_tracks));
fx.order.0.push(only_tracks_id);
// the file with both stays as it is, the other gets one route per segment
fx.selection = Selection::File {
file_ids: [id, only_tracks_id].into(),
};
Convert.apply(&mut fx.state()).unwrap();
assert_eq!(lengths(file(&fx, id)), lengths(&before));
let routes = file(&fx, only_tracks_id);
assert_eq!((routes.rte.len(), routes.trk.len()), (4, 0));
assert_eq!(routes.rte[0].seg.id, before.trk[0].trkseg[0].id);
assert_eq!(routes.rte[0].info, before.trk[0].info);
// and back: one track per route
fx.selection = Selection::File {
file_ids: [only_tracks_id].into(),
};
Convert.apply(&mut fx.state()).unwrap();
let tracks = file(&fx, only_tracks_id);
assert_eq!((tracks.rte.len(), tracks.trk.len()), (0, 4));
assert_eq!(tracks.trk[0].trkseg[0].id, before.trk[0].trkseg[0].id);
// nothing to do for a file with both
fx.selection = whole(id);
assert_eq!(
Convert.apply(&mut fx.state()),
Err(CommandError::NothingToDo)
);
}
@@ -24,7 +24,7 @@ impl Apply for ChangeLoopStart {
let index = self.index as usize;
let target = find_target(state, index, index)?.ok_or(CommandError::NothingToDo)?;
let mut file = (*state.files[&target.location.file_id]).clone();
let segment = &mut file.trk[target.location.trk].trkseg[target.location.seg];
let segment = file.segment_mut(target.location.index);
let (start, len) = (target.start, segment.len());
if start == 0 || start >= len {
// it already is the start, or it is not a trackpoint
+32 -9
View File
@@ -13,8 +13,8 @@ pub enum CleanType {
}
/// Removes the trackpoints and/or waypoints of the selection lying inside (or outside) a
/// rectangle. A selected file cleans its tracks and its waypoints, a selected track or segment
/// only trackpoints, selected waypoints only waypoints.
/// rectangle. A selected file cleans its routes, its tracks and its waypoints, a selected route,
/// track or segment only trackpoints, selected waypoints only waypoints.
#[derive(Debug)]
pub struct Clean {
pub bounds: LngLatBounds,
@@ -85,6 +85,11 @@ struct Cleaner<'a> {
impl Editor for Cleaner<'_> {
fn file(&mut self, file: &mut File) -> Edit {
let mut edit = Edit::Unchanged;
for rte in &mut file.rte {
if self.route(rte) == Edit::Changed {
edit = Edit::Changed;
}
}
for trk in &mut file.trk {
if self.track(trk) == Edit::Changed {
edit = Edit::Changed;
@@ -124,24 +129,21 @@ impl Editor for Cleaner<'_> {
}
}
/// Removes what the clean left empty: the emptied segments and the tracks that lost all their
/// segments. A file is never removed, even if nothing is left in it: it stays, empty. The
/// Removes what the clean left empty: the emptied segments, the routes of these segments and the
/// tracks that lost all their segments. A file is never removed, even if nothing is left in it: it stays, empty. The
/// selection falls back to the closest element still existing.
fn prune(state: &mut State, emptied: &HashSet<TrackSegmentId>, wpt_files: &HashSet<FileId>) {
let ids: Vec<FileId> = state
.files
.iter()
.filter(|(id, file)| {
wpt_files.contains(id)
|| file
.trk
.iter()
.any(|trk| trk.trkseg.iter().any(|seg| emptied.contains(&seg.id)))
wpt_files.contains(id) || file.segments().any(|seg| emptied.contains(&seg.id))
})
.map(|(id, _)| *id)
.collect();
for id in ids {
let mut file = (*state.files[&id]).clone();
file.rte.retain(|rte| !emptied.contains(&rte.seg.id));
file.trk.retain_mut(|trk| {
let len = trk.trkseg.len();
trk.trkseg.retain(|seg| !emptied.contains(&seg.id));
@@ -167,6 +169,27 @@ fn fixed_selection(state: &State) -> Selection {
Selection::File { file_ids }
}
}
Selection::Route { file_id, rte_ids } => {
let Some(file) = files.get(file_id) else {
return Selection::Empty;
};
let rte_ids: HashSet<_> = file
.rte
.iter()
.map(|rte| rte.id)
.filter(|id| rte_ids.contains(id))
.collect();
if rte_ids.is_empty() {
Selection::File {
file_ids: [*file_id].into(),
}
} else {
Selection::Route {
file_id: *file_id,
rte_ids,
}
}
}
Selection::Track { file_id, trk_ids } => {
let Some(file) = files.get(file_id) else {
return Selection::Empty;
+9 -8
View File
@@ -11,9 +11,9 @@ use crate::{
/// Keeps the trackpoints `start` to `end` (both included) of the selection, which are counted
/// over its segments one after the other (as in the `RoutingBuffer`), and removes the others.
///
/// The segments that have no trackpoint left are removed. So are the tracks that have no segment
/// left, if tracks or files are selected (not if only segments are), and the files that have no
/// trackpoint left, if files are selected. A range that goes past the last trackpoint stops at it.
/// The segments that have no trackpoint left are removed, and so are the routes of these segments.
/// So are the tracks that have no segment left, if tracks or files are selected (not if only
/// segments are), and the files that have no trackpoint left, if files are selected. A range that goes past the last trackpoint stops at it.
#[derive(Debug)]
pub struct Crop {
pub start: u32,
@@ -43,8 +43,7 @@ impl Apply for Crop {
.segment_locations(state.files, &state.order.0)
.into_iter()
.map(|location| {
let len =
state.files[&location.file_id].trk[location.trk].trkseg[location.seg].len();
let len = location.segment(state.files).len();
let keep = (len > 0 && start < offset + len && end >= offset)
.then(|| (start.saturating_sub(offset), (end - offset).min(len - 1)));
offset += len;
@@ -73,14 +72,14 @@ impl Apply for Crop {
let mut removed: HashSet<TrackSegmentId> = HashSet::new();
let mut had_points: HashSet<FileId> = HashSet::new();
for plan in &plans {
let SegmentLocation { file_id, trk, seg } = plan.location;
let SegmentLocation { file_id, index } = plan.location;
let file = files
.entry(file_id)
.or_insert_with(|| (*state.files[&file_id]).clone());
if plan.len > 0 {
had_points.insert(file_id);
}
let segment = &mut file.trk[trk].trkseg[seg];
let segment = file.segment_mut(index);
match plan.keep {
None => {
removed.insert(segment.id);
@@ -91,6 +90,8 @@ impl Apply for Crop {
let mut gone: Vec<FileId> = vec![];
for (file_id, file) in files.iter_mut() {
// a route has its segment, or is not there
file.rte.retain(|route| !removed.contains(&route.seg.id));
// the tracks that lost segments
let touched: Vec<bool> = file
.trk
@@ -108,7 +109,7 @@ impl Apply for Crop {
file.trk
.retain(|track| !(touched.next().unwrap() && track.trkseg.is_empty()));
}
if prune_files && file.trk.is_empty() && had_points.contains(file_id) {
if prune_files && file.has_no_path() && had_points.contains(file_id) {
gone.push(*file_id);
}
}
+1 -3
View File
@@ -14,9 +14,7 @@ impl Apply for Elevation<'_> {
.segment_locations(state.files, &state.order.0);
let count: usize = locations
.iter()
.map(|location| {
state.files[&location.file_id].trk[location.trk].trkseg[location.seg].len()
})
.map(|location| location.segment(state.files).len())
.sum();
if count == 0 {
return Err(CommandError::NothingToDo);
+62 -56
View File
@@ -1,22 +1,23 @@
use std::{collections::HashSet, rc::Rc};
use crate::{
Apply, CommandError, File, FileId, LngLat, LngLatBounds, Selection, State, Track, TrackId,
TrackSegment, Waypoint, Waypoints, copy_file, copy_waypoint, crossarc_lnglat, distance,
Apply, CommandError, File, FileId, LngLat, LngLatBounds, Route, Selection, State, Track,
TrackId, TrackSegment, Waypoint, Waypoints, copy_file, copy_waypoint, crossarc_lnglat,
distance,
};
/// Splits the selected elements that are made of several segments into elements that have one
/// each:
///
/// - files: a file with several tracks becomes one file per track, in which the track is split
/// into one track per segment. A file with a single track becomes one file per segment. The new
/// files are in the place of the file, which goes, and are selected. They are named after the
/// track, or the file, and numbered. The waypoints go to the files that have the segments
/// - files: a file with several routes and tracks becomes one file per route and per track, in
/// which the track is split into one track per segment. A file with a single track becomes one
/// file per segment. The new files are in the place of the file, which goes, and are selected.
/// They are named after the route or track, or the file, and numbered. The waypoints go to the files that have the segments
/// closest to them (to the first file if the file has no trackpoint).
/// - tracks: each selected track is replaced by one track per segment, which are selected. They
/// are named after the track and numbered.
///
/// What has a single segment is left as it is. There is nothing to do if nothing was changed.
/// What has a single segment is left as it is, which routes always are. There is nothing to do if nothing was changed.
///
/// With `keep_originals`, the files that are extracted stay where they are, and the new files come
/// right after each of them, and only the new files are selected. Extracting
@@ -46,8 +47,7 @@ fn extract_files(
let mut extracted: Vec<(FileId, Vec<File>)> = vec![];
for id in state.order.0.iter().filter(|id| file_ids.contains(id)) {
let file = &state.files[id];
let segments: usize = file.trk.iter().map(|track| track.trkseg.len()).sum();
if segments > 1 {
if file.segments().count() > 1 {
let mut files = extract_file(file);
if keep_originals {
// the tracks and segments are the ones of the file, which stays: they need new ids
@@ -95,60 +95,65 @@ fn extract_files(
/// The files that a file with several segments becomes.
fn extract_file(file: &File) -> Vec<File> {
// for each new file: its tracks and its name, and the segments that go with it
// for each new file: its routes, its tracks and its name, and the segments that go with it
struct Part {
routes: Vec<Route>,
tracks: Vec<Track>,
name: String,
}
let (parts, owner): (Vec<Part>, Vec<usize>) = if file.trk.len() > 1 {
// one file per track, the segments of a track follow it
let parts = file
.trk
.iter()
.enumerate()
.map(|(i, track)| Part {
tracks: split_track(track),
name: track
.info
.name
.clone()
.unwrap_or_else(|| format!("{} ({})", file.info.name, i + 1)),
})
.collect();
let owner = file
.trk
.iter()
.enumerate()
.flat_map(|(i, track)| track.trkseg.iter().map(move |_| i))
.collect();
(parts, owner)
} else {
// one file per segment
let track = &file.trk[0];
let base = track
.info
.name
.clone()
.unwrap_or_else(|| file.info.name.clone());
let parts = track
.trkseg
.iter()
.enumerate()
.map(|(i, segment)| Part {
tracks: vec![Track {
id: if i == 0 { track.id } else { Default::default() },
trkseg: vec![segment.clone()],
..track.clone()
}],
name: format!("{base} ({})", i + 1),
})
.collect();
let owner = (0..track.trkseg.len()).collect();
(parts, owner)
let name = |name: &Option<String>, i: usize| {
name.clone()
.unwrap_or_else(|| format!("{} ({})", file.info.name, i + 1))
};
let (parts, owner): (Vec<Part>, Vec<usize>) =
if file.rte.len() + file.trk.len() > 1 {
// one file per route and per track, the segments of a track follow it
let routes = file.rte.iter().enumerate().map(|(i, route)| Part {
routes: vec![route.clone()],
tracks: vec![],
name: name(&route.info.name, i),
});
let tracks = file.trk.iter().enumerate().map(|(i, track)| Part {
routes: vec![],
tracks: split_track(track),
name: name(&track.info.name, file.rte.len() + i),
});
let parts = routes.chain(tracks).collect();
let owner =
(0..file.rte.len())
.chain(file.trk.iter().enumerate().flat_map(|(i, track)| {
track.trkseg.iter().map(move |_| file.rte.len() + i)
}))
.collect();
(parts, owner)
} else {
// one file per segment
let track = &file.trk[0];
let base = track
.info
.name
.clone()
.unwrap_or_else(|| file.info.name.clone());
let parts = track
.trkseg
.iter()
.enumerate()
.map(|(i, segment)| Part {
routes: vec![],
tracks: vec![Track {
id: if i == 0 { track.id } else { Default::default() },
trkseg: vec![segment.clone()],
..track.clone()
}],
name: format!("{base} ({})", i + 1),
})
.collect();
let owner = (0..track.trkseg.len()).collect();
(parts, owner)
};
// the waypoints go with the segments that are the closest to them
let segments: Vec<&TrackSegment> = file.trk.iter().flat_map(|track| &track.trkseg).collect();
let segments: Vec<&TrackSegment> = file.segments().collect();
let mut waypoints: Vec<Vec<Waypoint>> = vec![vec![]; parts.len()];
for waypoint in file.wpt.iter() {
let mut owners: Vec<usize> = closest_segments(&segments, waypoint.coordinates)
@@ -176,6 +181,7 @@ fn extract_file(file: &File) -> Vec<File> {
.map(|(part, waypoints)| {
let mut extracted = File {
id: Default::default(),
rte: part.routes,
trk: part.tracks,
wpt: Waypoints::default(),
..file.clone()
@@ -39,7 +39,7 @@ impl Apply for InsertAnchor {
.selection
.segment_locations(state.files, &state.order.0)
{
let segment = &state.files[&location.file_id].trk[location.trk].trkseg[location.seg];
let segment = location.segment(state.files);
let mut points = segment.iter();
let Some(mut previous) = points.next() else {
continue;
@@ -81,7 +81,7 @@ impl Apply for InsertAnchor {
let closest = closest.ok_or(CommandError::NothingToDo)?;
let mut file = (*state.files[&closest.location.file_id]).clone();
let segment = &mut file.trk[closest.location.trk].trkseg[closest.location.seg];
let segment = file.segment_mut(closest.location.index);
let before = closest.before;
// on a trackpoint already, if it is closer than `SNAP_DISTANCE`
let snapped = if segment.len() == 1 {
+63 -15
View File
@@ -1,8 +1,9 @@
use std::{collections::HashSet, rc::Rc};
use crate::{
Apply, CommandError, FileId, GlobalStatistics, Selection, State, Statistics, Track, TrackId,
TrackSegment, TrackSegmentId, Waypoint, Waypoints, copy_track, replace_trackpoints,
Apply, CommandError, FileId, GlobalStatistics, Route, RouteId, Selection, State, Statistics,
Track, TrackId, TrackSegment, TrackSegmentId, Waypoint, Waypoints, copy_route, copy_track,
replace_trackpoints,
};
#[derive(Debug, Clone, Copy, PartialEq, Eq, serde::Deserialize)]
@@ -16,13 +17,16 @@ pub enum MergeType {
/// Merges the selected elements into the first of them:
///
/// - files: the other files are deleted. The tracks of all of them are in the first file, as
/// they are with `Group`, or all connected in the first track with `Connect`. The waypoints of
/// - files: the other files are deleted. The routes and tracks of all of them are in the first
/// file, as they are with `Group`, or all connected in the first of them (a route or a track,
/// the routes of a file coming before its tracks) with `Connect`. The waypoints of
/// all of them are in the first file too, except the ones that are the same. With
/// `keep_originals`, no file is deleted nor changed: the result is a new file (with the info of
/// the first one and new ids), which comes right after the last of the files and is selected.
/// - tracks of a file: the first track gets their segments, as they are with `Group`, or all
/// connected in one segment with `Connect`. The other tracks are removed.
/// - routes of a file: the first route gets their trackpoints, all connected with `Connect`, and
/// the other routes are removed. A route has a single segment, so there is nothing to `Group`.
/// - segments of a track: they are all connected in a segment at the position of the first one
/// with `Connect`. With `Group` they are put one after the other, at that position.
///
@@ -44,6 +48,7 @@ impl Apply for Merge {
let selection = state.selection.clone();
match &selection {
Selection::File { file_ids } => self.merge_files(state, file_ids),
Selection::Route { file_id, rte_ids } => self.merge_routes(state, *file_id, rte_ids),
Selection::Track { file_id, trk_ids } => self.merge_tracks(state, *file_id, trk_ids),
Selection::TrackSegment {
file_id,
@@ -77,21 +82,26 @@ impl Merge {
return Err(CommandError::NothingToDo);
};
let mut routes: Vec<Route> = vec![];
let mut tracks: Vec<Track> = vec![];
let mut segments: Vec<TrackSegment> = vec![];
let mut waypoints: Vec<Waypoint> = vec![];
for id in &ordered {
let file = &state.files[id];
routes.extend(file.rte.iter().cloned());
tracks.extend(file.trk.iter().cloned());
segments.extend(file.segments().cloned());
for waypoint in file.wpt.iter() {
if !waypoints.iter().any(|kept| kept.same_content(waypoint)) {
waypoints.push(waypoint.clone());
}
}
}
let segments: Vec<TrackSegment> = tracks
.iter()
.flat_map(|track| track.trkseg.iter().cloned())
.collect();
// the first route or track, which holds the connected segment
let first_is_route = ordered.iter().find_map(|id| {
let file = &state.files[id];
(!file.has_no_path()).then(|| !file.rte.is_empty())
});
match self.type_ {
MergeType::Group if ordered.len() == 1 => return Err(CommandError::NothingToDo),
@@ -99,24 +109,35 @@ impl Merge {
return Err(CommandError::NothingToDo);
}
MergeType::Connect if !segments.is_empty() => {
// a single track, which is the first one, with the connected segment
// a single route or track, which is the first one, with the connected segment
let merged = connect(&segments, segments[0].id, self.remove_gaps);
let first = Track {
trkseg: vec![merged],
..tracks[0].clone()
};
tracks = vec![first];
if first_is_route == Some(true) {
let first = Route {
seg: merged,
..routes[0].clone()
};
(routes, tracks) = (vec![first], vec![]);
} else {
let first = Track {
trkseg: vec![merged],
..tracks[0].clone()
};
(routes, tracks) = (vec![], vec![first]);
}
}
_ => {}
}
let mut target = (*state.files[&target_id]).clone();
target.rte = routes;
target.trk = tracks;
target.wpt = Waypoints::default();
target.wpt.insert_at(0, waypoints);
if self.keep_originals {
// a new file, whose tracks and segments do not share their ids with the originals
// a new file, whose routes, tracks and segments do not share their ids with the
// originals
target.id = Default::default();
target.rte = target.rte.iter().map(copy_route).collect();
target.trk = target.trk.iter().map(copy_track).collect();
let at = state
.order
@@ -139,6 +160,33 @@ impl Merge {
Ok(())
}
fn merge_routes(
&self,
state: &mut State,
file_id: FileId,
rte_ids: &HashSet<RouteId>,
) -> Result<(), CommandError> {
let file = state.files.get(&file_id).ok_or(CommandError::NothingToDo)?;
let selected: Vec<usize> = (0..file.rte.len())
.filter(|i| rte_ids.contains(&file.rte[*i].id))
.collect();
if selected.len() < 2 || self.type_ == MergeType::Group {
return Err(CommandError::NothingToDo);
}
let first = selected[0];
let segments: Vec<TrackSegment> =
selected.iter().map(|i| file.rte[*i].seg.clone()).collect();
let mut file = (**file).clone();
file.rte[first].seg = connect(&segments, segments[0].id, self.remove_gaps);
let mut index = 0;
file.rte.retain(|_| {
index += 1;
!selected[1..].contains(&(index - 1))
});
state.files.insert(file_id, Rc::new(file));
Ok(())
}
fn merge_tracks(
&self,
state: &mut State,
@@ -58,7 +58,8 @@ impl Apply for NewWaypoint<'_> {
.filter(|id| file_ids.contains(id))
.copied()
.collect(),
Selection::Track { file_id, .. }
Selection::Route { file_id, .. }
| Selection::Track { file_id, .. }
| Selection::TrackSegment { file_id, .. }
| Selection::Waypoints { file_id }
| Selection::Waypoint { file_id, .. } => vec![*file_id],
@@ -118,12 +119,14 @@ mod tests {
New {
name: "a",
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
New {
name: "b",
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
@@ -169,6 +172,7 @@ mod tests {
New {
name: "a",
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
@@ -196,6 +200,7 @@ mod tests {
New {
name,
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
@@ -234,6 +239,7 @@ mod tests {
New {
name: "a",
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
@@ -250,6 +256,7 @@ mod tests {
New {
name: "a",
trackpoint: None,
route: false,
}
.apply(&mut fx.state())
.unwrap();
+1 -1
View File
@@ -23,7 +23,7 @@ impl Apply for Reduce {
let mut changed = vec![];
let mut kept_points = vec![];
for location in &locations {
let segment = &state.files[&location.file_id].trk[location.trk].trkseg[location.seg];
let segment = location.segment(state.files);
let kept = reduce_indices(segment, self.tolerance);
if kept.len() == segment.len() {
continue;
+21 -9
View File
@@ -1,14 +1,15 @@
use std::rc::Rc;
use crate::{
Apply, CommandError, File, FileId, Selection, State, TrackId, TrackSegmentId, end_time,
reverse_segment, reverse_segments, reverse_tracks, start_time, track_end_time,
track_start_time,
Apply, CommandError, File, FileId, RouteId, Selection, State, TrackId, TrackSegmentId,
end_time, reverse_routes, reverse_segment, reverse_segments, reverse_tracks, start_time,
track_end_time, track_start_time,
};
/// Reverses the direction of the selected elements: the whole of each selected file (its tracks,
/// the segments of its tracks and their trackpoints are in the opposite order), of each selected
/// track (its segments and their trackpoints), or each selected segment on its own (its
/// Reverses the direction of the selected elements: the whole of each selected file (its routes,
/// its tracks, the segments of its tracks and their trackpoints are in the opposite order; the
/// routes and the tracks are two wholes, as the routes stay before the tracks), of each selected
/// track (its segments and their trackpoints), or each selected route or segment on its own (its
/// trackpoints). The timestamps are mirrored, see [`reverse_segment`].
#[derive(Debug)]
pub struct Reverse;
@@ -20,9 +21,7 @@ impl Apply for Reverse {
.segment_locations(state.files, &state.order.0);
let points: usize = locations
.iter()
.map(|location| {
state.files[&location.file_id].trk[location.trk].trkseg[location.seg].len()
})
.map(|location| location.segment(state.files).len())
.sum();
if points <= 1 {
return Err(CommandError::NothingToDo);
@@ -40,12 +39,18 @@ impl Apply for Reverse {
Selection::File { file_ids } => {
for file_id in file_ids {
edit(file_id, &|file| {
let end = file.rte.last().and_then(|rte| end_time(&rte.seg));
let start = file.rte.first().and_then(|rte| start_time(&rte.seg));
reverse_routes(&mut file.rte, end, start);
let end = file.trk.last().and_then(track_end_time);
let start = file.trk.first().and_then(track_start_time);
reverse_tracks(&mut file.trk, end, start);
});
}
}
Selection::Route { file_id, rte_ids } => edit(file_id, &|file| {
reverse_selected_routes(file, rte_ids);
}),
Selection::Track { file_id, trk_ids } => edit(file_id, &|file| {
reverse_selected_tracks(file, trk_ids);
}),
@@ -66,6 +71,13 @@ impl Apply for Reverse {
}
}
fn reverse_selected_routes(file: &mut File, ids: &std::collections::HashSet<RouteId>) {
for route in file.rte.iter_mut().filter(|route| ids.contains(&route.id)) {
let (end, start) = (end_time(&route.seg), start_time(&route.seg));
reverse_segment(&mut route.seg, end, start);
}
}
fn reverse_selected_tracks(file: &mut File, ids: &std::collections::HashSet<TrackId>) {
for track in file.trk.iter_mut().filter(|track| ids.contains(&track.id)) {
let end = track.trkseg.last().and_then(end_time);
@@ -14,7 +14,7 @@ impl Apply for RoundTrip {
.selection
.segment_locations(state.files, &state.order.0)
.into_iter()
.filter(|l| state.files[&l.file_id].trk[l.trk].trkseg[l.seg].len() >= 2)
.filter(|l| l.segment(state.files).len() >= 2)
.collect();
if locations.is_empty() {
return Err(CommandError::NothingToDo);
+17 -14
View File
@@ -1,11 +1,11 @@
use std::rc::Rc;
use crate::{
Apply, Categories, CommandError, FileId, SegmentLocation, Selection, State, Statistics, Track,
TrackSegment, Trackpoint, distance, replace_trackpoints,
Apply, Categories, CommandError, FileId, SegmentIndex, SegmentLocation, Selection, State,
Statistics, Track, TrackSegment, Trackpoint, distance, replace_trackpoints,
};
/// The values of an OSM attribute (surface, highway...) of the new trackpoints of a [`Route`], as
/// The values of an OSM attribute (surface, highway...) of the new trackpoints of a [`Reroute`], as
/// the intervals of consecutive trackpoints that share one: `starts[i]` is the index of the first
/// trackpoint of the interval `i`, and `values[i]` its value, 0 when unknown, else 1 + the index
/// of its name in `names`. No intervals means that the attribute is unknown everywhere.
@@ -64,7 +64,7 @@ impl RouteCategory<'_> {
/// anchors anyway. The ones right before and after the range, between which the route goes, become
/// shown at every zoom level as well.
#[derive(Debug)]
pub struct Route<'a> {
pub struct Reroute<'a> {
pub start: u32,
pub end: u32,
pub lng: &'a [f64],
@@ -89,7 +89,7 @@ pub(super) struct Target {
pub end: usize,
}
impl Apply for Route<'_> {
impl Apply for Reroute<'_> {
fn apply(self, state: &mut State) -> Result<(), CommandError> {
let (start, end) = (self.start as usize, self.end as usize);
let len = self.lng.len();
@@ -152,14 +152,17 @@ impl Apply for Route<'_> {
}
let seg = file.trk[trk].trkseg.len() - 1;
Target {
location: SegmentLocation { file_id, trk, seg },
location: SegmentLocation {
file_id,
index: SegmentIndex::Track { trk, seg },
},
start: 0,
end: 0,
}
}
};
let segment = &mut file.trk[target.location.trk].trkseg[target.location.seg];
let segment = file.segment_mut(target.location.index);
let new_len = points.len();
let (speed, start_time) = speed_and_start_time(segment, &target, &points);
let has_times = !segment.is_empty() && segment[0].time.is_some();
@@ -208,7 +211,7 @@ pub(super) fn find_target(
let segments: Vec<(SegmentLocation, usize)> = locations
.into_iter()
.map(|location| {
let segment = &state.files[&location.file_id].trk[location.trk].trkseg[location.seg];
let segment = location.segment(state.files);
(location, segment.len())
})
.collect();
@@ -382,8 +385,8 @@ mod tests {
lng: &'a [f64],
lat: &'a [f64],
ele: &'a [f64],
) -> Route<'a> {
Route {
) -> Reroute<'a> {
Reroute {
start,
end,
lng,
@@ -459,7 +462,7 @@ mod tests {
route(0, 0, &[1.0], &[], &[]).apply(&mut fx.state()),
route(0, len + 1, &[], &[], &[]).apply(&mut fx.state()),
route(len + 1, len + 1, &[1.0], &[1.0], &[1.0]).apply(&mut fx.state()),
Route {
Reroute {
anchors: &[1],
..route(0, 0, &[1.0], &[1.0], &[1.0])
}
@@ -589,7 +592,7 @@ mod tests {
// 5 new points replacing the 2 in the middle, two of them are anchors
let (lng, lat, ele) = ([1.0; 5], [2.0; 5], [3.0; 5]);
Route {
Reroute {
anchors: &[1, 3],
..route(1, 3, &lng, &lat, &ele)
}
@@ -639,7 +642,7 @@ mod tests {
let names = ["asphalt".to_string(), "gravel".to_string()];
let (lng, lat, ele) = ([1.0; 4], [2.0; 4], [3.0; 4]);
// asphalt, asphalt, unknown, gravel
Route {
Reroute {
surface: RouteCategory {
starts: &[0, 2, 3],
values: &[1, 0, 2],
@@ -683,7 +686,7 @@ mod tests {
},
];
for surface in broken {
let result = Route {
let result = Reroute {
surface,
..route(0, 0, &lng, &lat, &ele)
}
+106 -92
View File
@@ -1,8 +1,8 @@
use std::rc::Rc;
use crate::{
Apply, CommandError, File, Selection, State, Track, TrackSegment, Waypoints, copy_file,
copy_segment, copy_track, copy_waypoint,
Apply, CommandError, File, Route, SegmentIndex, Selection, State, Track, TrackSegment,
Waypoints, copy_file, copy_route, copy_segment, copy_track, copy_waypoint,
};
use super::route::find_target;
@@ -13,24 +13,27 @@ use super::route::find_target;
pub enum SplitType {
/// The file becomes two files, which both have the waypoints (copies, in the second one).
Files,
/// The route becomes two routes.
Routes,
/// The track becomes two tracks.
Tracks,
/// The segment becomes two segments.
/// The segment of a track becomes two segments.
Segments,
}
/// Cuts the file, the track or the segment of the trackpoint `at` of the selection (see
/// Cuts the file, the route, the track or the segment of the trackpoint `at` of the selection (see
/// `RoutingBuffer`) in two, there: the trackpoint is the last one of the first part and the first
/// one of the second part.
/// one of the second part. Cutting a route needs the trackpoint to be in a route, and cutting a
/// track or a segment needs it to be in a track: there is nothing to do otherwise.
///
/// The first part keeps the ids of what is cut, the second one has new ids, as the elements that
/// come after the trackpoint (the following segments of the track, the following tracks of the
/// file), which are moved to it. When the file is cut, the new file comes right after it. The
/// selection does not change.
/// come after the trackpoint (the following segments of the track, the following routes and
/// tracks of the file, the routes coming before the tracks), which are moved to it. When the file
/// is cut, the new file comes right after it. The selection does not change.
///
/// With `keep_originals`, a file that is cut (`SplitType::Files`) is left as it is: its two parts
/// are two new files, with new ids, which come right after it and are selected. Cutting a track
/// or a segment changes the file in place anyway, so the flag has no effect on them.
/// are two new files, with new ids, which come right after it and are selected. Cutting a route, a
/// track or a segment changes the file in place anyway, so the flag has no effect on them.
#[derive(Debug)]
pub struct Split {
pub at: u32,
@@ -43,104 +46,115 @@ impl Apply for Split {
let at = self.at as usize;
let target = find_target(state, at, at)?.ok_or(CommandError::NothingToDo)?;
let location = target.location;
let (file_id, trk, seg, index) =
(location.file_id, location.trk, location.seg, target.start);
let (file_id, index) = (location.file_id, target.start);
let in_route = matches!(location.index, SegmentIndex::Route(_));
let fits = match self.split_type {
SplitType::Files => true,
SplitType::Routes => in_route,
SplitType::Tracks | SplitType::Segments => !in_route,
};
if !fits {
return Err(CommandError::NothingToDo);
}
let mut file = (*state.files[&file_id]).clone();
if index >= file.trk[trk].trkseg[seg].len() {
if index >= file.segment(location.index).len() {
return Err(CommandError::InvalidData(
"trackpoint out of the segment".into(),
));
}
let track = file.trk[trk].clone();
let segment = &track.trkseg[seg];
let (first, second) = (keep_up_to(segment, index), keep_from(segment, index));
// the track is cut: what is before the segment, and its first part, then the second part
// and what is after it
let split_track = || {
let mut first_track = track.clone();
first_track.trkseg.truncate(seg);
first_track.trkseg.push(first.clone());
let mut second_track = Track {
id: Default::default(),
trkseg: vec![second.clone()],
..track.clone()
};
second_track
.trkseg
.extend(track.trkseg[seg + 1..].iter().map(copy_segment));
(first_track, second_track)
// with `SplitType::Files`, what goes to the second file: what comes after the trackpoint
let mut second_file = File {
id: Default::default(),
rte: vec![],
trk: vec![],
wpt: Waypoints::default(),
..file.clone()
};
match self.split_type {
SplitType::Segments => {
file.trk[trk].trkseg.splice(seg..=seg, [first, second]);
}
SplitType::Tracks => {
let (first_track, second_track) = split_track();
file.trk.splice(trk..=trk, [first_track, second_track]);
}
SplitType::Files if self.keep_originals => {
let (first_track, second_track) = split_track();
// the parts are new files: nothing is shared by id with the file, nor between them
let first_file = copy_file(&File {
trk: file.trk[..trk]
.iter()
.cloned()
.chain([first_track])
.collect(),
..file.clone()
});
let mut second_file = File {
match location.index {
SegmentIndex::Route(rte) => {
let route = file.rte[rte].clone();
let first = Route {
seg: keep_up_to(&route.seg, index),
..route.clone()
};
let second = Route {
id: Default::default(),
trk: vec![second_track],
wpt: Waypoints::default(),
..file.clone()
seg: keep_from(&route.seg, index),
..route.clone()
};
second_file
.trk
.extend(file.trk[trk + 1..].iter().map(copy_track));
second_file
.wpt
.insert_at(0, file.wpt.iter().map(copy_waypoint).collect());
let ids = [first_file.id, second_file.id];
let position = state.order.0.iter().position(|id| *id == file_id);
let at = position.map_or(state.order.0.len(), |p| p + 1);
state.order.0.splice(at..at, ids);
state.files.insert(first_file.id, Rc::new(first_file));
state.files.insert(second_file.id, Rc::new(second_file));
*state.selection = Selection::File {
file_ids: ids.into(),
};
return Ok(());
if self.split_type == SplitType::Routes {
file.rte.splice(rte..=rte, [first, second]);
state.files.insert(file_id, Rc::new(file));
return Ok(());
}
// the following routes, and the tracks, which come after the routes
second_file.rte = [second]
.into_iter()
.chain(file.rte[rte + 1..].iter().map(copy_route))
.collect();
second_file.trk = file.trk.iter().map(copy_track).collect();
file.rte.truncate(rte);
file.rte.push(first);
file.trk.clear();
}
SplitType::Files => {
let (first_track, second_track) = split_track();
let mut second_file = File {
SegmentIndex::Track { trk, seg } => {
let track = file.trk[trk].clone();
let segment = &track.trkseg[seg];
let (first, second) = (keep_up_to(segment, index), keep_from(segment, index));
if self.split_type == SplitType::Segments {
file.trk[trk].trkseg.splice(seg..=seg, [first, second]);
state.files.insert(file_id, Rc::new(file));
return Ok(());
}
// the track is cut: what is before the segment, and its first part, then the
// second part and what is after it
let mut first_track = track.clone();
first_track.trkseg.truncate(seg);
first_track.trkseg.push(first);
let mut second_track = Track {
id: Default::default(),
trk: vec![second_track],
wpt: Waypoints::default(),
..file.clone()
trkseg: vec![second],
..track.clone()
};
second_file
.trk
.extend(file.trk[trk + 1..].iter().map(copy_track));
second_file
.wpt
.insert_at(0, file.wpt.iter().map(copy_waypoint).collect());
second_track
.trkseg
.extend(track.trkseg[seg + 1..].iter().map(copy_segment));
if self.split_type == SplitType::Tracks {
file.trk.splice(trk..=trk, [first_track, second_track]);
state.files.insert(file_id, Rc::new(file));
return Ok(());
}
second_file.trk = [second_track]
.into_iter()
.chain(file.trk[trk + 1..].iter().map(copy_track))
.collect();
file.trk.truncate(trk);
file.trk.push(first_track);
let position = state.order.0.iter().position(|id| *id == file_id);
state.order.0.insert(
position.map_or(state.order.0.len(), |p| p + 1),
second_file.id,
);
state.files.insert(second_file.id, Rc::new(second_file));
}
}
state.files.insert(file_id, Rc::new(file));
// the file is cut: both parts have the waypoints
second_file
.wpt
.insert_at(0, file.wpt.iter().map(copy_waypoint).collect());
let position = state.order.0.iter().position(|id| *id == file_id);
let at = position.map_or(state.order.0.len(), |p| p + 1);
if self.keep_originals {
// the parts are new files: nothing is shared by id with the file, nor between them
let first_file = copy_file(&file);
let ids = [first_file.id, second_file.id];
state.order.0.splice(at..at, ids);
state.files.insert(first_file.id, Rc::new(first_file));
state.files.insert(second_file.id, Rc::new(second_file));
*state.selection = Selection::File {
file_ids: ids.into(),
};
} else {
state.order.0.insert(at, second_file.id);
state.files.insert(second_file.id, Rc::new(second_file));
state.files.insert(file_id, Rc::new(file));
}
Ok(())
}
}
+2 -4
View File
@@ -45,10 +45,8 @@ impl Apply for Time {
let locations = state
.selection
.segment_locations(state.files, &state.order.0);
let segments: Vec<&TrackSegment> = locations
.iter()
.map(|l| &state.files[&l.file_id].trk[l.trk].trkseg[l.seg])
.collect();
let segments: Vec<&TrackSegment> =
locations.iter().map(|l| l.segment(state.files)).collect();
if segments.iter().all(|segment| segment.is_empty()) {
return Err(CommandError::NothingToDo);
}
@@ -3,7 +3,7 @@ use std::collections::{HashMap, HashSet};
use crate::{FileId, FileWaypointsRevisionId, Files, TrackSegmentId, TrackSegmentRevisionId};
/// Coordinates buffers, as flat `[lng, lat, lng, lat, ...]` arrays so that they can be handed
/// to the UI without per-point calls: one per track segment (its trackpoints) and one per file
/// to the UI without per-point calls: one per segment of a route or track (its trackpoints) and one per file
/// (its waypoints). A buffer is only rebuilt when the revision of its segment (or the waypoints
/// of its file) changes, so editing a segment does not touch the buffers of the others.
#[derive(Debug, Default)]
@@ -30,7 +30,7 @@ impl CoordinatesCache {
self.waypoints
.insert(file.id, (file.wpt.rev_id, coordinates));
}
for seg in file.trk.iter().flat_map(|trk| &trk.trkseg) {
for seg in file.segments() {
segments.insert(seg.id);
if self
.segments
+56 -25
View File
@@ -1,9 +1,10 @@
use crate::{
File, FileId, FileWaypointsRevisionId, TrackId, TrackSegmentId, TrackSegmentRevisionId,
WaypointId,
File, FileId, FileWaypointsRevisionId, RouteId, TrackId, TrackInfo, TrackSegment,
TrackSegmentId, TrackSegmentRevisionId, WaypointId,
};
/// What the UI needs to display a file: its name and the structure of its tracks and waypoints,
/// What the UI needs to display a file: its name and the structure of its routes, tracks and
/// waypoints,
/// with the ids that reference every element. Coordinates are not part of it, they are read
/// from the buffers of the [`CoordinatesCache`](crate::CoordinatesCache) (the `i`-th waypoint
/// of a file is the `i`-th coordinates pair of its waypoint buffer, same for the trackpoints of
@@ -13,21 +14,47 @@ pub struct FileStructure {
pub id: FileId,
pub name: String,
pub desc: Option<String>,
pub routes: Vec<RouteNode>,
pub tracks: Vec<TrackNode>,
pub waypoints: Vec<WaypointNode>,
/// Changes whenever the waypoints of the file (hence their buffer) change.
pub wpt_rev_id: FileWaypointsRevisionId,
}
/// The name, description and style of a route or a track (the style only when the file defines
/// it).
#[derive(Debug, Clone, PartialEq)]
pub struct TrackNode {
pub id: TrackId,
pub struct PathInfo {
pub name: Option<String>,
pub desc: Option<String>,
/// Style of the track, only when the file defines it.
pub color: Option<String>,
pub opacity: Option<f64>,
pub width: Option<f64>,
}
impl From<&TrackInfo> for PathInfo {
fn from(info: &TrackInfo) -> Self {
Self {
name: info.name.clone(),
desc: info.desc.clone(),
color: info.color.clone(),
opacity: info.opacity,
width: info.width,
}
}
}
#[derive(Debug, Clone, PartialEq)]
pub struct RouteNode {
pub id: RouteId,
pub info: PathInfo,
pub segment: SegmentNode,
}
#[derive(Debug, Clone, PartialEq)]
pub struct TrackNode {
pub id: TrackId,
pub info: PathInfo,
pub segments: Vec<SegmentNode>,
}
@@ -46,31 +73,38 @@ pub struct WaypointNode {
pub sym: Option<String>,
}
impl From<&TrackSegment> for SegmentNode {
fn from(seg: &TrackSegment) -> Self {
Self {
id: seg.id,
rev_id: seg.rev_id,
len: seg.len(),
}
}
}
impl FileStructure {
pub fn new(file: &File) -> Self {
Self {
id: file.id,
name: file.info.name.clone(),
desc: file.info.desc.clone(),
routes: file
.rte
.iter()
.map(|rte| RouteNode {
id: rte.id,
info: (&rte.info).into(),
segment: (&rte.seg).into(),
})
.collect(),
tracks: file
.trk
.iter()
.map(|trk| TrackNode {
id: trk.id,
name: trk.info.name.clone(),
desc: trk.info.desc.clone(),
color: trk.info.color.clone(),
opacity: trk.info.opacity,
width: trk.info.width,
segments: trk
.trkseg
.iter()
.map(|seg| SegmentNode {
id: seg.id,
rev_id: seg.rev_id,
len: seg.len(),
})
.collect(),
info: (&trk.info).into(),
segments: trk.trkseg.iter().map(Into::into).collect(),
})
.collect(),
waypoints: file
@@ -112,11 +146,8 @@ mod tests {
);
assert_eq!(node.tracks.len(), file.trk.len());
for (n, t) in node.tracks.iter().zip(&file.trk) {
assert_eq!((n.id, &n.name, &n.desc), (t.id, &t.info.name, &t.info.desc));
assert_eq!(
(&n.color, n.opacity, n.width),
(&t.info.color, t.info.opacity, t.info.width)
);
assert_eq!(n.id, t.id);
assert_eq!(n.info, PathInfo::from(&t.info));
assert_eq!(n.segments.len(), t.trkseg.len());
for (n, s) in n.segments.iter().zip(&t.trkseg) {
assert_eq!((n.id, n.rev_id, n.len), (s.id, s.rev_id, s.len()));
+1 -1
View File
@@ -39,7 +39,7 @@ impl RoutingBuffer {
.map(|files| selection.segment_locations(files, order))
.unwrap_or_default()
{
let segment = &files.unwrap()[&location.file_id].trk[location.trk].trkseg[location.seg];
let segment = location.segment(files.unwrap());
signature.push((segment.id, segment.rev_id));
if segment.is_empty() {
continue;
+18 -11
View File
@@ -18,15 +18,13 @@ impl StatisticsCache {
let mut stale: HashSet<TrackSegmentRevisionId> = self.map.keys().copied().collect();
if let Some(state) = state {
for file in state.values() {
for trk in file.trk.iter() {
for trkseg in trk.trkseg.iter() {
match self.map.entry(trkseg.rev_id) {
Entry::Occupied(_) => {
stale.remove(&trkseg.rev_id);
}
Entry::Vacant(e) => {
e.insert(Statistics::compute(trkseg));
}
for trkseg in file.segments() {
match self.map.entry(trkseg.rev_id) {
Entry::Occupied(_) => {
stale.remove(&trkseg.rev_id);
}
Entry::Vacant(e) => {
e.insert(Statistics::compute(trkseg));
}
}
}
@@ -72,6 +70,15 @@ impl StatisticsCache {
}
}
}
Selection::Route { file_id, rte_ids } => {
if let Some(file) = state.get(file_id) {
for rte in file.rte.iter() {
if rte_ids.contains(&rte.id) {
self.add_segment_stats(&rte.seg, &mut selected);
}
}
}
}
Selection::Track { file_id, trk_ids } => {
if let Some(file) = state.get(file_id) {
for trk in file.trk.iter() {
@@ -129,8 +136,8 @@ impl StatisticsCache {
}
fn add_file_stats<'a>(&'a self, file: &'a File, selected: &mut Vec<SelectedSegment<'a>>) {
for trk in file.trk.iter() {
self.add_track_stats(trk, selected);
for trkseg in file.segments() {
self.add_segment_stats(trkseg, selected);
}
}
+3 -9
View File
@@ -82,7 +82,7 @@ impl<H: History> Engine<H> {
.find(|wpt| wpt.id == *id)
}
/// A trackpoint of a segment of a file. `None` if it does not exist.
/// A trackpoint of a segment (of a route or a track) of a file. `None` if it does not exist.
pub fn trackpoint(
&self,
file_id: &FileId,
@@ -93,9 +93,7 @@ impl<H: History> Engine<H> {
.history
.current()?
.get(file_id)?
.trk
.iter()
.flat_map(|trk| &trk.trkseg)
.segments()
.find(|seg| seg.id == *segment_id)?;
(index < segment.len()).then(|| &segment[index])
}
@@ -109,11 +107,7 @@ impl<H: History> Engine<H> {
self.selection
.segment_locations(files, &self.order.0)
.iter()
.flat_map(|location| {
reduction_distances(
&files[&location.file_id].trk[location.trk].trkseg[location.seg],
)
})
.flat_map(|location| reduction_distances(location.segment(files)))
.collect()
}
+44 -13
View File
@@ -3,10 +3,18 @@ use std::rc::Rc;
use uuid::Uuid;
use crate::{
File, FileId, Files, Selection, Track, TrackId, TrackSegment, TrackSegmentId, Waypoint,
WaypointId,
File, FileId, Files, Route, RouteId, Selection, Track, TrackId, TrackSegment, TrackSegmentId,
Waypoint, WaypointId,
};
/// A route in the clipboard, with the name of the file it comes from (the name of the new file
/// when it is pasted on its own).
#[derive(Debug, Clone)]
pub struct ClipboardRoute {
pub route: Route,
pub file_name: String,
}
/// A track in the clipboard, with the name of the file it comes from (the name of the new file
/// when it is pasted on its own).
#[derive(Debug, Clone)]
@@ -31,6 +39,7 @@ pub struct ClipboardSegment {
#[derive(Debug, Clone)]
pub enum ClipboardContent {
Files(Vec<Rc<File>>),
Routes(Vec<ClipboardRoute>),
Tracks(Vec<ClipboardTrack>),
Segments(Vec<ClipboardSegment>),
Waypoints(Vec<Waypoint>),
@@ -84,6 +93,19 @@ impl Clipboard {
.filter_map(|id| files.get(id).cloned())
.collect(),
),
Selection::Route { file_id, rte_ids } => {
let file = files.get(file_id)?;
ClipboardContent::Routes(
file.rte
.iter()
.filter(|rte| rte_ids.contains(&rte.id))
.map(|rte| ClipboardRoute {
route: rte.clone(),
file_name: file.info.name.clone(),
})
.collect(),
)
}
Selection::Track { file_id, trk_ids } => {
let file = files.get(file_id)?;
ClipboardContent::Tracks(
@@ -144,32 +166,37 @@ impl Clipboard {
pub fn is_empty(&self) -> bool {
match &self.content {
ClipboardContent::Files(files) => files.is_empty(),
ClipboardContent::Routes(routes) => routes.is_empty(),
ClipboardContent::Tracks(tracks) => tracks.is_empty(),
ClipboardContent::Segments(segments) => segments.is_empty(),
ClipboardContent::Waypoints(waypoints) => waypoints.is_empty(),
}
}
/// Whether the content can be pasted when `target` is selected: files and tracks can be
/// pasted anywhere above them in the file tree, and nothing selected stands for the list of
/// files.
/// Whether the content can be pasted when `target` is selected: files, routes and tracks can
/// be pasted anywhere above them in the file tree (routes and tracks are at the same level),
/// and nothing selected stands for the list of files.
///
/// | copied | pasted when selected |
/// |-----------|------------------------------------------|
/// | files | nothing, a file |
/// | tracks | nothing, a file, a track |
/// | segments | nothing, a file, a track, a segment |
/// | waypoints | a file, the waypoints, a waypoint |
/// | copied | pasted when selected |
/// |-----------|-----------------------------------------------|
/// | files | nothing, a file |
/// | routes | nothing, a file, a route, a track |
/// | tracks | nothing, a file, a route, a track |
/// | segments | nothing, a file, a route, a track, a segment |
/// | waypoints | a file, the waypoints, a waypoint |
pub fn can_paste(&self, target: &Selection) -> bool {
use ClipboardContent as C;
use Selection::*;
matches!(
(&self.content, target),
(C::Files(_), Empty | File { .. })
| (C::Tracks(_), Empty | File { .. } | Track { .. })
| (
C::Routes(_) | C::Tracks(_),
Empty | File { .. } | Route { .. } | Track { .. }
)
| (
C::Segments(_),
Empty | File { .. } | Track { .. } | TrackSegment { .. }
Empty | File { .. } | Route { .. } | Track { .. } | TrackSegment { .. }
)
| (
C::Waypoints(_),
@@ -183,6 +210,7 @@ impl Clipboard {
#[derive(Debug, Clone, PartialEq, Eq)]
pub enum ClipboardIds {
Files(Vec<FileId>),
Routes(Vec<RouteId>),
Tracks(Vec<TrackId>),
Segments(Vec<TrackSegmentId>),
Waypoints(Vec<WaypointId>),
@@ -194,6 +222,9 @@ impl ClipboardContent {
ClipboardContent::Files(files) => {
ClipboardIds::Files(files.iter().map(|file| file.id).collect())
}
ClipboardContent::Routes(routes) => {
ClipboardIds::Routes(routes.iter().map(|rte| rte.route.id).collect())
}
ClipboardContent::Tracks(tracks) => {
ClipboardIds::Tracks(tracks.iter().map(|trk| trk.track.id).collect())
}
+81 -27
View File
@@ -1,6 +1,8 @@
use std::collections::HashSet;
use crate::{FileId, Files, TrackId, TrackSegmentId, WaypointId};
use crate::{
File, FileId, Files, RouteId, SegmentIndex, TrackId, TrackSegment, TrackSegmentId, WaypointId,
};
/// How a new selection combines with the current one.
#[derive(Debug, Clone, Copy, PartialEq, Eq, serde::Deserialize)]
@@ -23,6 +25,10 @@ pub enum Selection {
File {
file_ids: HashSet<FileId>,
},
Route {
file_id: FileId,
rte_ids: HashSet<RouteId>,
},
Track {
file_id: FileId,
trk_ids: HashSet<TrackId>,
@@ -41,12 +47,19 @@ pub enum Selection {
},
}
/// Where a track segment is: its file, then its track and its own position in it.
/// Where a segment is: its file, then its place in it (its route, or its track and its own
/// position in it).
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
pub struct SegmentLocation {
pub file_id: FileId,
pub trk: usize,
pub seg: usize,
pub index: SegmentIndex,
}
impl SegmentLocation {
/// The segment, in `files`. Panics if it is not there.
pub fn segment<'a>(&self, files: &'a Files) -> &'a TrackSegment {
files[&self.file_id].segment(self.index)
}
}
impl Selection {
@@ -89,23 +102,18 @@ impl Selection {
}
impl Selection {
/// The track segments covered by the selection, in order: the segments of several selected
/// files follow `order` (the files missing from it come last), then the tracks and segments
/// follow the file. Waypoints are not segments: selecting only waypoints covers none.
/// The segments covered by the selection, in order: the segments of several selected files
/// follow `order` (the files missing from it come last), then the routes, tracks and segments
/// follow the file (see [`File::segments`]). Waypoints are not segments: selecting only
/// waypoints covers none.
pub fn segment_locations(&self, files: &Files, order: &[FileId]) -> Vec<SegmentLocation> {
let mut locations = vec![];
let mut add = |file_id: FileId,
trk_filter: &dyn Fn(&crate::Track) -> bool,
seg_filter: &dyn Fn(&crate::TrackSegment) -> bool| {
let mut add = |file_id: FileId, filter: &dyn Fn(&File, SegmentIndex) -> bool| {
let Some(file) = files.get(&file_id) else {
return;
};
for (trk, track) in file.trk.iter().enumerate().filter(|(_, t)| trk_filter(t)) {
for (seg, segment) in track.trkseg.iter().enumerate() {
if seg_filter(segment) {
locations.push(SegmentLocation { file_id, trk, seg });
}
}
for index in file.segment_indices().filter(|index| filter(file, *index)) {
locations.push(SegmentLocation { file_id, index });
}
};
match self {
@@ -116,25 +124,35 @@ impl Selection {
.filter(|id| file_ids.contains(id))
.chain(file_ids.iter().filter(|id| !order.contains(id)));
for id in ordered {
add(*id, &|_| true, &|_| true);
add(*id, &|_, _| true);
}
}
Selection::Track { file_id, trk_ids } => {
add(*file_id, &|trk| trk_ids.contains(&trk.id), &|_| true)
}
Selection::Route { file_id, rte_ids } => add(*file_id, &|file, index| match index {
SegmentIndex::Route(rte) => rte_ids.contains(&file.rte[rte].id),
SegmentIndex::Track { .. } => false,
}),
Selection::Track { file_id, trk_ids } => add(*file_id, &|file, index| match index {
SegmentIndex::Track { trk, .. } => trk_ids.contains(&file.trk[trk].id),
SegmentIndex::Route(_) => false,
}),
Selection::TrackSegment {
file_id,
trk_id,
trkseg_ids,
} => add(*file_id, &|trk| trk.id == *trk_id, &|seg| {
trkseg_ids.contains(&seg.id)
} => add(*file_id, &|file, index| match index {
SegmentIndex::Track { trk, seg } => {
let track = &file.trk[trk];
track.id == *trk_id && trkseg_ids.contains(&track.trkseg[seg].id)
}
SegmentIndex::Route(_) => false,
}),
}
locations
}
/// Drops what does not exist anymore (e.g. after an undo): files, and the tracks, segments
/// and waypoints of the files that remain. A selection left with nothing becomes empty.
/// Drops what does not exist anymore (e.g. after an undo): files, and the routes, tracks,
/// segments and waypoints of the files that remain. A selection left with nothing becomes
/// empty.
pub fn retain_existing(&mut self, files: &Files) {
let empty = match self {
Selection::Empty => false,
@@ -142,6 +160,13 @@ impl Selection {
file_ids.retain(|id| files.contains_key(id));
file_ids.is_empty()
}
Selection::Route { file_id, rte_ids } => match files.get(file_id) {
Some(file) => {
rte_ids.retain(|id| file.rte.iter().any(|rte| rte.id == *id));
rte_ids.is_empty()
}
None => true,
},
Selection::Track { file_id, trk_ids } => match files.get(file_id) {
Some(file) => {
trk_ids.retain(|id| file.trk.iter().any(|trk| trk.id == *id));
@@ -185,6 +210,13 @@ impl Selection {
(Selection::File { file_ids }, Selection::File { file_ids: other }) => {
file_ids.extend(other)
}
(
Selection::Route { file_id, rte_ids },
Selection::Route {
file_id: other_file,
rte_ids: other,
},
) if *file_id == other_file => rte_ids.extend(other),
(
Selection::Track { file_id, trk_ids },
Selection::Track {
@@ -232,6 +264,16 @@ impl Selection {
toggle_ids(file_ids, other);
file_ids.is_empty()
}
(
Selection::Route { file_id, rte_ids },
Selection::Route {
file_id: other_file,
rte_ids: other,
},
) if *file_id == other_file => {
toggle_ids(rte_ids, other);
rte_ids.is_empty()
}
(
Selection::Track { file_id, trk_ids },
Selection::Track {
@@ -325,7 +367,8 @@ impl Selection {
match self {
Selection::Empty => Selection::Empty,
Selection::File { .. } => self.clone(),
Selection::Track { file_id, .. }
Selection::Route { file_id, .. }
| Selection::Track { file_id, .. }
| Selection::TrackSegment { file_id, .. }
| Selection::Waypoints { file_id }
| Selection::Waypoint { file_id, .. } => Selection::File {
@@ -335,7 +378,7 @@ impl Selection {
}
/// Selects all the elements of the same kind as the selected ones, in the same place: all
/// the files, all the tracks of the file, all the segments of the track, all the waypoints
/// the files, all the routes (or tracks) of the file, all the segments of the track, all the waypoints
/// of the file. All the files when nothing is selected. `None` when there is nothing to
/// select (the waypoints node, or a file that does not exist).
pub fn all_at_level(&self, files: &Files, order: &[FileId]) -> Option<Selection> {
@@ -343,6 +386,10 @@ impl Selection {
Selection::Empty | Selection::File { .. } => Some(Selection::File {
file_ids: order.iter().copied().collect(),
}),
Selection::Route { file_id, .. } => Some(Selection::Route {
file_id: *file_id,
rte_ids: files.get(file_id)?.rte.iter().map(|rte| rte.id).collect(),
}),
Selection::Track { file_id, .. } => Some(Selection::Track {
file_id: *file_id,
trk_ids: files.get(file_id)?.trk.iter().map(|trk| trk.id).collect(),
@@ -371,7 +418,7 @@ impl Selection {
/// The element next to the selected ones, as a selection of that one element: the next one
/// (`down`) or the previous one among the elements of the same kind, in the same place.
/// The tracks, segments and waypoints stop at their ends, the files wrap around and skip
/// The routes, tracks, segments and waypoints stop at their ends, the files wrap around and skip
/// the ones that are already selected. `None` when there is no such element.
pub fn neighbour(&self, files: &Files, order: &[FileId], down: bool) -> Option<Selection> {
match self {
@@ -393,6 +440,13 @@ impl Selection {
}
}
}
Selection::Route { file_id, rte_ids } => {
let ids: Vec<RouteId> = files.get(file_id)?.rte.iter().map(|r| r.id).collect();
Some(Selection::Route {
file_id: *file_id,
rte_ids: [neighbour(&ids, rte_ids, down)?].into(),
})
}
Selection::Track { file_id, trk_ids } => {
let ids: Vec<TrackId> = files.get(file_id)?.trk.iter().map(|t| t.id).collect();
Some(Selection::Track {
+2
View File
@@ -87,6 +87,7 @@ fn new(engine: &mut TestEngine, name: &str) -> bool {
Command::New(New {
name,
trackpoint: None,
route: false,
}),
)
}
@@ -1489,6 +1490,7 @@ mod per_file_history {
.execute(Action::Edit(Command::New(New {
name,
trackpoint: None,
route: false,
})))
.unwrap();
*engine.order().last().unwrap()
+88 -36
View File
@@ -18,9 +18,9 @@ use serde::{Deserialize, Serialize};
use uuid::Uuid;
use crate::{
File, FileId, FileInfo, Track, TrackId, TrackInfo, TrackSegment, TrackSegmentId,
TrackpointCategories, TrackpointChunk, TrackpointChunkId, Waypoint, WaypointChunk,
WaypointChunkId, Waypoints,
File, FileId, FileInfo, Route, RouteId, Track, TrackId, TrackInfo, TrackSegment,
TrackSegmentId, TrackpointCategories, TrackpointChunk, TrackpointChunkId, Waypoint,
WaypointChunk, WaypointChunkId, Waypoints,
};
/// The version that is written.
@@ -54,10 +54,17 @@ fn parse<'a, T: Deserialize<'a>>(data: &'a [u8]) -> Result<T, DecodeError> {
/// A file without its trackpoints and waypoints, which are in chunks.
pub struct FileRecord {
info: FileInfo,
routes: Vec<RouteRecord>,
tracks: Vec<TrackRecord>,
waypoints: Vec<WaypointChunkId>,
}
struct RouteRecord {
id: RouteId,
info: TrackInfo,
segment: SegmentRecord,
}
struct TrackRecord {
id: TrackId,
info: TrackInfo,
@@ -69,24 +76,35 @@ struct SegmentRecord {
chunks: Vec<TrackpointChunkId>,
}
impl From<v1::SegmentRecord1> for SegmentRecord {
fn from(segment: v1::SegmentRecord1) -> Self {
Self {
id: TrackSegmentId(segment.id),
chunks: segment.chunks.into_iter().map(TrackpointChunkId).collect(),
}
}
}
impl From<v1::FileRecord1> for FileRecord {
fn from(record: v1::FileRecord1) -> Self {
Self {
info: record.info.into(),
routes: record
.routes
.into_iter()
.map(|route| RouteRecord {
id: RouteId(route.id),
info: route.info.into(),
segment: route.segment.into(),
})
.collect(),
tracks: record
.tracks
.into_iter()
.map(|track| TrackRecord {
id: TrackId(track.id),
info: track.info.into(),
segments: track
.segments
.into_iter()
.map(|segment| SegmentRecord {
id: TrackSegmentId(segment.id),
chunks: segment.chunks.into_iter().map(TrackpointChunkId).collect(),
})
.collect(),
segments: track.segments.into_iter().map(Into::into).collect(),
})
.collect(),
waypoints: record.waypoints.into_iter().map(WaypointChunkId).collect(),
@@ -108,16 +126,27 @@ impl FileRecord {
info: self.info,
..Default::default()
};
for track in self.tracks {
let mut trkseg = Vec::with_capacity(track.segments.len());
for segment in track.segments {
let mut built = TrackSegment::default();
built.id = segment.id;
for chunk in &segment.chunks {
built.push_shared(trackpoint_chunks(chunk)?);
}
trkseg.push(built);
let segment = |record: SegmentRecord| {
let mut built = TrackSegment::default();
built.id = record.id;
for chunk in &record.chunks {
built.push_shared(trackpoint_chunks(chunk)?);
}
Some(built)
};
for route in self.routes {
file.rte.push(Route {
id: route.id,
info: route.info,
seg: segment(route.segment)?,
});
}
for track in self.tracks {
let trkseg = track
.segments
.into_iter()
.map(segment)
.collect::<Option<Vec<_>>>()?;
file.trk.push(Track {
id: track.id,
info: track.info,
@@ -292,6 +321,20 @@ mod tests {
});
let mut empty = TrackSegment::default();
empty.id = TrackSegmentId(id(6));
// a route, whose segment shares the chunk of the track
let mut route_segment = TrackSegment::default();
route_segment.id = TrackSegmentId(id(10));
route_segment.push_shared(segment.chunks()[0].clone());
file.rte.push(Route {
id: RouteId(id(9)),
info: TrackInfo {
name: Some("route".into()),
type_: Some("Hiking".into()),
color: Some("00ff00".into()),
..Default::default()
},
seg: route_segment,
});
file.trk.push(Track {
id: TrackId(id(2)),
info: TrackInfo {
@@ -352,9 +395,7 @@ mod tests {
HashMap<WaypointChunkId, Rc<WaypointChunk>>,
) {
let trackpoints = file
.trk
.iter()
.flat_map(|track| &track.trkseg)
.segments()
.flat_map(|segment| segment.chunks().iter().map(|c| (c.id, c.clone())))
.collect();
let waypoints = file
@@ -373,19 +414,28 @@ mod tests {
})
}
fn assert_same_segment(a: &TrackSegment, b: &TrackSegment) {
assert_eq!(a.id, b.id);
assert_eq!(a.len(), b.len());
for (a, b) in a.chunks().iter().zip(b.chunks()) {
assert!(Rc::ptr_eq(a, b), "the chunks are the ones that were given");
}
}
fn assert_same_file(a: &File, b: &File) {
assert_eq!(a.id, b.id);
assert_eq!(a.info, b.info);
assert_eq!(a.rte.len(), b.rte.len());
for (a, b) in a.rte.iter().zip(&b.rte) {
assert_eq!((a.id, &a.info), (b.id, &b.info));
assert_same_segment(&a.seg, &b.seg);
}
assert_eq!(a.trk.len(), b.trk.len());
for (a, b) in a.trk.iter().zip(&b.trk) {
assert_eq!((a.id, &a.info), (b.id, &b.info));
assert_eq!(a.trkseg.len(), b.trkseg.len());
for (a, b) in a.trkseg.iter().zip(&b.trkseg) {
assert_eq!(a.id, b.id);
assert_eq!(a.len(), b.len());
for (a, b) in a.chunks().iter().zip(b.chunks()) {
assert!(Rc::ptr_eq(a, b), "the chunks are the ones that were given");
}
assert_same_segment(a, b);
}
}
assert_eq!(a.wpt.len(), b.wpt.len());
@@ -425,7 +475,7 @@ mod tests {
.unwrap()
.build(file.id, &|_| None, &|_| None)
.unwrap();
assert!(built.trk.is_empty() && built.wpt.is_empty());
assert!(built.rte.is_empty() && built.trk.is_empty() && built.wpt.is_empty());
}
/// The names of the categories of a trackpoint.
@@ -611,13 +661,15 @@ mod tests {
"70733a2f2f6578616d706c652e636f6d2f736f6d656f6e65021868747470733a2f2f6578616d706c652e636f",
"6d2f66696c65012074657874206f662068747470733a2f2f6578616d706c652e636f6d2f66696c651a687474",
"70733a2f2f6578616d706c652e636f6d2f66696c652d32012274657874206f662068747470733a2f2f657861",
"6d706c652e636f6d2f66696c652d320180a0abfef96201100000000000000000000000000000000201057472",
"61636b0103636d740104646573630103737263021968747470733a2f2f6578616d706c652e636f6d2f747261",
"636b012174657874206f662068747470733a2f2f6578616d706c652e636f6d2f747261636b1b68747470733a",
"2f2f6578616d706c652e636f6d2f747261636b2d32012374657874206f662068747470733a2f2f6578616d70",
"6c652e636f6d2f747261636b2d3201074379636c696e67010666663030303001000000000000e03f01000000",
"0000001840021000000000000000000000000000000003011000000000000000000000000000000004100000",
"000000000000000000000000000600011000000000000000000000000000000005",
"6d706c652e636f6d2f66696c652d320180a0abfef9620110000000000000000000000000000000090105726f",
"75746500000000010648696b696e6701063030666630300000100000000000000000000000000000000a0110",
"000000000000000000000000000000040110000000000000000000000000000000020105747261636b010363",
"6d740104646573630103737263021968747470733a2f2f6578616d706c652e636f6d2f747261636b01217465",
"7874206f662068747470733a2f2f6578616d706c652e636f6d2f747261636b1b68747470733a2f2f6578616d",
"706c652e636f6d2f747261636b2d32012374657874206f662068747470733a2f2f6578616d706c652e636f6d",
"2f747261636b2d3201074379636c696e67010666663030303001000000000000e03f01000000000000184002",
"1000000000000000000000000000000003011000000000000000000000000000000004100000000000000000",
"000000000000000600011000000000000000000000000000000005",
);
const TRACKPOINTS_V1: &str = concat!(
"01020667726176656c0761737068616c74010470617468010668696b696e6701046d74623003000000000000",
+29 -10
View File
@@ -8,8 +8,8 @@ use serde::{Deserialize, Serialize};
use uuid::Uuid;
use crate::{
Author, Categories, File, FileInfo, Link, LngLat, TrackInfo, Trackpoint, TrackpointCategories,
Waypoint, WaypointId,
Author, Categories, File, FileInfo, Link, LngLat, TrackInfo, TrackSegment, Trackpoint,
TrackpointCategories, Waypoint, WaypointId,
};
#[derive(Serialize, Deserialize)]
@@ -133,10 +133,18 @@ impl From<TrackInfo1> for TrackInfo {
#[derive(Serialize, Deserialize)]
pub struct FileRecord1 {
pub info: FileInfo1,
pub routes: Vec<RouteRecord1>,
pub tracks: Vec<TrackRecord1>,
pub waypoints: Vec<Uuid>,
}
#[derive(Serialize, Deserialize)]
pub struct RouteRecord1 {
pub id: Uuid,
pub info: TrackInfo1,
pub segment: SegmentRecord1,
}
#[derive(Serialize, Deserialize)]
pub struct TrackRecord1 {
pub id: Uuid,
@@ -150,24 +158,35 @@ pub struct SegmentRecord1 {
pub chunks: Vec<Uuid>,
}
impl SegmentRecord1 {
fn new(segment: &TrackSegment) -> Self {
Self {
id: segment.id.0,
chunks: segment.chunks().iter().map(|chunk| chunk.id.0).collect(),
}
}
}
impl FileRecord1 {
pub fn new(file: &File) -> Self {
Self {
info: (&file.info).into(),
routes: file
.rte
.iter()
.map(|route| RouteRecord1 {
id: route.id.0,
info: (&route.info).into(),
segment: SegmentRecord1::new(&route.seg),
})
.collect(),
tracks: file
.trk
.iter()
.map(|track| TrackRecord1 {
id: track.id.0,
info: (&track.info).into(),
segments: track
.trkseg
.iter()
.map(|segment| SegmentRecord1 {
id: segment.id.0,
chunks: segment.chunks().iter().map(|chunk| chunk.id.0).collect(),
})
.collect(),
segments: track.trkseg.iter().map(SegmentRecord1::new).collect(),
})
.collect(),
waypoints: file.wpt.chunks().iter().map(|chunk| chunk.id.0).collect(),
+1 -2
View File
@@ -217,8 +217,7 @@ impl Persister {
.files
.get(id)
.is_none_or(|saved| !Rc::ptr_eq(saved, file));
let segments = file.trk.iter().flat_map(|track| &track.trkseg);
for chunk in segments.flat_map(|segment| segment.chunks()) {
for chunk in file.segments().flat_map(|segment| segment.chunks()) {
let key = ChunkKey::trackpoints(chunk.id.0);
if chunks.insert(key) && changed && !self.chunks.contains(&key) {
batch
+9
View File
@@ -7,6 +7,15 @@ edition = "2024"
[lib]
name = "gpx_studio_desktop_lib"
crate-type = ["staticlib", "cdylib", "rlib"]
# the tests are in `tests/app`: on Windows, a test binary of the library or of the app could not
# start without the manifest, which `build.rs` only gives to the binaries of `tests`
test = false
doctest = false
[[bin]]
name = "gpx-studio-desktop"
path = "src/main.rs"
test = false
[dev-dependencies]
tauri = { version = "2", features = ["test"] }
+17
View File
@@ -1,3 +1,20 @@
fn main() {
// On Windows, the dialogs need Common Controls v6, which a binary only gets with a manifest that
// asks for it: tauri-build gives one to the app, but not to the test binaries, which would stop
// with STATUS_ENTRYPOINT_NOT_FOUND before running any test. This gives them the same manifest
// (`-tests` only reaches the binaries of `tests`, hence the tests are there).
if std::env::var("CARGO_CFG_TARGET_OS").as_deref() == Ok("windows")
&& std::env::var("CARGO_CFG_TARGET_ENV").as_deref() == Ok("msvc")
{
let manifest = std::path::Path::new(env!("CARGO_MANIFEST_DIR"))
.join("tests")
.join("windows-test.manifest");
println!("cargo:rerun-if-changed={}", manifest.display());
println!("cargo:rustc-link-arg-tests=/MANIFEST:EMBED");
println!(
"cargo:rustc-link-arg-tests=/MANIFESTINPUT:{}",
manifest.display()
);
}
tauri_build::build()
}
+18 -21
View File
@@ -14,7 +14,6 @@ use tauri::{AppHandle, Runtime, State};
use tauri_plugin_dialog::{DialogExt, FilePath};
use uuid::Uuid;
use crate::commands::edit;
use crate::state::AppState;
use crate::workspace::{Entry, Unsaved, WorkspaceState, free_path};
@@ -313,7 +312,6 @@ pub struct ExportArguments {
atemp: bool,
power: bool,
osm: bool,
as_route: bool,
}
impl From<ExportArguments> for ExportOptions {
@@ -325,7 +323,6 @@ impl From<ExportArguments> for ExportOptions {
atemp: arguments.atemp,
power: arguments.power,
osm: arguments.osm,
as_route: arguments.as_route,
}
}
}
@@ -388,28 +385,28 @@ pub fn create_folder(
state.core(move |core| core.workspace.create_folder(&parent, &name))
}
/// Renames a file or a folder of the open folder. An open file is renamed like in its info (it is
/// selected), which can be undone; the rest is renamed on the disk, and the open files that are in
/// a folder that is renamed follow.
/// Renames a file or a folder of the open folder, on the disk only: an open file keeps its name,
/// and follows. Not undoable.
#[tauri::command(async)]
pub fn rename_path(
state: State<AppState>,
path: PathBuf,
name: String,
) -> Result<api::Outcome, String> {
pub fn rename_path(state: State<AppState>, path: PathBuf, name: String) -> Result<(), String> {
state.core(move |core| {
if let Some(id) = core.workspace.linked(&path) {
let selected = api::select(&mut core.engine, &[id], gpx_engine::SelectMode::Replace);
let desc = core
.engine
.file_structure(&FileId(id))
.and_then(|structure| structure.desc.clone())
.unwrap_or_default();
return Ok(selected.merge(edit::rename_selection(core, name, desc)));
}
core.workspace.rename_path(&path, &name)?;
core.commit(None);
Ok(api::Outcome::failed_none())
Ok(())
})
}
/// Renames the file on the disk of an open file, next to where it is (see `Workspace::rename_file`).
/// Not undoable, the name of the file in the engine is not changed. Says whether it was renamed:
/// not for a file that is not on the disk.
#[tauri::command(async)]
pub fn rename_file(state: State<AppState>, file_id: Uuid, name: String) -> Result<bool, String> {
state.core(move |core| {
let renamed = core.workspace.rename_file(file_id, &name)?;
if renamed {
core.commit(None);
}
Ok(renamed)
})
}
+15 -19
View File
@@ -1,31 +1,17 @@
//! Editing the metadata, the style, the structure and the waypoints of the files, and the tools:
//! reversing, cropping, reducing, timestamps, merging, cleaning...
use gpx_engine::{CleanType, FileId, MergeType, Selection, SplitType, api};
use gpx_engine::{CleanType, MergeType, SplitType, api};
use tauri::State;
use uuid::Uuid;
use crate::state::{AppState, Core};
use crate::state::AppState;
/// Names and describes the selected files or tracks. A file on the disk is renamed with it (and back
/// when it is undone): its name is one that a file can have, and not the one of another file.
/// Names and describes the selected files or tracks. The files on the disk keep their name: see
/// `disk::rename_file`.
#[tauri::command(async)]
pub fn metadata(state: State<AppState>, name: String, desc: String) -> api::Outcome {
state.core(move |core| rename_selection(core, name, desc))
}
/// See `metadata`.
pub fn rename_selection(core: &mut Core, name: String, desc: String) -> api::Outcome {
let mut name = name;
if let Selection::File { file_ids } = core.engine.selection() {
for FileId(id) in file_ids.clone() {
match core.workspace.name_for(id, &name) {
Ok(valid) => name = valid,
Err(error) => return api::Outcome::failed(error),
}
}
}
core.edit(move |e| api::metadata(e, &name, &desc))
state.run(move |e| api::metadata(e, &name, &desc))
}
#[tauri::command(async)]
@@ -38,6 +24,16 @@ pub fn style(
state.run(move |e| api::style(e, color.as_deref(), opacity, width))
}
#[tauri::command(async)]
pub fn convert(state: State<AppState>) -> api::Outcome {
state.run(api::convert)
}
#[tauri::command(async)]
pub fn new_route(state: State<AppState>) -> api::Outcome {
state.run(api::new_route)
}
#[tauri::command(async)]
pub fn new_track(state: State<AppState>) -> api::Outcome {
state.run(api::new_track)
+9 -3
View File
@@ -16,10 +16,16 @@ pub fn new_file(
lng: Option<f64>,
lat: Option<f64>,
ele: Option<f64>,
route: Option<bool>,
folder: Option<PathBuf>,
) -> api::Outcome {
let route = route.unwrap_or_default();
// in `folder` if it is given, otherwise like the selected file, otherwise in the open folder
state.core(move |core| core.creating(folder, true, |e| api::new_file(e, &name, lng, lat, ele)))
state.core(move |core| {
core.creating(folder, true, |e| {
api::new_file(e, &name, lng, lat, ele, route)
})
})
}
/// The files are in the body, one after the other, the headers `lengths` and `names` (JSON,
@@ -109,10 +115,10 @@ pub fn move_elements(
#[tauri::command(async)]
pub fn undo(state: State<AppState>) -> api::Outcome {
state.edit(api::undo)
state.run(api::undo)
}
#[tauri::command(async)]
pub fn redo(state: State<AppState>) -> api::Outcome {
state.edit(api::redo)
state.run(api::redo)
}
-2
View File
@@ -79,7 +79,6 @@ pub fn export_file(
atemp: bool,
power: bool,
osm: bool,
as_route: bool,
) -> Bytes {
let options = ExportOptions {
time,
@@ -88,7 +87,6 @@ pub fn export_file(
atemp,
power,
osm,
as_route,
};
optional(state.read(move |e| api::export_file(e, file_id, options)))
}
+10
View File
@@ -21,6 +21,16 @@ pub fn arrow_select(state: State<AppState>, down: bool, add: bool) -> api::Outco
state.run(move |e| api::arrow_select(e, down, add))
}
#[tauri::command(async)]
pub fn select_routes(
state: State<AppState>,
file_id: Uuid,
route_ids: Vec<Uuid>,
mode: SelectMode,
) -> api::Outcome {
state.run(move |e| api::select_routes(e, file_id, &route_ids, mode))
}
#[tauri::command(async)]
pub fn select_tracks(
state: State<AppState>,
+10 -9
View File
@@ -3,20 +3,20 @@
mod bytes;
mod commands;
mod state;
mod store;
mod workspace;
pub mod state;
pub mod store;
pub mod workspace;
use commands::*;
use tauri::{Emitter, Manager};
/// The engine of the app: each file has its own history, so that undoing only concerns the
/// selected files, and opening or closing a file is not something to undo.
pub(crate) type Engine = gpx_engine::Engine<gpx_engine::PerFileHistory>;
pub type Engine = gpx_engine::Engine<gpx_engine::PerFileHistory>;
/// The GPX files among what the system gave to the app (the arguments of the command line, when a
/// file is opened with it).
fn gpx_paths(arguments: impl IntoIterator<Item = String>) -> Vec<std::path::PathBuf> {
pub fn gpx_paths(arguments: impl IntoIterator<Item = String>) -> Vec<std::path::PathBuf> {
arguments
.into_iter()
.map(std::path::PathBuf::from)
@@ -108,12 +108,13 @@ fn menu<R: tauri::Runtime>(app: &tauri::AppHandle<R>) -> tauri::Result<tauri::me
}
/// Gives the builder the commands that reach the engine (which it has to be given as `AppState`).
fn register_commands<R: tauri::Runtime>(builder: tauri::Builder<R>) -> tauri::Builder<R> {
pub fn register_commands<R: tauri::Runtime>(builder: tauri::Builder<R>) -> tauri::Builder<R> {
builder.invoke_handler(tauri::generate_handler![
// selection
selection::select,
selection::select_all,
selection::arrow_select,
selection::select_routes,
selection::select_tracks,
selection::select_segments,
selection::select_waypoint_group,
@@ -134,6 +135,8 @@ fn register_commands<R: tauri::Runtime>(builder: tauri::Builder<R>) -> tauri::Bu
// edits and tools
edit::metadata,
edit::style,
edit::convert,
edit::new_route,
edit::new_track,
edit::new_track_segment,
edit::new_waypoint,
@@ -223,6 +226,7 @@ fn register_commands<R: tauri::Runtime>(builder: tauri::Builder<R>) -> tauri::Bu
disk::list_folder,
disk::create_folder,
disk::rename_path,
disk::rename_file,
disk::move_paths,
disk::delete_paths,
// the system
@@ -230,6 +234,3 @@ fn register_commands<R: tauri::Runtime>(builder: tauri::Builder<R>) -> tauri::Bu
system::reveal_path,
])
}
#[cfg(test)]
mod tests;
+7 -45
View File
@@ -94,11 +94,8 @@ impl Core {
self.commit(None);
}
/// Runs `f` on the engine, which can change: what changed is kept, and the files on the disk are
/// renamed like the files of the engine were (an edit of the name, or its undo). Gives what
/// could not be renamed.
fn editing<T>(&mut self, f: impl FnOnce(&mut Engine) -> T) -> (T, Vec<String>) {
let names = self.workspace.names(&self.engine);
/// Runs `f` on the engine, which can change: what changed is kept.
fn editing<T>(&mut self, f: impl FnOnce(&mut Engine) -> T) -> T {
let preview = self.workspace.preview;
let preview_before = preview.and_then(|id| signature(&self.engine, id));
let result = f(&mut self.engine);
@@ -107,32 +104,18 @@ impl Core {
if let Some(id) = preview.filter(|_| kept) {
self.workspace.keep(&[id]);
}
let (renamed, errors) = self.workspace.follow_names(&mut self.engine, &names);
// and it goes when something else is selected
let dropped = self.workspace.drop_preview(&mut self.engine);
if kept || renamed || dropped.is_some() {
if kept || dropped.is_some() {
self.commit(dropped.as_ref());
} else {
self.persist();
}
(result, errors)
result
}
/// Like `editing`, for an edit that can rename files: the outcome tells the first thing that
/// could not be renamed (a key of the dictionaries of the frontend, which shows it).
pub fn edit(&mut self, f: impl FnOnce(&mut Engine) -> api::Outcome) -> api::Outcome {
let (outcome, errors) = self.editing(f);
match errors.first() {
Some(error) if !outcome.has_error() => {
outcome.merge(api::Outcome::failed(error.clone()))
}
_ => outcome,
}
}
/// After files were written (`result`): what changed in the engine (the files are called like
/// the files they were written to) is kept and told, with where the files are. Says whether
/// they were written.
/// After files were written (`result`): what changed in the engine is kept and told, with where
/// the files are. Says whether they were written.
pub fn wrote(&mut self, result: Result<api::Outcome, String>) -> Result<bool, String> {
self.commit(result.as_ref().ok().filter(|outcome| outcome.changed));
result.map(|_| true)
@@ -297,13 +280,6 @@ impl AppState {
})
}
/// An engine that keeps its data in a directory of its own, and has nobody to tell.
#[cfg(test)]
pub fn temporary() -> Self {
let directory = std::env::temp_dir().join(format!("gpx-studio-{}", uuid::Uuid::new_v4()));
Self::open(directory, Arc::new(|_, _| {})).expect("the storage can be opened")
}
/// Opens files that the system gave to the app, see `Core::open_external`.
pub fn open_external(&self, paths: Vec<PathBuf>) {
self.core(move |core| core.open_external(&paths));
@@ -325,21 +301,7 @@ impl AppState {
/// Runs `f` on the engine, which can change: what changed is kept (see `Core::editing`).
pub fn run<T: Send + 'static>(&self, f: impl FnOnce(&mut Engine) -> T + Send + 'static) -> T {
self.core(move |core| {
let (result, errors) = core.editing(f);
for error in errors {
eprintln!("{error}");
}
result
})
}
/// Like `run`, for an edit that can rename files: the outcome tells what could not be renamed.
pub fn edit(
&self,
f: impl FnOnce(&mut Engine) -> api::Outcome + Send + 'static,
) -> api::Outcome {
self.core(move |core| core.edit(f))
self.core(move |core| core.editing(f))
}
/// Reads from the engine.
-66
View File
@@ -217,69 +217,3 @@ impl Storage for FileStorage {
Ok(())
}
}
#[cfg(test)]
mod tests {
use crate::Engine;
use gpx_engine::{Action, Command, Load, LoadFiles, Persister};
use super::*;
fn directory() -> PathBuf {
std::env::temp_dir().join(format!("gpx-studio-store-{}", Uuid::new_v4()))
}
fn engine_with_a_file() -> Engine {
let mut engine = Engine::default();
let data = std::fs::read("../gpx-rs/data/simple.gpx").unwrap();
engine
.execute(Action::Edit(Command::LoadFiles(LoadFiles {
files: vec![Load {
data: &data,
name: "simple",
}],
})))
.unwrap();
engine
}
#[test]
fn keeps_the_files_and_the_settings_between_sessions() {
let root = directory();
let engine = engine_with_a_file();
let order = engine.order().to_vec();
let storage = FileStorage::open(&root).unwrap();
pollster::block_on(Persister::default().save(&storage, engine.snapshot())).unwrap();
pollster::block_on(storage.put_setting("units", "\"metric\"")).unwrap();
// another session
let storage = FileStorage::open(&root).unwrap();
let restored = pollster::block_on(Persister::default().restore(&storage)).unwrap();
let mut other = Engine::default();
other.restore(restored);
assert_eq!(other.order(), order);
assert_eq!(
other.file_statistics(&order[0]).unwrap().total_distance,
engine.file_statistics(&order[0]).unwrap().total_distance
);
assert_eq!(storage.settings()["units"], "\"metric\"");
fs::remove_dir_all(root).unwrap();
}
#[test]
fn refuses_a_commit_that_is_not_based_on_the_head() {
let root = directory();
let storage = FileStorage::open(&root).unwrap();
let batch = Batch {
order: Some(vec![]),
..Default::default()
};
assert_eq!(pollster::block_on(storage.commit(&batch)), Ok(Revision(1)));
assert_eq!(
pollster::block_on(storage.commit(&batch)),
Err(CommitError::Stale { head: Revision(1) })
);
fs::remove_dir_all(root).unwrap();
}
}
+29 -897
View File
@@ -148,7 +148,7 @@ pub struct Entry {
pub is_dir: bool,
}
fn is_gpx(path: &Path) -> bool {
pub fn is_gpx(path: &Path) -> bool {
path.extension()
.and_then(|extension| extension.to_str())
.is_some_and(|extension| extension.eq_ignore_ascii_case("gpx"))
@@ -156,7 +156,7 @@ fn is_gpx(path: &Path) -> bool {
/// Compares names the way people read them: the numbers in them by their value ("track 2" before
/// "track 10"), the rest regardless of case.
fn natural_cmp(a: &str, b: &str) -> Ordering {
pub fn natural_cmp(a: &str, b: &str) -> Ordering {
let (mut a, mut b) = (a.chars().peekable(), b.chars().peekable());
loop {
let order = match (a.peek().copied(), b.peek().copied()) {
@@ -695,8 +695,8 @@ impl Workspace {
/// Writes files to the disk, replacing what is there: they are the same as their file from
/// then on. A file that is not on the disk goes in its folder, or at its path in `paths`, or in
/// `folder`, under its name (changed if that is the one of something else), which the file then
/// has (see `write_file`). Stops at the first error, the files that were written stay written.
/// `folder`, under its name (changed if that is the one of something else). Stops at the first
/// error, the files that were written stay written.
pub fn save(
&mut self,
engine: &mut Engine,
@@ -725,8 +725,7 @@ impl Workspace {
/// Writes the new files among `ids` in their folder, under their name. A name that is the one of
/// something else is changed (`name (2)`). Files that would be called the same, like the parts
/// of a split, are numbered from 1 (`name (1)`, `name (2)`). The files are called like the file
/// they are written to.
/// of a split, are numbered from 1 (`name (1)`, `name (2)`). The files keep their name.
pub fn save_new(&mut self, engine: &mut Engine, ids: &[Uuid]) -> Result<api::Outcome, String> {
let names: Vec<Option<(Uuid, PathBuf, String)>> = ids
.iter()
@@ -760,30 +759,6 @@ impl Workspace {
Ok(outcome)
}
/// The names of the files that are on the disk, and whether they are the files on the disk, to
/// tell which ones were renamed by an edit (see `follow_names`).
pub fn names(&self, engine: &Engine) -> BTreeMap<Uuid, (String, bool)> {
self.links
.keys()
.filter_map(|id| {
let name = engine.file_structure(&FileId(*id))?.name.clone();
Some((*id, (name, !self.is_dirty(engine, *id))))
})
.collect()
}
/// The name that a file can be given. A file on the disk is called like its file: the name is
/// one that a file can have (see `file_name`), and not the one of another file next to it
/// (`NAME_TAKEN`). The other files get theirs when they are written.
pub fn name_for(&self, id: Uuid, name: &str) -> Result<String, String> {
if !self.links.contains_key(&id) {
return Ok(name.to_owned());
}
let name = file_name(name);
self.renamed_path(id, &name)?;
Ok(name)
}
/// Where a file on the disk goes when it is called `name`: next to where it is, with the same
/// extension. `None` if it is there already, an error if something else is.
fn renamed_path(&self, id: Uuid, name: &str) -> Result<Option<PathBuf>, String> {
@@ -813,49 +788,24 @@ impl Workspace {
Ok(Some(path))
}
/// Renames the files on the disk whose name changed in the engine since it had `before` (see
/// `names`): an edit of the name, or its undo, renames the file. A file that was the one on the
/// disk is written too, so that it still is: the rename is not something to save. A file with
/// edits is not, its edits are not saved for it. A file that cannot be renamed keeps its place,
/// with the reason in the errors. Says whether a file was renamed.
pub fn follow_names(
&mut self,
engine: &mut Engine,
before: &BTreeMap<Uuid, (String, bool)>,
) -> (bool, Vec<String>) {
let mut renamed = false;
let mut errors = vec![];
for (id, (old, clean)) in before {
let Some(structure) = engine.file_structure(&FileId(*id)) else {
continue;
};
if structure.name == *old {
continue;
}
let path = match self.renamed_path(*id, &structure.name) {
Ok(Some(path)) => path,
Ok(None) => continue,
Err(error) => {
errors.push(error);
continue;
}
};
let link = self.links.get_mut(id).expect("the file is on the disk");
// a file that is gone is written at its new place when it is saved
if !link.missing
&& let Err(error) = fs::rename(&link.path, &path)
{
/// Renames the file on the disk of an open file, next to where it is: the name is one that a
/// file can have (see `file_name`), and not the one of another file (`NAME_TAKEN`). It is not
/// an edit, nothing is undone; the file has the same content, so it is as saved as it was.
/// Nothing happens to a file that is not on the disk. Says whether it was renamed.
pub fn rename_file(&mut self, id: Uuid, name: &str) -> Result<bool, String> {
let Some(path) = self.renamed_path(id, name)? else {
return Ok(false);
};
let link = self.links.get_mut(&id).expect("the file is on the disk");
// a file that is gone is written at its new place when it is saved
if !link.missing {
fs::rename(&link.path, &path).map_err(|error| {
eprintln!("{} cannot be renamed: {error}", link.path.display());
errors.push(RENAME_FAILED.to_owned());
continue;
}
link.path = path.clone();
renamed = true;
if *clean && let Err(error) = self.write_file(engine, *id, &path) {
errors.push(error);
}
RENAME_FAILED.to_owned()
})?;
}
(renamed, errors)
link.path = path;
Ok(true)
}
/// Writes a file to a path of the user's choice, which is the place of the file from then on.
@@ -873,30 +823,21 @@ impl Workspace {
self.write_file(engine, id, path)
}
/// Writes a file to `path`, which it is linked to from then on. It is called like the file it is
/// written to (a file has a single name), as if it always had been: that is not an edit to
/// undo. Says what changed in the engine.
/// Writes a file to `path`, which it is linked to from then on. Says what changed in the engine
/// (nothing: its name is its own, whatever the name of the file it is written to).
fn write_file(
&mut self,
engine: &mut Engine,
id: Uuid,
path: &Path,
) -> Result<api::Outcome, String> {
let Some(structure) = engine.file_structure(&FileId(id)) else {
return Ok(api::Outcome::failed_none());
};
let name = stem(path);
let renamed = if structure.name != name {
api::rename_file(engine, id, &name)
} else {
api::Outcome::failed_none()
};
let unchanged = api::Outcome::failed_none();
let Some(bytes) = api::export_file(engine, id, ExportOptions::ALL) else {
return Ok(renamed);
return Ok(unchanged);
};
write_atomically(path, &bytes).map_err(|error| format!("{}: {error}", path.display()))?;
let Some(signature) = signature(engine, id) else {
return Ok(renamed);
return Ok(unchanged);
};
// the folder, if there is one, may watch what was just written: the hash tells it is known
self.links.insert(
@@ -911,7 +852,7 @@ impl Workspace {
},
);
self.dirs.remove(&id);
Ok(renamed)
Ok(unchanged)
}
/// The folders and the GPX files that are in a folder of the open folder (or in it), the
@@ -1026,8 +967,8 @@ impl Workspace {
Ok(path)
}
/// Renames a file or a folder of the tree that is not an open file (which is renamed in the
/// engine, see `follow_names`). A file keeps its extension. Gives the new path.
/// Renames a file or a folder of the tree. A file keeps its extension, and an open file keeps its
/// name (see `rename_file`). Gives the new path.
pub fn rename_path(&mut self, path: &Path, name: &str) -> Result<PathBuf, String> {
let path = self.tree_item(path)?;
let name = file_name(name);
@@ -1099,812 +1040,3 @@ impl Workspace {
})
}
}
#[cfg(test)]
mod tests {
use gpx_engine::SelectMode;
use super::*;
fn gpx(name: &str) -> String {
format!(
r#"<?xml version="1.0"?><gpx version="1.1" creator="test" xmlns="http://www.topografix.com/GPX/1/1"><metadata><name>{name}</name></metadata><trk><name>{name}</name><trkseg><trkpt lat="50.0" lon="4.0"><ele>10</ele></trkpt><trkpt lat="50.1" lon="4.1"><ele>20</ele></trkpt></trkseg></trk></gpx>"#
)
}
/// The GPX files of a folder and of the folders in it.
pub(super) fn gpx_files(folder: &Path) -> Vec<PathBuf> {
let mut found = vec![];
for entry in fs::read_dir(folder).unwrap().flatten() {
let path = entry.path();
if path.is_dir() {
found.extend(gpx_files(&path));
} else if is_gpx(&path) {
found.push(path);
}
}
found.sort();
found
}
struct Fixture {
folder: PathBuf,
engine: Engine,
workspace: Workspace,
}
impl Fixture {
fn new() -> Self {
let folder =
canonical(&std::env::temp_dir()).join(format!("gpx-ws-{}", Uuid::new_v4()));
fs::create_dir_all(&folder).unwrap();
Self {
folder,
engine: Engine::default(),
workspace: Workspace::default(),
}
}
fn write(&self, name: &str, title: &str) -> PathBuf {
let path = self.folder.join(name);
fs::write(&path, gpx(title)).unwrap();
path
}
/// Opens the folder, and all the files that are in it.
fn open_folder(&mut self) {
let folder = self.folder.clone();
self.workspace.open_folder(&mut self.engine, &folder, false);
let files = gpx_files(&folder);
self.workspace.open_files(&mut self.engine, &files, false);
}
fn reconcile(&mut self, changed: Option<&[PathBuf]>) -> Option<api::Outcome> {
self.workspace.reconcile(&mut self.engine, changed)
}
fn save(&mut self, ids: &[Uuid]) {
self.workspace
.save(&mut self.engine, ids, &BTreeMap::new(), None)
.unwrap();
}
fn ids(&self) -> Vec<Uuid> {
self.engine.order().iter().map(|id| id.0).collect()
}
fn id_named(&self, name: &str) -> Uuid {
self.ids()
.into_iter()
.find(|id| self.name(*id) == name)
.unwrap()
}
fn name(&self, id: Uuid) -> String {
self.engine
.file_structure(&FileId(id))
.unwrap()
.name
.clone()
}
/// What the file holds: the name of its track (its own name is the one of the file).
fn content(&self, id: Uuid) -> String {
self.engine.file_structure(&FileId(id)).unwrap().tracks[0]
.name
.clone()
.unwrap_or_default()
}
fn rename(&mut self, id: Uuid, name: &str) {
api::select(&mut self.engine, &[id], SelectMode::Replace);
assert!(!api::metadata(&mut self.engine, name, "").has_error());
}
fn status(&self, id: Uuid) -> FileStatus {
self.workspace
.state(&self.engine)
.files
.remove(&id.to_string())
.unwrap()
}
}
impl Drop for Fixture {
fn drop(&mut self) {
let _ = fs::remove_dir_all(&self.folder);
}
}
#[test]
fn an_opened_file_is_clean_until_it_is_edited_and_clean_again_once_saved() {
let mut fx = Fixture::new();
let path = fx.write("a.gpx", "a");
fx.workspace
.open_files(&mut fx.engine, std::slice::from_ref(&path), true);
let id = fx.ids()[0];
assert!(!fx.workspace.is_dirty(&fx.engine, id));
assert_eq!(fx.status(id).path.as_deref(), path.to_str());
fx.rename(id, "renamed");
assert!(fx.workspace.is_dirty(&fx.engine, id));
// nothing is written before it is saved
assert!(fs::read_to_string(&path).unwrap().contains(">a<"));
fx.save(&[id]);
assert!(!fx.workspace.is_dirty(&fx.engine, id));
assert!(fs::read_to_string(&path).unwrap().contains(">renamed<"));
// going back before the save is an edit again
api::select(&mut fx.engine, &[id], SelectMode::Replace);
api::undo(&mut fx.engine);
assert!(fx.workspace.is_dirty(&fx.engine, id));
api::redo(&mut fx.engine);
assert!(!fx.workspace.is_dirty(&fx.engine, id));
}
#[test]
fn a_file_that_is_not_on_the_disk_is_unsaved_until_it_is_given_a_place() {
let mut fx = Fixture::new();
api::new_file(&mut fx.engine, "fresh", None, None, None);
let id = fx.ids()[0];
assert!(fx.workspace.is_dirty(&fx.engine, id));
assert_eq!(
fx.workspace.unlinked(&fx.engine, &[id]),
[(id, "fresh".to_string())]
);
// nowhere to put it without being told
fx.save(&[id]);
assert!(fx.workspace.is_dirty(&fx.engine, id));
let path = fx.folder.join("fresh.gpx");
let paths = BTreeMap::from([(id, path.clone())]);
fx.workspace
.save(&mut fx.engine, &[id], &paths, None)
.unwrap();
assert!(path.exists());
assert!(!fx.workspace.is_dirty(&fx.engine, id));
assert!(fx.workspace.unlinked(&fx.engine, &[id]).is_empty());
}
#[test]
fn several_files_without_a_place_go_in_a_folder_with_names_that_are_free() {
let mut fx = Fixture::new();
fx.write("fresh.gpx", "already there");
api::new_file(&mut fx.engine, "fresh", None, None, None);
api::new_file(&mut fx.engine, "a/b: c", None, None, None);
api::new_file(&mut fx.engine, "fresh", None, None, None);
let ids = fx.ids();
let folder = fx.folder.clone();
fx.workspace
.save(&mut fx.engine, &ids, &BTreeMap::new(), Some(&folder))
.unwrap();
for name in ["fresh (2).gpx", "fresh (3).gpx", "a_b_ c.gpx"] {
assert!(fx.folder.join(name).exists(), "{name}");
}
assert!(ids.iter().all(|id| !fx.workspace.is_dirty(&fx.engine, *id)));
}
#[test]
fn what_this_app_writes_is_not_a_change_of_the_disk() {
let mut fx = Fixture::new();
let path = fx.write("a.gpx", "a");
fx.open_folder();
let id = fx.ids()[0];
api::select(&mut fx.engine, &[id], SelectMode::Replace);
assert!(!api::style(&mut fx.engine, Some("#ff0000"), None, None).has_error());
fx.save(&[id]);
assert!(fx.reconcile(Some(&[path])).is_none());
assert!(!fx.status(id).replaced && !fx.status(id).dirty);
}
#[test]
fn the_folders_of_the_files_outside_the_open_folder_are_watched_too() {
let mut fx = Fixture::new();
let inside = fx.write("in.gpx", "in");
let elsewhere = canonical(&std::env::temp_dir()).join(format!("gpx-ws-{}", Uuid::new_v4()));
fs::create_dir_all(&elsewhere).unwrap();
let (one, two) = (elsewhere.join("one.gpx"), elsewhere.join("two.gpx"));
fs::write(&one, gpx("one")).unwrap();
fs::write(&two, gpx("two")).unwrap();
let folder = fx.folder.clone();
// no folder: only the folder of the files
fx.workspace.open_files(&mut fx.engine, &[one, two], false);
assert_eq!(fx.workspace.watched(), [(elsewhere.clone(), false)]);
// the open folder as a whole, and the file in it needs nothing more
fx.workspace.open_folder(&mut fx.engine, &folder, false);
fx.workspace.open_files(&mut fx.engine, &[inside], false);
assert_eq!(
fx.workspace.watched(),
[(folder, true), (elsewhere.clone(), false)]
);
// closing the files stops it
let ids = fx.ids();
fx.workspace.close_files(&mut fx.engine, &ids, true);
assert_eq!(fx.workspace.watched(), [(fx.folder.clone(), true)]);
fs::remove_dir_all(elsewhere).unwrap();
}
#[test]
fn opening_a_folder_does_not_open_its_files() {
let mut fx = Fixture::new();
fx.write("a.gpx", "a");
let folder = fx.folder.clone();
let outcome = fx.workspace.open_folder(&mut fx.engine, &folder, false);
assert!(!outcome.changed);
assert!(fx.ids().is_empty());
assert_eq!(fx.workspace.folder.as_deref(), Some(folder.as_path()));
}
#[test]
fn folders_are_made_renamed_and_moved_and_the_open_files_follow() {
let mut fx = Fixture::new();
let a = fx.write("a.gpx", "a");
fx.open_folder();
let id = fx.ids()[0];
let root = fx.folder.clone();
let path_of = |fx: &Fixture| PathBuf::from(fx.status(id).path.unwrap());
let first = fx.workspace.create_folder(&root, "New folder").unwrap();
let second = fx.workspace.create_folder(&root, "New folder").unwrap();
assert_eq!(first, root.join("New folder"));
assert_eq!(second, root.join("New folder 2"));
fx.workspace.move_paths(&[a], &first).unwrap();
assert_eq!(path_of(&fx), first.join("a.gpx"));
assert!(first.join("a.gpx").exists() && !fx.status(id).dirty);
let trips = fx.workspace.rename_path(&first, "Trips").unwrap();
assert_eq!(path_of(&fx), root.join("Trips/a.gpx"));
// a folder does not go into itself, nor onto something else
assert!(
fx.workspace
.move_paths(std::slice::from_ref(&trips), &trips)
.is_err()
);
assert_eq!(
fx.workspace.rename_path(&trips, "New folder 2"),
Err(NAME_TAKEN.to_owned())
);
fx.workspace.move_paths(&[trips], &second).unwrap();
assert_eq!(path_of(&fx), second.join("Trips/a.gpx"));
// a file keeps its extension
let closed = fx.write("c.gpx", "c");
assert_eq!(
fx.workspace.rename_path(&closed, "d").unwrap(),
root.join("d.gpx")
);
// the open folder and what is not in it are left alone
assert!(fx.workspace.rename_path(&root, "x").is_err());
let elsewhere = canonical(&std::env::temp_dir());
assert!(fx.workspace.rename_path(&elsewhere, "x").is_err());
assert!(fx.workspace.create_folder(&elsewhere, "x").is_err());
}
#[test]
fn deleted_files_are_closed_unless_they_have_edits() {
let mut fx = Fixture::new();
let a = fx.write("a.gpx", "a");
let b = fx.write("b.gpx", "b");
fx.open_folder();
let (id_a, id_b) = (fx.id_named("a"), fx.id_named("b"));
api::select(&mut fx.engine, &[id_b], SelectMode::Replace);
assert!(!api::style(&mut fx.engine, Some("#ff0000"), None, None).has_error());
let outcome = fx
.workspace
.delete_paths(&mut fx.engine, &[a.clone(), b.clone()], |paths| {
paths.iter().for_each(|path| fs::remove_file(path).unwrap());
Ok(())
})
.unwrap();
assert!(outcome.changed);
assert_eq!(fx.ids(), [id_b]);
assert!(!fx.ids().contains(&id_a));
let status = fx.status(id_b);
assert!(status.dirty && status.missing);
assert!(!a.exists() && !b.exists());
}
#[test]
fn the_tree_of_the_folder_shows_folders_first_and_only_gpx_files() {
let mut fx = Fixture::new();
fx.write("b.gpx", "b");
fx.write("A.GPX", "a");
fx.write("b 10.gpx", "b 10");
fx.write("b 2.gpx", "b 2");
fx.write("notes.txt", "not a track");
fx.write(".hidden.gpx", "hidden");
fs::create_dir(fx.folder.join("zoo")).unwrap();
fs::create_dir(fx.folder.join("Alps")).unwrap();
fs::create_dir(fx.folder.join(".git")).unwrap();
fx.write("zoo/c.gpx", "c");
let folder = fx.folder.clone();
fx.workspace.open_folder(&mut fx.engine, &folder, false);
let names = |entries: Vec<Entry>| -> Vec<(String, bool)> {
entries.into_iter().map(|e| (e.name, e.is_dir)).collect()
};
assert_eq!(
names(fx.workspace.list_folder(&folder).unwrap()),
[
("Alps".to_string(), true),
("zoo".to_string(), true),
("A.GPX".to_string(), false),
("b.gpx".to_string(), false),
("b 2.gpx".to_string(), false),
("b 10.gpx".to_string(), false),
]
);
assert_eq!(
names(fx.workspace.list_folder(&folder.join("zoo")).unwrap()),
[("c.gpx".to_string(), false)]
);
// only what is in the open folder can be listed
let elsewhere = canonical(&std::env::temp_dir());
assert!(fx.workspace.list_folder(&elsewhere).is_err());
}
#[test]
fn names_are_sorted_by_the_numbers_they_hold() {
let mut names = vec![
"track 10",
"track 2",
"Track 1",
"track",
"track 02b",
"track 2a",
"b",
"A",
];
names.sort_by(|a, b| natural_cmp(a, b).then_with(|| a.cmp(b)));
assert_eq!(
names,
[
"A",
"b",
"track",
"Track 1",
"track 2",
"track 2a",
"track 02b",
"track 10",
]
);
}
#[test]
fn a_file_without_edits_is_read_again_when_the_disk_changes() {
let mut fx = Fixture::new();
let a = fx.write("a.gpx", "a");
fx.open_folder();
let id = fx.ids()[0];
fx.write("a.gpx", "theirs");
let outcome = fx.reconcile(Some(std::slice::from_ref(&a))).unwrap();
assert!(outcome.changed);
// the same file, now the one on the disk
assert_eq!(fx.ids(), [id]);
assert_eq!(fx.content(id), "theirs");
let status = fx.status(id);
assert!(!status.dirty && !status.replaced);
// looking again changes nothing
assert!(fx.reconcile(Some(std::slice::from_ref(&a))).is_none());
}
#[test]
fn a_file_with_edits_is_flagged_when_the_disk_changes_and_not_read_again() {
let mut fx = Fixture::new();
let a = fx.write("a.gpx", "a");
fx.open_folder();
let id = fx.ids()[0];
fx.rename(id, "mine");
fx.write("a.gpx", "theirs");
let outcome = fx.reconcile(Some(std::slice::from_ref(&a))).unwrap();
assert!(!outcome.changed);
assert_eq!(fx.name(id), "mine");
let status = fx.status(id);
assert!(status.dirty && status.replaced && !status.missing);
// looking again changes nothing
assert!(fx.reconcile(Some(std::slice::from_ref(&a))).is_none());
// the disk goes back to what was read: only the edits are left
fx.write("a.gpx", "a");
assert!(fx.reconcile(Some(std::slice::from_ref(&a))).is_some());
let status = fx.status(id);
assert!(status.dirty && !status.replaced);
}
#[test]
fn the_user_can_take_the_disk_instead_of_the_edits() {
let mut fx = Fixture::new();
fx.write("a.gpx", "a");
fx.open_folder();
let id = fx.ids()[0];
fx.rename(id, "mine");
fx.write("a.gpx", "theirs");
fx.reconcile(None);
assert!(fx.status(id).replaced);
fx.workspace.reload(&mut fx.engine, &[id]).unwrap();
assert_eq!(fx.content(id), "theirs");
let status = fx.status(id);
assert!(!status.dirty && !status.replaced);
// and it is one undo step
api::select(&mut fx.engine, &[id], SelectMode::Replace);
api::undo(&mut fx.engine);
assert_eq!(fx.content(id), "mine");
}
#[test]
fn saving_overwrites_what_replaced_the_file() {
let mut fx = Fixture::new();
let a = fx.write("a.gpx", "a");
fx.open_folder();
let id = fx.ids()[0];
fx.rename(id, "mine");
fx.write("a.gpx", "theirs");
fx.reconcile(None);
assert!(fx.status(id).replaced);
fx.save(&[id]);
assert!(fs::read_to_string(a).unwrap().contains(">mine<"));
assert!(!fx.status(id).replaced && !fx.status(id).dirty);
}
#[test]
fn a_file_that_is_gone_from_the_disk_stays_open_and_dirty_until_it_is_saved_again() {
let mut fx = Fixture::new();
let a = fx.write("a.gpx", "a");
fx.open_folder();
let id = fx.ids()[0];
fs::remove_file(&a).unwrap();
assert!(fx.reconcile(Some(std::slice::from_ref(&a))).is_some());
assert_eq!(fx.ids(), [id]);
let status = fx.status(id);
assert!(status.dirty && status.missing && status.path.is_some());
// saving puts it back
fx.save(&[id]);
assert!(a.exists());
let status = fx.status(id);
assert!(!status.dirty && !status.missing);
}
#[test]
fn a_file_that_comes_back_different_is_read_again() {
let mut fx = Fixture::new();
let a = fx.write("a.gpx", "a");
fx.open_folder();
let id = fx.ids()[0];
fs::remove_file(&a).unwrap();
fx.reconcile(None);
assert!(fx.status(id).missing);
fx.write("a.gpx", "back");
fx.reconcile(None);
assert_eq!(fx.content(id), "back");
let status = fx.status(id);
assert!(!status.dirty && !status.missing);
}
#[test]
fn what_changed_while_the_app_was_closed_is_found_when_it_starts() {
let mut fx = Fixture::new();
let a = fx.write("a.gpx", "a");
fx.open_folder();
let id = fx.ids()[0];
fx.write("a.gpx", "changed");
assert!(fx.reconcile(None).is_some());
assert_eq!(fx.content(id), "changed");
fs::remove_file(a).unwrap();
assert!(fx.reconcile(None).is_some());
assert!(fx.status(id).missing);
}
#[test]
fn new_files_made_from_a_file_of_the_folder_are_saved_next_to_it() {
let mut fx = Fixture::new();
fs::create_dir(fx.folder.join("sub")).unwrap();
fx.write("sub/b.gpx", "b");
fx.write("a.gpx", "a");
fx.open_folder();
let b = fx.id_named("b");
let sub = fx.folder.join("sub");
assert_eq!(fx.workspace.source_dir(&[b]), Some(canonical(&sub)));
let before = fx.ids();
api::select(&mut fx.engine, &[b], SelectMode::Replace);
let outcome = api::duplicate(&mut fx.engine);
let copy = fx
.ids()
.into_iter()
.find(|id| !before.contains(id))
.unwrap();
fx.workspace.inherit(outcome.added(), &canonical(&sub));
assert_eq!(fx.status(copy).folder.as_deref(), canonical(&sub).to_str());
assert!(fx.status(copy).path.is_none() && fx.status(copy).dirty);
// nobody needs to be asked where it goes
assert!(fx.workspace.unlinked(&fx.engine, &[copy]).is_empty());
fx.save(&[copy]);
assert!(sub.join("b (2).gpx").exists());
let status = fx.status(copy);
assert!(!status.dirty && status.folder.is_none());
assert_eq!(
status.path.as_deref(),
canonical(&sub.join("b (2).gpx")).to_str()
);
// the copy of the copy is next to it as well
assert_eq!(fx.workspace.source_dir(&[copy]), Some(canonical(&sub)));
}
#[test]
fn a_file_outside_the_open_folder_hands_nothing_down() {
let mut fx = Fixture::new();
let elsewhere = canonical(&std::env::temp_dir()).join(format!("gpx-ws-{}", Uuid::new_v4()));
fs::create_dir_all(&elsewhere).unwrap();
let outside = elsewhere.join("o.gpx");
fs::write(&outside, gpx("o")).unwrap();
let folder = fx.folder.clone();
fx.workspace.open_folder(&mut fx.engine, &folder, false);
fx.workspace
.open_files(&mut fx.engine, std::slice::from_ref(&outside), true);
let id = fx.ids()[0];
assert_eq!(fx.workspace.source_dir(&[id]), None);
// a new file with no source is not linked to a folder by `source_dir`: the caller falls back
fs::remove_dir_all(elsewhere).unwrap();
}
#[test]
fn closing_files_asks_what_to_do_with_the_unsaved_ones() {
let mut fx = Fixture::new();
fx.write("a.gpx", "a");
fx.write("b.gpx", "b");
fx.open_folder();
let (a, b) = (fx.id_named("a"), fx.id_named("b"));
fx.rename(a, "edited");
assert_eq!(fx.workspace.unsaved(&fx.engine, &[a, b]), [a]);
// without a choice what has edits stays
let outcome = fx.workspace.close_files(&mut fx.engine, &[a, b], false);
assert!(outcome.changed);
assert_eq!(fx.ids(), [a]);
// saved first
fx.save(&[a]);
fx.workspace.close_files(&mut fx.engine, &[a], false);
assert!(fx.ids().is_empty());
assert!(
fs::read_to_string(fx.folder.join("a.gpx"))
.unwrap()
.contains("edited")
);
assert!(fx.workspace.links.is_empty());
}
#[test]
fn closing_files_can_discard_their_edits() {
let mut fx = Fixture::new();
fx.write("a.gpx", "a");
fx.open_folder();
let a = fx.ids()[0];
fx.rename(a, "edited");
api::new_file(&mut fx.engine, "fresh", None, None, None);
let fresh = fx.id_named("fresh");
fx.workspace.close_files(&mut fx.engine, &[a, fresh], true);
assert!(fx.ids().is_empty());
assert!(
fs::read_to_string(fx.folder.join("a.gpx"))
.unwrap()
.contains(">a<")
);
}
#[test]
fn closing_the_folder_closes_every_file() {
let mut fx = Fixture::new();
fx.write("a.gpx", "a");
fx.open_folder();
api::new_file(&mut fx.engine, "fresh", None, None, None);
assert_eq!(fx.ids().len(), 2);
// the new file has nowhere to go: it has to be discarded
let outcome = fx.workspace.close_folder(&mut fx.engine, false);
assert!(outcome.changed);
assert_eq!(fx.ids().len(), 1);
assert!(fx.workspace.folder.is_none());
let outcome = fx.workspace.close_folder(&mut fx.engine, true);
assert!(outcome.changed);
assert!(fx.ids().is_empty());
}
#[test]
fn opening_another_folder_closes_the_first_with_all_its_files() {
let mut fx = Fixture::new();
fx.write("a.gpx", "a");
fx.open_folder();
let a = fx.ids()[0];
fx.rename(a, "edited");
let other = canonical(&std::env::temp_dir()).join(format!("gpx-ws-{}", Uuid::new_v4()));
fs::create_dir_all(&other).unwrap();
// the edits stay, nothing is lost without a choice
fx.workspace.open_folder(&mut fx.engine, &other, false);
assert_eq!(fx.ids(), [a]);
assert_eq!(fx.workspace.folder.as_deref(), Some(other.as_path()));
let first = fx.folder.clone();
fx.workspace.open_folder(&mut fx.engine, &first, true);
assert!(fx.ids().is_empty());
assert!(
fs::read_to_string(fx.folder.join("a.gpx"))
.unwrap()
.contains(">a<")
);
fs::remove_dir_all(other).unwrap();
}
#[test]
fn opening_the_same_folder_again_changes_nothing() {
let mut fx = Fixture::new();
fx.write("a.gpx", "a");
fx.open_folder();
let folder = fx.folder.clone();
let outcome = fx.workspace.open_folder(&mut fx.engine, &folder, true);
assert!(!outcome.changed);
assert_eq!(fx.ids().len(), 1);
}
#[test]
fn a_failed_save_stops_at_the_file_and_leaves_it_dirty() {
let mut fx = Fixture::new();
fx.write("a.gpx", "a");
fx.open_folder();
let id = fx.ids()[0];
fx.rename(id, "edited");
// the place of the file is taken by a folder
fs::remove_file(fx.folder.join("a.gpx")).unwrap();
fs::create_dir(fx.folder.join("a.gpx")).unwrap();
let result = fx
.workspace
.save(&mut fx.engine, &[id], &BTreeMap::new(), None);
assert!(result.is_err());
assert!(fx.workspace.is_dirty(&fx.engine, id));
}
#[test]
fn saving_as_gives_the_file_another_place() {
let mut fx = Fixture::new();
let a = fx.write("a.gpx", "a");
fx.write("b.gpx", "b");
fx.open_folder();
let id = fx.id_named("a");
let other = fx.id_named("b");
fx.rename(id, "edited");
let target = fx.folder.join("copy.gpx");
fx.workspace.save_as(&mut fx.engine, id, &target).unwrap();
assert!(fs::read_to_string(&target).unwrap().contains("edited"));
assert!(fs::read_to_string(a).unwrap().contains(">a<"));
let status = fx.status(id);
assert!(!status.dirty);
assert_eq!(status.path.as_deref(), canonical(&target).to_str());
// the place of another open file is not available
let taken = fx.folder.join("b.gpx");
assert!(fx.workspace.save_as(&mut fx.engine, id, &taken).is_err());
assert!(!fx.workspace.is_dirty(&fx.engine, other));
}
}
#[cfg(test)]
mod restart_tests {
use gpx_engine::Persister;
use super::*;
use crate::store::FileStorage;
/// What the app does when it starts again: the engine and the links are read from what was
/// kept.
fn restart(engine: &Engine, workspace: &Workspace, root: &Path) -> (Engine, Workspace) {
let storage = FileStorage::open(root).unwrap();
pollster::block_on(Persister::default().save(&storage, engine.snapshot())).unwrap();
let mut restarted = Engine::default();
let restored = pollster::block_on(Persister::default().restore(&storage)).unwrap();
restarted.restore(restored);
let links = serde_json::to_string(workspace).unwrap();
(restarted, serde_json::from_str(&links).unwrap())
}
fn setup() -> (PathBuf, Engine, Workspace, Vec<Uuid>) {
let root = std::env::temp_dir().join(format!("gpx-restart-{}", Uuid::new_v4()));
let folder = root.join("tracks");
fs::create_dir_all(&folder).unwrap();
for name in ["with_tracks", "with_time", "with_waypoint"] {
let _ = fs::copy(
format!("../gpx-rs/data/{name}.gpx"),
folder.join(format!("{name}.gpx")),
);
}
let mut engine = Engine::default();
let mut workspace = Workspace::default();
workspace.open_folder(&mut engine, &folder, false);
workspace.open_files(&mut engine, &tests::gpx_files(&folder), false);
let ids = engine.order().iter().map(|id| id.0).collect();
(root, engine, workspace, ids)
}
#[test]
fn files_that_were_not_edited_are_not_unsaved_after_a_restart() {
let (root, engine, workspace, ids) = setup();
assert!(ids.len() >= 2);
let (restarted, mut workspace) = restart(&engine, &workspace, &root.join("store"));
// the files are the same, but the engine knows their contents by new identities
assert!(ids.iter().any(|id| workspace.is_dirty(&restarted, *id)));
workspace.verify(&restarted);
assert!(ids.iter().all(|id| !workspace.is_dirty(&restarted, *id)));
fs::remove_dir_all(root).unwrap();
}
#[test]
fn files_that_were_edited_are_still_unsaved_after_a_restart() {
let (root, mut engine, workspace, ids) = setup();
api::select(&mut engine, &[ids[0]], gpx_engine::SelectMode::Replace);
api::metadata(&mut engine, "edited", "");
assert!(workspace.is_dirty(&engine, ids[0]));
let (restarted, mut workspace) = restart(&engine, &workspace, &root.join("store"));
workspace.verify(&restarted);
assert!(workspace.is_dirty(&restarted, ids[0]));
assert!(
ids[1..]
.iter()
.all(|id| !workspace.is_dirty(&restarted, *id))
);
fs::remove_dir_all(root).unwrap();
}
#[test]
fn links_made_before_the_content_was_kept_are_checked_against_the_disk() {
let (root, mut engine, mut workspace, ids) = setup();
api::select(&mut engine, &[ids[0]], gpx_engine::SelectMode::Replace);
api::metadata(&mut engine, "edited", "");
for link in workspace.links.values_mut() {
link.export_hash = 0;
}
let (restarted, mut workspace) = restart(&engine, &workspace, &root.join("store"));
workspace.verify(&restarted);
assert!(workspace.is_dirty(&restarted, ids[0]));
assert!(
ids[1..]
.iter()
.all(|id| !workspace.is_dirty(&restarted, *id))
);
fs::remove_dir_all(root).unwrap();
}
#[test]
fn the_folders_of_new_files_and_the_old_flag_name_survive_a_restart() {
let (root, engine, mut workspace, ids) = setup();
let folder = workspace.folder.clone().unwrap();
let fresh = Uuid::new_v4();
workspace.inherit(&[fresh.to_string()], &folder);
workspace.links.get_mut(&ids[0]).unwrap().replaced = true;
let json = serde_json::to_string(&workspace).unwrap();
// links written by the previous version had `conflict`, which is the same thing
let json = json.replace("\"replaced\"", "\"conflict\"");
assert!(json.contains("\"conflict\":true"));
let read: Workspace = serde_json::from_str(&json).unwrap();
assert!(read.links[&ids[0]].replaced);
assert_eq!(read.dirs.get(&fresh), Some(&folder));
let _ = engine;
fs::remove_dir_all(root).unwrap();
}
}
@@ -6,14 +6,23 @@ use tauri::test::{INVOKE_KEY, get_ipc_response, mock_builder, mock_context, noop
use tauri::webview::InvokeRequest;
use tauri::{WebviewWindow, ipc::CallbackFn, ipc::InvokeBody, ipc::InvokeResponseBody};
/// The origin of the pages of the app, the one Tauri allows the commands from: Windows has no
/// custom protocols, the pages come from `http://tauri.localhost` there.
const ORIGIN: &str = if cfg!(windows) {
"http://tauri.localhost"
} else {
"tauri://localhost"
};
struct App(WebviewWindow<tauri::test::MockRuntime>);
impl App {
fn new() -> Self {
let app =
crate::register_commands(mock_builder().manage(crate::state::AppState::temporary()))
.build(mock_context(noop_assets()))
.unwrap();
let app = gpx_studio_desktop_lib::register_commands(
mock_builder().manage(crate::temporary_state()),
)
.build(mock_context(noop_assets()))
.unwrap();
let window = tauri::WebviewWindowBuilder::new(&app, "main", Default::default())
.build()
.unwrap();
@@ -34,7 +43,7 @@ impl App {
cmd: cmd.into(),
callback: CallbackFn(0),
error: CallbackFn(1),
url: "tauri://localhost".parse().unwrap(),
url: ORIGIN.parse().unwrap(),
body,
headers,
invoke_key: INVOKE_KEY.to_string(),
@@ -50,6 +59,20 @@ impl App {
}
}
/// The error of a command that fails.
fn error(&self, cmd: &str, args: Value) -> Value {
let request = InvokeRequest {
cmd: cmd.into(),
callback: CallbackFn(0),
error: CallbackFn(1),
url: ORIGIN.parse().unwrap(),
body: InvokeBody::Json(args),
headers: Default::default(),
invoke_key: INVOKE_KEY.to_string(),
};
get_ipc_response(&self.0, request).expect_err(&format!("{cmd} did not fail"))
}
fn bytes(&self, cmd: &str, args: Value) -> Vec<u8> {
match self.request(cmd, InvokeBody::Json(args), Default::default()) {
InvokeResponseBody::Raw(bytes) => bytes,
@@ -257,7 +280,7 @@ fn waypoints_exports_and_routing_revisions() {
let exported = app.bytes(
"export_file",
json!({ "fileId": file, "time": true, "hr": true, "cad": true, "atemp": true, "power": true, "osm": true, "asRoute": false }),
json!({ "fileId": file, "time": true, "hr": true, "cad": true, "atemp": true, "power": true, "osm": true }),
);
assert_eq!(exported[0], 1);
assert!(
@@ -297,8 +320,8 @@ fn keeps_settings_for_the_session() {
#[test]
fn the_files_are_still_there_in_the_next_session() {
use crate::state::AppState;
use gpx_engine::api;
use gpx_studio_desktop_lib::state::AppState;
let directory =
std::env::temp_dir().join(format!("gpx-studio-session-{}", uuid::Uuid::new_v4()));
@@ -333,14 +356,14 @@ fn only_gpx_files_that_exist_are_taken_from_what_the_system_gives() {
text.to_string_lossy().into_owned(),
directory.join("missing.gpx").to_string_lossy().into_owned(),
];
assert_eq!(crate::gpx_paths(arguments), [track]);
assert_eq!(gpx_studio_desktop_lib::gpx_paths(arguments), [track]);
std::fs::remove_dir_all(directory).unwrap();
}
#[test]
fn a_file_opened_with_the_app_is_opened_selected_and_announced() {
use crate::state::AppState;
use gpx_engine::api;
use gpx_studio_desktop_lib::state::AppState;
use std::sync::{Arc, Mutex};
let directory = std::env::temp_dir().join(format!("gpx-studio-open-{}", uuid::Uuid::new_v4()));
@@ -427,7 +450,7 @@ fn a_file_outside_the_open_folder_is_read_again_when_the_disk_changes() {
let (events, received) = mpsc::channel();
let directory = std::env::temp_dir().join(format!("gpx-store-{}", uuid::Uuid::new_v4()));
let state = crate::state::AppState::open(
let state = gpx_studio_desktop_lib::state::AppState::open(
directory.clone(),
std::sync::Arc::new(move |name, _| {
let _ = events.send(name.to_string());
@@ -447,8 +470,8 @@ fn a_file_outside_the_open_folder_is_read_again_when_the_disk_changes() {
});
let id = uuid::Uuid::parse_str(&id).unwrap();
// the name comes from the path, which does not change: the content does
let signature = |state: &crate::state::AppState| {
state.core(move |core| crate::workspace::signature(&core.engine, id))
let signature = |state: &gpx_studio_desktop_lib::state::AppState| {
state.core(move |core| gpx_studio_desktop_lib::workspace::signature(&core.engine, id))
};
let before = signature(&state);
@@ -469,7 +492,7 @@ fn a_file_outside_the_open_folder_is_read_again_when_the_disk_changes() {
#[test]
fn files_made_from_a_file_of_the_open_folder_are_written_next_to_it_at_once() {
let state = crate::state::AppState::temporary();
let state = crate::temporary_state();
let folder = std::env::temp_dir().join(format!("gpx-tests-{}", uuid::Uuid::new_v4()));
std::fs::create_dir_all(folder.join("sub")).unwrap();
std::fs::copy("../gpx-rs/data/simple.gpx", folder.join("sub/track.gpx")).unwrap();
@@ -483,7 +506,7 @@ fn files_made_from_a_file_of_the_open_folder_are_written_next_to_it_at_once() {
assert_eq!(opened.added().len(), 1);
let copy = core.creating(None, false, gpx_engine::api::duplicate);
let fresh = core.creating(None, true, |e| {
gpx_engine::api::new_file(e, "fresh", None, None, None)
gpx_engine::api::new_file(e, "fresh", None, None, None, false)
});
// a copy of the file goes next to it, a new file with nothing selected in the selection
// of the file that was just made
@@ -500,21 +523,25 @@ fn files_made_from_a_file_of_the_open_folder_are_written_next_to_it_at_once() {
.values()
.map(|file| file.path.clone().unwrap())
.collect();
// the copy is in the subfolder, named after the file
// the copy is in the subfolder, named after its name (the one in the file, not of the file)
assert!(
paths.iter().any(|path| path.ends_with("sub/track (2).gpx")),
paths
.iter()
.any(|path| std::path::Path::new(path).ends_with("sub/simple.gpx")),
"{paths:?}"
);
// the new file went where the selected file is
assert!(
paths.iter().any(|path| path.ends_with("sub/fresh.gpx")),
paths
.iter()
.any(|path| std::path::Path::new(path).ends_with("sub/fresh.gpx")),
"{paths:?}"
);
}
#[test]
fn the_parts_of_a_split_are_numbered_from_one() {
let state = crate::state::AppState::temporary();
let state = crate::temporary_state();
let folder = std::env::temp_dir().join(format!("gpx-tests-{}", uuid::Uuid::new_v4()));
std::fs::create_dir_all(&folder).unwrap();
std::fs::copy("../gpx-rs/data/simple.gpx", folder.join("track.gpx")).unwrap();
@@ -536,8 +563,8 @@ fn the_parts_of_a_split_are_numbered_from_one() {
.map(|entry| entry.file_name().to_string_lossy().into_owned())
.collect();
names.sort();
// the file stays, its two parts are numbered from 1
assert_eq!(names, ["track (1).gpx", "track (2).gpx", "track.gpx"]);
// the file stays, its two parts are numbered from 1, under their name (the one in the file)
assert_eq!(names, ["simple (1).gpx", "simple (2).gpx", "track.gpx"]);
std::fs::remove_dir_all(folder).unwrap();
}
@@ -581,8 +608,18 @@ fn name_of(app: &App, id: &str) -> String {
.to_string()
}
/// The name of the file on the disk of an open file, without its extension.
fn file_stem_of(app: &App, id: &str) -> String {
let path = app.call("workspace_state", json!({}))["files"][id]["path"].clone();
std::path::Path::new(path.as_str().unwrap())
.file_stem()
.unwrap()
.to_string_lossy()
.into_owned()
}
#[test]
fn a_file_is_called_like_its_file_and_renaming_it_renames_the_file() {
fn a_file_keeps_its_name_and_its_file_is_renamed_apart() {
let app = App::new();
let folder = folder_with(&["outside", "taken"]);
let outcome = app.call(
@@ -590,76 +627,62 @@ fn a_file_is_called_like_its_file_and_renaming_it_renames_the_file() {
json!({ "paths": [folder.join("outside.gpx")] }),
);
let id = outcome["added"][0].as_str().unwrap().to_string();
let dirty = |app: &App| app.call("workspace_state", json!({}))["files"][&id]["dirty"].clone();
let inside = |name: &str| std::fs::read_to_string(folder.join(format!("{name}.gpx"))).unwrap();
// the name in the file is not the name of the file: the file wins, and nothing is unsaved
assert_eq!(name_of(&app, &id), "outside");
assert_eq!(
app.call("workspace_state", json!({}))["files"][&id]["dirty"],
false
);
// the name in the file is its name, not the one of the file
assert_eq!(name_of(&app, &id), "Inside");
assert_eq!(file_stem_of(&app, &id), "outside");
assert_eq!(dirty(&app), false);
// renaming renames the file on the disk
// renaming it is an edit of the file only, which is undone
app.call("select", json!({ "fileIds": [id], "mode": "replace" }));
let outcome = app.call("metadata", json!({ "name": "renamed", "desc": "" }));
assert!(outcome.get("error").is_none());
assert!(folder.join("renamed.gpx").exists());
assert!(!folder.join("outside.gpx").exists());
let path = app.call("workspace_state", json!({}))["files"][&id]["path"].clone();
assert!(path.as_str().unwrap().ends_with("renamed.gpx"));
// the file had nothing unsaved: it is written with its new name, and still has nothing unsaved
let dirty = |app: &App| app.call("workspace_state", json!({}))["files"][&id]["dirty"].clone();
let inside = |name: &str| std::fs::read_to_string(folder.join(format!("{name}.gpx"))).unwrap();
assert_eq!(dirty(&app), false);
assert!(inside("renamed").contains("<name>renamed</name>"));
// undone and redone, on the disk too, still with nothing unsaved
assert_eq!(name_of(&app, &id), "renamed");
assert_eq!(file_stem_of(&app, &id), "outside");
assert_eq!(dirty(&app), true);
app.call("undo", json!({}));
assert_eq!(name_of(&app, &id), "outside");
assert!(folder.join("outside.gpx").exists() && !folder.join("renamed.gpx").exists());
assert_eq!(dirty(&app), false);
assert!(inside("outside").contains("<name>outside</name>"));
app.call("redo", json!({}));
assert_eq!(name_of(&app, &id), "renamed");
assert!(folder.join("renamed.gpx").exists() && !folder.join("outside.gpx").exists());
assert_eq!(name_of(&app, &id), "Inside");
assert_eq!(dirty(&app), false);
// the name of another file is not taken, nothing changes
let outcome = app.call("metadata", json!({ "name": "taken", "desc": "" }));
assert_eq!(outcome["error"], crate::workspace::NAME_TAKEN);
assert_eq!(name_of(&app, &id), "renamed");
assert!(folder.join("renamed.gpx").exists());
// the file on the disk is renamed apart: not an edit, the file is as saved as it was
let renamed = app.call("rename_file", json!({ "fileId": id, "name": "moved" }));
assert_eq!(renamed, true);
assert!(folder.join("moved.gpx").exists() && !folder.join("outside.gpx").exists());
assert_eq!(file_stem_of(&app, &id), "moved");
assert_eq!(name_of(&app, &id), "Inside");
assert_eq!(dirty(&app), false);
assert_eq!(app.call("can_undo", json!({})), false);
// saving writes to the new place, under the name of the file
app.call("metadata", json!({ "name": "saved", "desc": "" }));
app.call("save_files", json!({ "fileIds": [id] }));
assert!(inside("moved").contains("<name>saved</name>"));
assert_eq!(name_of(&app, &id), "saved");
// the name of another file is not taken
let error = app.error("rename_file", json!({ "fileId": id, "name": "taken" }));
assert_eq!(error, gpx_studio_desktop_lib::workspace::NAME_TAKEN);
assert!(folder.join("moved.gpx").exists());
// a name is one that a file can have
app.call("metadata", json!({ "name": "a/b", "desc": "" }));
assert_eq!(name_of(&app, &id), "a_b");
app.call("rename_file", json!({ "fileId": id, "name": "a/b" }));
assert!(folder.join("a_b.gpx").exists());
// a file with edits is renamed, but not written: its edits are not saved for it
app.call("style", json!({ "color": "#ff0000" }));
assert_eq!(dirty(&app), true);
app.call("metadata", json!({ "name": "edited", "desc": "" }));
assert!(folder.join("edited.gpx").exists());
assert_eq!(dirty(&app), true);
assert!(!inside("edited").contains("ff0000"));
app.call("undo", json!({}));
app.call("undo", json!({}));
assert_eq!(name_of(&app, &id), "a_b");
assert_eq!(dirty(&app), false);
// only the case changes: the same file
app.call("metadata", json!({ "name": "A_B", "desc": "" }));
assert_eq!(name_of(&app, &id), "A_B");
app.call("rename_file", json!({ "fileId": id, "name": "A_B" }));
let names: Vec<String> = std::fs::read_dir(&folder)
.unwrap()
.map(|entry| entry.unwrap().file_name().to_string_lossy().into_owned())
.collect();
assert!(names.contains(&"A_B.gpx".to_string()));
assert_eq!(names.len(), 2);
assert_eq!(name_of(&app, &id), "saved");
std::fs::remove_dir_all(folder).unwrap();
}
#[test]
fn a_new_file_written_under_another_name_takes_it() {
fn a_new_file_written_under_another_name_keeps_its_own() {
let app = App::new();
let folder = folder_with(&["route"]);
// a new file made in the folder is written there at once: route.gpx is taken
@@ -669,7 +692,8 @@ fn a_new_file_written_under_another_name_takes_it() {
);
let id = created["added"][0].as_str().unwrap().to_string();
assert!(folder.join("route (2).gpx").exists());
assert_eq!(name_of(&app, &id), "route (2)");
// the file keeps its name, whatever the name of the file it was written to
assert_eq!(name_of(&app, &id), "route");
assert_eq!(
app.call("workspace_state", json!({}))["files"][&id]["dirty"],
false
@@ -698,7 +722,7 @@ fn a_file_open_as_a_preview_is_replaced_until_it_is_kept_or_edited() {
.as_array()
.unwrap()
.iter()
.map(|id| name_of(app, id.as_str().unwrap()))
.map(|id| file_stem_of(app, id.as_str().unwrap()))
.collect()
};
+15
View File
@@ -0,0 +1,15 @@
//! The tests of the desktop app. They are integration tests, in a single binary, rather than unit
//! tests: on Windows, only test binaries can be given the manifest that the app needs to start
//! (see `build.rs`).
mod commands;
mod store;
mod workspace;
use gpx_studio_desktop_lib::state::AppState;
/// An engine that keeps its data in a directory of its own, and has nobody to tell.
fn temporary_state() -> AppState {
let directory = std::env::temp_dir().join(format!("gpx-studio-{}", uuid::Uuid::new_v4()));
AppState::open(directory, std::sync::Arc::new(|_, _| {})).expect("the storage can be opened")
}
+69
View File
@@ -0,0 +1,69 @@
//! The storage of the engine on the file system (`store`).
use std::fs;
use std::path::PathBuf;
use gpx_engine::{
Action, Batch, Command, CommitError, Load, LoadFiles, Persister, Revision, Storage,
};
use gpx_studio_desktop_lib::Engine;
use gpx_studio_desktop_lib::store::FileStorage;
use uuid::Uuid;
fn directory() -> PathBuf {
std::env::temp_dir().join(format!("gpx-studio-store-{}", Uuid::new_v4()))
}
fn engine_with_a_file() -> Engine {
let mut engine = Engine::default();
let data = std::fs::read("../gpx-rs/data/simple.gpx").unwrap();
engine
.execute(Action::Edit(Command::LoadFiles(LoadFiles {
files: vec![Load {
data: &data,
name: "simple",
}],
})))
.unwrap();
engine
}
#[test]
fn keeps_the_files_and_the_settings_between_sessions() {
let root = directory();
let engine = engine_with_a_file();
let order = engine.order().to_vec();
let storage = FileStorage::open(&root).unwrap();
pollster::block_on(Persister::default().save(&storage, engine.snapshot())).unwrap();
pollster::block_on(storage.put_setting("units", "\"metric\"")).unwrap();
// another session
let storage = FileStorage::open(&root).unwrap();
let restored = pollster::block_on(Persister::default().restore(&storage)).unwrap();
let mut other = Engine::default();
other.restore(restored);
assert_eq!(other.order(), order);
assert_eq!(
other.file_statistics(&order[0]).unwrap().total_distance,
engine.file_statistics(&order[0]).unwrap().total_distance
);
assert_eq!(storage.settings()["units"], "\"metric\"");
fs::remove_dir_all(root).unwrap();
}
#[test]
fn refuses_a_commit_that_is_not_based_on_the_head() {
let root = directory();
let storage = FileStorage::open(&root).unwrap();
let batch = Batch {
order: Some(vec![]),
..Default::default()
};
assert_eq!(pollster::block_on(storage.commit(&batch)), Ok(Revision(1)));
assert_eq!(
pollster::block_on(storage.commit(&batch)),
Err(CommitError::Stale { head: Revision(1) })
);
fs::remove_dir_all(root).unwrap();
}
+811
View File
@@ -0,0 +1,811 @@
//! The places of the files of the engine on the disk (`workspace`).
use std::collections::BTreeMap;
use std::fs;
use std::path::{Path, PathBuf};
use gpx_engine::{FileId, SelectMode, api};
use gpx_studio_desktop_lib::Engine;
use gpx_studio_desktop_lib::workspace::*;
use uuid::Uuid;
fn gpx(name: &str) -> String {
format!(
r#"<?xml version="1.0"?><gpx version="1.1" creator="test" xmlns="http://www.topografix.com/GPX/1/1"><metadata><name>{name}</name></metadata><trk><name>{name}</name><trkseg><trkpt lat="50.0" lon="4.0"><ele>10</ele></trkpt><trkpt lat="50.1" lon="4.1"><ele>20</ele></trkpt></trkseg></trk></gpx>"#
)
}
/// The GPX files of a folder and of the folders in it.
pub fn gpx_files(folder: &Path) -> Vec<PathBuf> {
let mut found = vec![];
for entry in fs::read_dir(folder).unwrap().flatten() {
let path = entry.path();
if path.is_dir() {
found.extend(gpx_files(&path));
} else if is_gpx(&path) {
found.push(path);
}
}
found.sort();
found
}
struct Fixture {
folder: PathBuf,
engine: Engine,
workspace: Workspace,
}
impl Fixture {
fn new() -> Self {
let folder = canonical(&std::env::temp_dir()).join(format!("gpx-ws-{}", Uuid::new_v4()));
fs::create_dir_all(&folder).unwrap();
Self {
folder,
engine: Engine::default(),
workspace: Workspace::default(),
}
}
fn write(&self, name: &str, title: &str) -> PathBuf {
let path = self.folder.join(name);
fs::write(&path, gpx(title)).unwrap();
path
}
/// Opens the folder, and all the files that are in it.
fn open_folder(&mut self) {
let folder = self.folder.clone();
self.workspace.open_folder(&mut self.engine, &folder, false);
let files = gpx_files(&folder);
self.workspace.open_files(&mut self.engine, &files, false);
}
fn reconcile(&mut self, changed: Option<&[PathBuf]>) -> Option<api::Outcome> {
self.workspace.reconcile(&mut self.engine, changed)
}
fn save(&mut self, ids: &[Uuid]) {
self.workspace
.save(&mut self.engine, ids, &BTreeMap::new(), None)
.unwrap();
}
fn ids(&self) -> Vec<Uuid> {
self.engine.order().iter().map(|id| id.0).collect()
}
fn id_named(&self, name: &str) -> Uuid {
self.ids()
.into_iter()
.find(|id| self.name(*id) == name)
.unwrap()
}
fn name(&self, id: Uuid) -> String {
self.engine
.file_structure(&FileId(id))
.unwrap()
.name
.clone()
}
/// What the file holds: the name of its track (its own name is the one of the file).
fn content(&self, id: Uuid) -> String {
self.engine.file_structure(&FileId(id)).unwrap().tracks[0]
.info
.name
.clone()
.unwrap_or_default()
}
fn rename(&mut self, id: Uuid, name: &str) {
api::select(&mut self.engine, &[id], SelectMode::Replace);
assert!(!api::metadata(&mut self.engine, name, "").has_error());
}
fn status(&self, id: Uuid) -> FileStatus {
self.workspace
.state(&self.engine)
.files
.remove(&id.to_string())
.unwrap()
}
}
impl Drop for Fixture {
fn drop(&mut self) {
let _ = fs::remove_dir_all(&self.folder);
}
}
#[test]
fn an_opened_file_is_clean_until_it_is_edited_and_clean_again_once_saved() {
let mut fx = Fixture::new();
let path = fx.write("a.gpx", "a");
fx.workspace
.open_files(&mut fx.engine, std::slice::from_ref(&path), true);
let id = fx.ids()[0];
assert!(!fx.workspace.is_dirty(&fx.engine, id));
assert_eq!(fx.status(id).path.as_deref(), path.to_str());
fx.rename(id, "renamed");
assert!(fx.workspace.is_dirty(&fx.engine, id));
// nothing is written before it is saved
assert!(fs::read_to_string(&path).unwrap().contains(">a<"));
fx.save(&[id]);
assert!(!fx.workspace.is_dirty(&fx.engine, id));
assert!(fs::read_to_string(&path).unwrap().contains(">renamed<"));
// going back before the save is an edit again
api::select(&mut fx.engine, &[id], SelectMode::Replace);
api::undo(&mut fx.engine);
assert!(fx.workspace.is_dirty(&fx.engine, id));
api::redo(&mut fx.engine);
assert!(!fx.workspace.is_dirty(&fx.engine, id));
}
#[test]
fn a_file_that_is_not_on_the_disk_is_unsaved_until_it_is_given_a_place() {
let mut fx = Fixture::new();
api::new_file(&mut fx.engine, "fresh", None, None, None, false);
let id = fx.ids()[0];
assert!(fx.workspace.is_dirty(&fx.engine, id));
assert_eq!(
fx.workspace.unlinked(&fx.engine, &[id]),
[(id, "fresh".to_string())]
);
// nowhere to put it without being told
fx.save(&[id]);
assert!(fx.workspace.is_dirty(&fx.engine, id));
let path = fx.folder.join("fresh.gpx");
let paths = BTreeMap::from([(id, path.clone())]);
fx.workspace
.save(&mut fx.engine, &[id], &paths, None)
.unwrap();
assert!(path.exists());
assert!(!fx.workspace.is_dirty(&fx.engine, id));
assert!(fx.workspace.unlinked(&fx.engine, &[id]).is_empty());
}
#[test]
fn several_files_without_a_place_go_in_a_folder_with_names_that_are_free() {
let mut fx = Fixture::new();
fx.write("fresh.gpx", "already there");
api::new_file(&mut fx.engine, "fresh", None, None, None, false);
api::new_file(&mut fx.engine, "a/b: c", None, None, None, false);
api::new_file(&mut fx.engine, "fresh", None, None, None, false);
let ids = fx.ids();
let folder = fx.folder.clone();
fx.workspace
.save(&mut fx.engine, &ids, &BTreeMap::new(), Some(&folder))
.unwrap();
for name in ["fresh (2).gpx", "fresh (3).gpx", "a_b_ c.gpx"] {
assert!(fx.folder.join(name).exists(), "{name}");
}
assert!(ids.iter().all(|id| !fx.workspace.is_dirty(&fx.engine, *id)));
}
#[test]
fn what_this_app_writes_is_not_a_change_of_the_disk() {
let mut fx = Fixture::new();
let path = fx.write("a.gpx", "a");
fx.open_folder();
let id = fx.ids()[0];
api::select(&mut fx.engine, &[id], SelectMode::Replace);
assert!(!api::style(&mut fx.engine, Some("#ff0000"), None, None).has_error());
fx.save(&[id]);
assert!(fx.reconcile(Some(&[path])).is_none());
assert!(!fx.status(id).replaced && !fx.status(id).dirty);
}
#[test]
fn the_folders_of_the_files_outside_the_open_folder_are_watched_too() {
let mut fx = Fixture::new();
let inside = fx.write("in.gpx", "in");
let elsewhere = canonical(&std::env::temp_dir()).join(format!("gpx-ws-{}", Uuid::new_v4()));
fs::create_dir_all(&elsewhere).unwrap();
let (one, two) = (elsewhere.join("one.gpx"), elsewhere.join("two.gpx"));
fs::write(&one, gpx("one")).unwrap();
fs::write(&two, gpx("two")).unwrap();
let folder = fx.folder.clone();
// no folder: only the folder of the files
fx.workspace.open_files(&mut fx.engine, &[one, two], false);
assert_eq!(fx.workspace.watched(), [(elsewhere.clone(), false)]);
// the open folder as a whole, and the file in it needs nothing more
fx.workspace.open_folder(&mut fx.engine, &folder, false);
fx.workspace.open_files(&mut fx.engine, &[inside], false);
assert_eq!(
fx.workspace.watched(),
[(folder, true), (elsewhere.clone(), false)]
);
// closing the files stops it
let ids = fx.ids();
fx.workspace.close_files(&mut fx.engine, &ids, true);
assert_eq!(fx.workspace.watched(), [(fx.folder.clone(), true)]);
fs::remove_dir_all(elsewhere).unwrap();
}
#[test]
fn opening_a_folder_does_not_open_its_files() {
let mut fx = Fixture::new();
fx.write("a.gpx", "a");
let folder = fx.folder.clone();
let outcome = fx.workspace.open_folder(&mut fx.engine, &folder, false);
assert!(!outcome.changed);
assert!(fx.ids().is_empty());
assert_eq!(fx.workspace.folder.as_deref(), Some(folder.as_path()));
}
#[test]
fn folders_are_made_renamed_and_moved_and_the_open_files_follow() {
let mut fx = Fixture::new();
let a = fx.write("a.gpx", "a");
fx.open_folder();
let id = fx.ids()[0];
let root = fx.folder.clone();
let path_of = |fx: &Fixture| PathBuf::from(fx.status(id).path.unwrap());
let first = fx.workspace.create_folder(&root, "New folder").unwrap();
let second = fx.workspace.create_folder(&root, "New folder").unwrap();
assert_eq!(first, root.join("New folder"));
assert_eq!(second, root.join("New folder 2"));
fx.workspace.move_paths(&[a], &first).unwrap();
assert_eq!(path_of(&fx), first.join("a.gpx"));
assert!(first.join("a.gpx").exists() && !fx.status(id).dirty);
let trips = fx.workspace.rename_path(&first, "Trips").unwrap();
assert_eq!(path_of(&fx), root.join("Trips/a.gpx"));
// a folder does not go into itself, nor onto something else
assert!(
fx.workspace
.move_paths(std::slice::from_ref(&trips), &trips)
.is_err()
);
assert_eq!(
fx.workspace.rename_path(&trips, "New folder 2"),
Err(NAME_TAKEN.to_owned())
);
fx.workspace.move_paths(&[trips], &second).unwrap();
assert_eq!(path_of(&fx), second.join("Trips/a.gpx"));
// a file keeps its extension
let closed = fx.write("c.gpx", "c");
assert_eq!(
fx.workspace.rename_path(&closed, "d").unwrap(),
root.join("d.gpx")
);
// the open folder and what is not in it are left alone
assert!(fx.workspace.rename_path(&root, "x").is_err());
let elsewhere = canonical(&std::env::temp_dir());
assert!(fx.workspace.rename_path(&elsewhere, "x").is_err());
assert!(fx.workspace.create_folder(&elsewhere, "x").is_err());
}
#[test]
fn deleted_files_are_closed_unless_they_have_edits() {
let mut fx = Fixture::new();
let a = fx.write("a.gpx", "a");
let b = fx.write("b.gpx", "b");
fx.open_folder();
let (id_a, id_b) = (fx.id_named("a"), fx.id_named("b"));
api::select(&mut fx.engine, &[id_b], SelectMode::Replace);
assert!(!api::style(&mut fx.engine, Some("#ff0000"), None, None).has_error());
let outcome = fx
.workspace
.delete_paths(&mut fx.engine, &[a.clone(), b.clone()], |paths| {
paths.iter().for_each(|path| fs::remove_file(path).unwrap());
Ok(())
})
.unwrap();
assert!(outcome.changed);
assert_eq!(fx.ids(), [id_b]);
assert!(!fx.ids().contains(&id_a));
let status = fx.status(id_b);
assert!(status.dirty && status.missing);
assert!(!a.exists() && !b.exists());
}
#[test]
fn the_tree_of_the_folder_shows_folders_first_and_only_gpx_files() {
let mut fx = Fixture::new();
fx.write("b.gpx", "b");
fx.write("A.GPX", "a");
fx.write("b 10.gpx", "b 10");
fx.write("b 2.gpx", "b 2");
fx.write("notes.txt", "not a track");
fx.write(".hidden.gpx", "hidden");
fs::create_dir(fx.folder.join("zoo")).unwrap();
fs::create_dir(fx.folder.join("Alps")).unwrap();
fs::create_dir(fx.folder.join(".git")).unwrap();
fx.write("zoo/c.gpx", "c");
let folder = fx.folder.clone();
fx.workspace.open_folder(&mut fx.engine, &folder, false);
let names = |entries: Vec<Entry>| -> Vec<(String, bool)> {
entries.into_iter().map(|e| (e.name, e.is_dir)).collect()
};
assert_eq!(
names(fx.workspace.list_folder(&folder).unwrap()),
[
("Alps".to_string(), true),
("zoo".to_string(), true),
("A.GPX".to_string(), false),
("b.gpx".to_string(), false),
("b 2.gpx".to_string(), false),
("b 10.gpx".to_string(), false),
]
);
assert_eq!(
names(fx.workspace.list_folder(&folder.join("zoo")).unwrap()),
[("c.gpx".to_string(), false)]
);
// only what is in the open folder can be listed
let elsewhere = canonical(&std::env::temp_dir());
assert!(fx.workspace.list_folder(&elsewhere).is_err());
}
#[test]
fn names_are_sorted_by_the_numbers_they_hold() {
let mut names = vec![
"track 10",
"track 2",
"Track 1",
"track",
"track 02b",
"track 2a",
"b",
"A",
];
names.sort_by(|a, b| natural_cmp(a, b).then_with(|| a.cmp(b)));
assert_eq!(
names,
[
"A",
"b",
"track",
"Track 1",
"track 2",
"track 2a",
"track 02b",
"track 10",
]
);
}
#[test]
fn a_file_without_edits_is_read_again_when_the_disk_changes() {
let mut fx = Fixture::new();
let a = fx.write("a.gpx", "a");
fx.open_folder();
let id = fx.ids()[0];
fx.write("a.gpx", "theirs");
let outcome = fx.reconcile(Some(std::slice::from_ref(&a))).unwrap();
assert!(outcome.changed);
// the same file, now the one on the disk
assert_eq!(fx.ids(), [id]);
assert_eq!(fx.content(id), "theirs");
let status = fx.status(id);
assert!(!status.dirty && !status.replaced);
// looking again changes nothing
assert!(fx.reconcile(Some(std::slice::from_ref(&a))).is_none());
}
#[test]
fn a_file_with_edits_is_flagged_when_the_disk_changes_and_not_read_again() {
let mut fx = Fixture::new();
let a = fx.write("a.gpx", "a");
fx.open_folder();
let id = fx.ids()[0];
fx.rename(id, "mine");
fx.write("a.gpx", "theirs");
let outcome = fx.reconcile(Some(std::slice::from_ref(&a))).unwrap();
assert!(!outcome.changed);
assert_eq!(fx.name(id), "mine");
let status = fx.status(id);
assert!(status.dirty && status.replaced && !status.missing);
// looking again changes nothing
assert!(fx.reconcile(Some(std::slice::from_ref(&a))).is_none());
// the disk goes back to what was read: only the edits are left
fx.write("a.gpx", "a");
assert!(fx.reconcile(Some(std::slice::from_ref(&a))).is_some());
let status = fx.status(id);
assert!(status.dirty && !status.replaced);
}
#[test]
fn the_user_can_take_the_disk_instead_of_the_edits() {
let mut fx = Fixture::new();
fx.write("a.gpx", "a");
fx.open_folder();
let id = fx.ids()[0];
fx.rename(id, "mine");
fx.write("a.gpx", "theirs");
fx.reconcile(None);
assert!(fx.status(id).replaced);
fx.workspace.reload(&mut fx.engine, &[id]).unwrap();
assert_eq!(fx.content(id), "theirs");
let status = fx.status(id);
assert!(!status.dirty && !status.replaced);
// and it is one undo step
api::select(&mut fx.engine, &[id], SelectMode::Replace);
api::undo(&mut fx.engine);
assert_eq!(fx.content(id), "mine");
}
#[test]
fn saving_overwrites_what_replaced_the_file() {
let mut fx = Fixture::new();
let a = fx.write("a.gpx", "a");
fx.open_folder();
let id = fx.ids()[0];
fx.rename(id, "mine");
fx.write("a.gpx", "theirs");
fx.reconcile(None);
assert!(fx.status(id).replaced);
fx.save(&[id]);
assert!(fs::read_to_string(a).unwrap().contains(">mine<"));
assert!(!fx.status(id).replaced && !fx.status(id).dirty);
}
#[test]
fn a_file_that_is_gone_from_the_disk_stays_open_and_dirty_until_it_is_saved_again() {
let mut fx = Fixture::new();
let a = fx.write("a.gpx", "a");
fx.open_folder();
let id = fx.ids()[0];
fs::remove_file(&a).unwrap();
assert!(fx.reconcile(Some(std::slice::from_ref(&a))).is_some());
assert_eq!(fx.ids(), [id]);
let status = fx.status(id);
assert!(status.dirty && status.missing && status.path.is_some());
// saving puts it back
fx.save(&[id]);
assert!(a.exists());
let status = fx.status(id);
assert!(!status.dirty && !status.missing);
}
#[test]
fn a_file_that_comes_back_different_is_read_again() {
let mut fx = Fixture::new();
let a = fx.write("a.gpx", "a");
fx.open_folder();
let id = fx.ids()[0];
fs::remove_file(&a).unwrap();
fx.reconcile(None);
assert!(fx.status(id).missing);
fx.write("a.gpx", "back");
fx.reconcile(None);
assert_eq!(fx.content(id), "back");
let status = fx.status(id);
assert!(!status.dirty && !status.missing);
}
#[test]
fn what_changed_while_the_app_was_closed_is_found_when_it_starts() {
let mut fx = Fixture::new();
let a = fx.write("a.gpx", "a");
fx.open_folder();
let id = fx.ids()[0];
fx.write("a.gpx", "changed");
assert!(fx.reconcile(None).is_some());
assert_eq!(fx.content(id), "changed");
fs::remove_file(a).unwrap();
assert!(fx.reconcile(None).is_some());
assert!(fx.status(id).missing);
}
#[test]
fn new_files_made_from_a_file_of_the_folder_are_saved_next_to_it() {
let mut fx = Fixture::new();
fs::create_dir(fx.folder.join("sub")).unwrap();
fx.write("sub/b.gpx", "b");
fx.write("a.gpx", "a");
fx.open_folder();
let b = fx.id_named("b");
let sub = fx.folder.join("sub");
assert_eq!(fx.workspace.source_dir(&[b]), Some(canonical(&sub)));
let before = fx.ids();
api::select(&mut fx.engine, &[b], SelectMode::Replace);
let outcome = api::duplicate(&mut fx.engine);
let copy = fx
.ids()
.into_iter()
.find(|id| !before.contains(id))
.unwrap();
fx.workspace.inherit(outcome.added(), &canonical(&sub));
assert_eq!(fx.status(copy).folder.as_deref(), canonical(&sub).to_str());
assert!(fx.status(copy).path.is_none() && fx.status(copy).dirty);
// nobody needs to be asked where it goes
assert!(fx.workspace.unlinked(&fx.engine, &[copy]).is_empty());
fx.save(&[copy]);
assert!(sub.join("b (2).gpx").exists());
let status = fx.status(copy);
assert!(!status.dirty && status.folder.is_none());
assert_eq!(
status.path.as_deref(),
canonical(&sub.join("b (2).gpx")).to_str()
);
// the copy of the copy is next to it as well
assert_eq!(fx.workspace.source_dir(&[copy]), Some(canonical(&sub)));
}
#[test]
fn a_file_outside_the_open_folder_hands_nothing_down() {
let mut fx = Fixture::new();
let elsewhere = canonical(&std::env::temp_dir()).join(format!("gpx-ws-{}", Uuid::new_v4()));
fs::create_dir_all(&elsewhere).unwrap();
let outside = elsewhere.join("o.gpx");
fs::write(&outside, gpx("o")).unwrap();
let folder = fx.folder.clone();
fx.workspace.open_folder(&mut fx.engine, &folder, false);
fx.workspace
.open_files(&mut fx.engine, std::slice::from_ref(&outside), true);
let id = fx.ids()[0];
assert_eq!(fx.workspace.source_dir(&[id]), None);
// a new file with no source is not linked to a folder by `source_dir`: the caller falls back
fs::remove_dir_all(elsewhere).unwrap();
}
#[test]
fn closing_files_asks_what_to_do_with_the_unsaved_ones() {
let mut fx = Fixture::new();
fx.write("a.gpx", "a");
fx.write("b.gpx", "b");
fx.open_folder();
let (a, b) = (fx.id_named("a"), fx.id_named("b"));
fx.rename(a, "edited");
assert_eq!(fx.workspace.unsaved(&fx.engine, &[a, b]), [a]);
// without a choice what has edits stays
let outcome = fx.workspace.close_files(&mut fx.engine, &[a, b], false);
assert!(outcome.changed);
assert_eq!(fx.ids(), [a]);
// saved first
fx.save(&[a]);
fx.workspace.close_files(&mut fx.engine, &[a], false);
assert!(fx.ids().is_empty());
assert!(
fs::read_to_string(fx.folder.join("a.gpx"))
.unwrap()
.contains("edited")
);
assert!(fx.workspace.links.is_empty());
}
#[test]
fn closing_files_can_discard_their_edits() {
let mut fx = Fixture::new();
fx.write("a.gpx", "a");
fx.open_folder();
let a = fx.ids()[0];
fx.rename(a, "edited");
api::new_file(&mut fx.engine, "fresh", None, None, None, false);
let fresh = fx.id_named("fresh");
fx.workspace.close_files(&mut fx.engine, &[a, fresh], true);
assert!(fx.ids().is_empty());
assert!(
fs::read_to_string(fx.folder.join("a.gpx"))
.unwrap()
.contains(">a<")
);
}
#[test]
fn closing_the_folder_closes_every_file() {
let mut fx = Fixture::new();
fx.write("a.gpx", "a");
fx.open_folder();
api::new_file(&mut fx.engine, "fresh", None, None, None, false);
assert_eq!(fx.ids().len(), 2);
// the new file has nowhere to go: it has to be discarded
let outcome = fx.workspace.close_folder(&mut fx.engine, false);
assert!(outcome.changed);
assert_eq!(fx.ids().len(), 1);
assert!(fx.workspace.folder.is_none());
let outcome = fx.workspace.close_folder(&mut fx.engine, true);
assert!(outcome.changed);
assert!(fx.ids().is_empty());
}
#[test]
fn opening_another_folder_closes_the_first_with_all_its_files() {
let mut fx = Fixture::new();
fx.write("a.gpx", "a");
fx.open_folder();
let a = fx.ids()[0];
fx.rename(a, "edited");
let other = canonical(&std::env::temp_dir()).join(format!("gpx-ws-{}", Uuid::new_v4()));
fs::create_dir_all(&other).unwrap();
// the edits stay, nothing is lost without a choice
fx.workspace.open_folder(&mut fx.engine, &other, false);
assert_eq!(fx.ids(), [a]);
assert_eq!(fx.workspace.folder.as_deref(), Some(other.as_path()));
let first = fx.folder.clone();
fx.workspace.open_folder(&mut fx.engine, &first, true);
assert!(fx.ids().is_empty());
assert!(
fs::read_to_string(fx.folder.join("a.gpx"))
.unwrap()
.contains(">a<")
);
fs::remove_dir_all(other).unwrap();
}
#[test]
fn opening_the_same_folder_again_changes_nothing() {
let mut fx = Fixture::new();
fx.write("a.gpx", "a");
fx.open_folder();
let folder = fx.folder.clone();
let outcome = fx.workspace.open_folder(&mut fx.engine, &folder, true);
assert!(!outcome.changed);
assert_eq!(fx.ids().len(), 1);
}
#[test]
fn a_failed_save_stops_at_the_file_and_leaves_it_dirty() {
let mut fx = Fixture::new();
fx.write("a.gpx", "a");
fx.open_folder();
let id = fx.ids()[0];
fx.rename(id, "edited");
// the place of the file is taken by a folder
fs::remove_file(fx.folder.join("a.gpx")).unwrap();
fs::create_dir(fx.folder.join("a.gpx")).unwrap();
let result = fx
.workspace
.save(&mut fx.engine, &[id], &BTreeMap::new(), None);
assert!(result.is_err());
assert!(fx.workspace.is_dirty(&fx.engine, id));
}
#[test]
fn saving_as_gives_the_file_another_place() {
let mut fx = Fixture::new();
let a = fx.write("a.gpx", "a");
fx.write("b.gpx", "b");
fx.open_folder();
let id = fx.id_named("a");
let other = fx.id_named("b");
fx.rename(id, "edited");
let target = fx.folder.join("copy.gpx");
fx.workspace.save_as(&mut fx.engine, id, &target).unwrap();
assert!(fs::read_to_string(&target).unwrap().contains("edited"));
assert!(fs::read_to_string(a).unwrap().contains(">a<"));
let status = fx.status(id);
assert!(!status.dirty);
assert_eq!(status.path.as_deref(), canonical(&target).to_str());
// the place of another open file is not available
let taken = fx.folder.join("b.gpx");
assert!(fx.workspace.save_as(&mut fx.engine, id, &taken).is_err());
assert!(!fx.workspace.is_dirty(&fx.engine, other));
}
mod restart {
use gpx_engine::Persister;
use super::*;
use gpx_studio_desktop_lib::store::FileStorage;
/// What the app does when it starts again: the engine and the links are read from what was
/// kept.
fn restart(engine: &Engine, workspace: &Workspace, root: &Path) -> (Engine, Workspace) {
let storage = FileStorage::open(root).unwrap();
pollster::block_on(Persister::default().save(&storage, engine.snapshot())).unwrap();
let mut restarted = Engine::default();
let restored = pollster::block_on(Persister::default().restore(&storage)).unwrap();
restarted.restore(restored);
let links = serde_json::to_string(workspace).unwrap();
(restarted, serde_json::from_str(&links).unwrap())
}
fn setup() -> (PathBuf, Engine, Workspace, Vec<Uuid>) {
let root = std::env::temp_dir().join(format!("gpx-restart-{}", Uuid::new_v4()));
let folder = root.join("tracks");
fs::create_dir_all(&folder).unwrap();
for name in ["with_tracks", "with_time", "with_waypoint"] {
let _ = fs::copy(
format!("../gpx-rs/data/{name}.gpx"),
folder.join(format!("{name}.gpx")),
);
}
let mut engine = Engine::default();
let mut workspace = Workspace::default();
workspace.open_folder(&mut engine, &folder, false);
workspace.open_files(&mut engine, &gpx_files(&folder), false);
let ids = engine.order().iter().map(|id| id.0).collect();
(root, engine, workspace, ids)
}
#[test]
fn files_that_were_not_edited_are_not_unsaved_after_a_restart() {
let (root, engine, workspace, ids) = setup();
assert!(ids.len() >= 2);
let (restarted, mut workspace) = restart(&engine, &workspace, &root.join("store"));
// the files are the same, but the engine knows their contents by new identities
assert!(ids.iter().any(|id| workspace.is_dirty(&restarted, *id)));
workspace.verify(&restarted);
assert!(ids.iter().all(|id| !workspace.is_dirty(&restarted, *id)));
fs::remove_dir_all(root).unwrap();
}
#[test]
fn files_that_were_edited_are_still_unsaved_after_a_restart() {
let (root, mut engine, workspace, ids) = setup();
api::select(&mut engine, &[ids[0]], gpx_engine::SelectMode::Replace);
api::metadata(&mut engine, "edited", "");
assert!(workspace.is_dirty(&engine, ids[0]));
let (restarted, mut workspace) = restart(&engine, &workspace, &root.join("store"));
workspace.verify(&restarted);
assert!(workspace.is_dirty(&restarted, ids[0]));
assert!(
ids[1..]
.iter()
.all(|id| !workspace.is_dirty(&restarted, *id))
);
fs::remove_dir_all(root).unwrap();
}
#[test]
fn links_made_before_the_content_was_kept_are_checked_against_the_disk() {
let (root, mut engine, mut workspace, ids) = setup();
api::select(&mut engine, &[ids[0]], gpx_engine::SelectMode::Replace);
api::metadata(&mut engine, "edited", "");
for link in workspace.links.values_mut() {
link.export_hash = 0;
}
let (restarted, mut workspace) = restart(&engine, &workspace, &root.join("store"));
workspace.verify(&restarted);
assert!(workspace.is_dirty(&restarted, ids[0]));
assert!(
ids[1..]
.iter()
.all(|id| !workspace.is_dirty(&restarted, *id))
);
fs::remove_dir_all(root).unwrap();
}
#[test]
fn the_folders_of_new_files_and_the_old_flag_name_survive_a_restart() {
let (root, engine, mut workspace, ids) = setup();
let folder = workspace.folder.clone().unwrap();
let fresh = Uuid::new_v4();
workspace.inherit(&[fresh.to_string()], &folder);
workspace.links.get_mut(&ids[0]).unwrap().replaced = true;
let json = serde_json::to_string(&workspace).unwrap();
// links written by the previous version had `conflict`, which is the same thing
let json = json.replace("\"replaced\"", "\"conflict\"");
assert!(json.contains("\"conflict\":true"));
let read: Workspace = serde_json::from_str(&json).unwrap();
assert!(read.links[&ids[0]].replaced);
assert_eq!(read.dirs.get(&fresh), Some(&folder));
let _ = engine;
fs::remove_dir_all(root).unwrap();
}
}
+14
View File
@@ -0,0 +1,14 @@
<assembly xmlns="urn:schemas-microsoft-com:asm.v1" manifestVersion="1.0">
<dependency>
<dependentAssembly>
<assemblyIdentity
type="win32"
name="Microsoft.Windows.Common-Controls"
version="6.0.0.0"
processorArchitecture="*"
publicKeyToken="6595b64144ccf1df"
language="*"
/>
</dependentAssembly>
</dependency>
</assembly>
+11
View File
@@ -18,6 +18,17 @@ pub fn style(color: Option<String>, opacity: Option<f64>, width: Option<f64>) ->
run(|e| api::style(e, color.as_deref(), opacity, width))
}
/// Converts the selected routes to tracks, or the selected tracks or segments to routes.
#[wasm_bindgen]
pub fn convert() -> Outcome {
run(api::convert)
}
#[wasm_bindgen]
pub fn new_route() -> Outcome {
run(api::new_route)
}
#[wasm_bindgen]
pub fn new_track() -> Outcome {
run(api::new_track)
+11 -4
View File
@@ -10,11 +10,18 @@ use crate::ts::*;
// File commands
/// Creates a file and selects it. With `lng` and `lat`, it starts with a track and a segment that
/// hold that trackpoint (at `ele`, 0 by default).
/// Creates a file and selects it. With `route`, it starts with a route. With `lng` and `lat`, it
/// starts with a segment that holds that trackpoint (at `ele`, 0 by default): the one of its route
/// with `route`, of a track otherwise.
#[wasm_bindgen]
pub fn new_file(name: &str, lng: Option<f64>, lat: Option<f64>, ele: Option<f64>) -> Outcome {
run(|e| api::new_file(e, name, lng, lat, ele))
pub fn new_file(
name: &str,
lng: Option<f64>,
lat: Option<f64>,
ele: Option<f64>,
route: bool,
) -> Outcome {
run(|e| api::new_file(e, name, lng, lat, ele, route))
}
/// Loads files as a single command (one undo step). The files are in `data`, one after the
+1 -5
View File
@@ -45,10 +45,8 @@ pub fn selection_statistics() -> GlobalStatistics {
}
/// A file as GPX (UTF-8 bytes), `undefined` if it does not exist. The options say which data of the trackpoints
/// to keep, and with `as_route` that the segments are written as routes (`rte`) instead of tracks:
/// the anchors are the route points, and the trackpoints between them the points of their path.
/// to keep.
#[wasm_bindgen]
#[allow(clippy::too_many_arguments)]
pub fn export_file(
file_id: &str,
time: bool,
@@ -57,7 +55,6 @@ pub fn export_file(
atemp: bool,
power: bool,
osm: bool,
as_route: bool,
) -> Option<Vec<u8>> {
let id = parse_uuid(file_id)?;
let options = ExportOptions {
@@ -67,7 +64,6 @@ pub fn export_file(
atemp,
power,
osm,
as_route,
};
with_engine(|e| api::export_file(e, id, options)).flatten()
}
+8
View File
@@ -41,6 +41,14 @@ pub fn arrow_select(down: bool, add: bool) -> Outcome {
// with the selection when it holds elements of the same kind in the same place (same file, same
// track for segments), otherwise they replace it.
#[wasm_bindgen]
pub fn select_routes(file_id: &str, route_ids_bytes: &[u8], mode: SelectMode) -> Outcome {
match parse_uuid(file_id).zip(uuid_list(route_ids_bytes)) {
Some((file_id, ids)) => run(|e| api::select_routes(e, file_id, &ids, mode.into())),
None => fail("the ids are not UUIDs"),
}
}
#[wasm_bindgen]
pub fn select_tracks(file_id: &str, track_ids_bytes: &[u8], mode: SelectMode) -> Outcome {
match parse_uuid(file_id).zip(uuid_list(track_ids_bytes)) {
+27 -21
View File
@@ -16,6 +16,7 @@ use crate::types::*;
const SIMPLE: &[u8] = include_bytes!("../../gpx-rs/data/simple.gpx");
const WITH_TIME: &[u8] = include_bytes!("../../gpx-rs/data/with_time.gpx");
const WITH_SURFACE: &[u8] = include_bytes!("../../gpx-rs/data/with_surface.gpx");
const WITH_ROUTES: &[u8] = include_bytes!("../../gpx-rs/data/with_routes.gpx");
thread_local! {
static LAST: std::cell::RefCell<JsValue> = const { std::cell::RefCell::new(JsValue::UNDEFINED) };
@@ -83,7 +84,7 @@ fn uuid_bytes(ids: &[String]) -> Vec<u8> {
fn test_new_file_is_listed_and_selected() {
start();
assert!(run(crate::command::files::new_file(
"route", None, None, None
"route", None, None, None, false
)));
assert_eq!(last_error(), None);
@@ -112,10 +113,7 @@ fn test_unknown_file_has_no_structure() {
assert!(crate::queries::files::file_structure(&id).is_undefined());
assert!(crate::queries::files::file_structure("not an id").is_undefined());
assert!(crate::queries::files::file_statistics(&id).is_undefined());
assert!(
crate::queries::files::export_file(&id, true, true, true, true, true, true, false)
.is_none()
);
assert!(crate::queries::files::export_file(&id, true, true, true, true, true, true).is_none());
assert_eq!(
crate::queries::coordinates::segment_coordinates(&id).length(),
0
@@ -200,7 +198,9 @@ fn test_the_reason_of_a_failure_can_be_read() {
assert!(!run(crate::command::tools::reverse()));
assert_eq!(last_error().as_deref(), Some("nothing to do"));
// the next call starts afresh
assert!(run(crate::command::files::new_file("a", None, None, None)));
assert!(run(crate::command::files::new_file(
"a", None, None, None, false
)));
assert_eq!(last_error(), None);
// invalid data
@@ -253,7 +253,9 @@ fn test_the_reason_of_a_failure_can_be_read() {
assert!(last_error().is_some());
// answers are not failures
assert!(run(crate::command::files::new_file("b", None, None, None)));
assert!(run(crate::command::files::new_file(
"b", None, None, None, false
)));
assert!(!crate::command::files::can_paste());
assert_eq!(last_error(), None);
}
@@ -262,7 +264,9 @@ fn test_the_reason_of_a_failure_can_be_read() {
fn test_undo_and_redo() {
started_with(&[]);
assert!(!can_undo() && !can_redo());
assert!(run(crate::command::files::new_file("a", None, None, None)));
assert!(run(crate::command::files::new_file(
"a", None, None, None, false
)));
assert!(can_undo() && !can_redo());
assert!(run(undo()));
assert!(order().is_empty());
@@ -279,8 +283,7 @@ fn test_export_and_exportable_data() {
let ids = order();
let bytes =
crate::queries::files::export_file(&ids[0], true, true, true, true, true, true, false)
.unwrap();
crate::queries::files::export_file(&ids[0], true, true, true, true, true, true).unwrap();
assert!(String::from_utf8(bytes).unwrap().contains("<gpx"));
// the time is the first bit
@@ -300,18 +303,17 @@ fn test_export_and_exportable_data() {
}
#[wasm_bindgen_test]
fn test_export_as_route() {
started_with(&[(SIMPLE, "plain")]);
let id = order().remove(0);
let text = |as_route| {
fn test_export_tracks_and_routes() {
started_with(&[(SIMPLE, "track"), (WITH_ROUTES, "routes")]);
let text = |id: &str| {
let bytes =
crate::queries::files::export_file(&id, true, true, true, true, true, true, as_route)
.unwrap();
crate::queries::files::export_file(id, true, true, true, true, true, true).unwrap();
String::from_utf8(bytes).unwrap()
};
let tracks = text(false);
let ids = order();
let tracks = text(&ids[0]);
assert!(tracks.contains("<trk>") && !tracks.contains("<rte>"));
let routes = text(true);
let routes = text(&ids[1]);
assert!(routes.contains("<rte>") && !routes.contains("<trk>"));
assert!(routes.contains("<rtept"));
}
@@ -390,7 +392,7 @@ fn test_move_elements_takes_js_objects() {
fn test_waypoints() {
started_with(&[]);
assert!(run(crate::command::files::new_file(
"file", None, None, None
"file", None, None, None, false
)));
let file = order().remove(0);
assert!(run(crate::command::waypoints::new_waypoint(
@@ -494,7 +496,9 @@ fn test_editing_tools_change_the_statistics() {
#[wasm_bindgen_test]
fn test_last_update_describes_the_last_call_only() {
started_with(&[]);
assert!(run(crate::command::files::new_file("a", None, None, None)));
assert!(run(crate::command::files::new_file(
"a", None, None, None, false
)));
let update = last_outcome();
assert_eq!(array(get(&update, "added")).length(), 1);
assert_eq!(get(&update, "orderChanged").as_bool(), Some(true));
@@ -506,7 +510,9 @@ fn test_last_update_describes_the_last_call_only() {
assert_eq!(get(&update, "orderChanged").as_bool(), Some(false));
assert_eq!(get(&update, "selectionChanged").as_bool(), Some(false));
assert!(run(crate::command::files::new_file("b", None, None, None)));
assert!(run(crate::command::files::new_file(
"b", None, None, None, false
)));
assert!(!run(crate::command::tools::crop(3, 1)));
let update = last_outcome();
assert_eq!(array(get(&update, "added")).length(), 0);
+18 -5
View File
@@ -28,16 +28,19 @@ const FILE_STRUCTURE_TS: &str = r#"
export type Selection =
| { type: 'empty' }
| { type: 'file'; fileIds: string[] }
| { type: 'route'; fileId: string; routeIds: string[] }
| { type: 'track'; fileId: string; trackIds: string[] }
| { type: 'segment'; fileId: string; trackId: string; segmentIds: string[] }
| { type: 'waypoints'; fileId: string }
| { type: 'waypoint'; fileId: string; waypointIds: string[] };
/**
* Where moved elements go: a list of the file tree and the position in it, counted among the
* elements of the list that are not moved.
* elements of the list that are not moved. Routes and tracks keep their kind: moved among the
* other kind, routes go after the routes, tracks before the tracks.
*/
export type MoveTarget =
| { type: 'files'; index: number }
| { type: 'routes'; fileId: string; index: number }
| { type: 'tracks'; fileId: string; index: number }
| { type: 'segments'; fileId: string; trackId: string; index: number }
| { type: 'waypoints'; fileId: string; index: number };
@@ -83,7 +86,7 @@ export interface SelectionPoint {
}
/** What was copied or cut, to be pasted: the kind of the elements and their ids. */
export interface Clipboard {
type: 'files' | 'tracks' | 'segments' | 'waypoints';
type: 'files' | 'routes' | 'tracks' | 'segments' | 'waypoints';
ids: string[];
cut: boolean;
}
@@ -119,6 +122,8 @@ export interface FileStructure {
id: string;
name: string;
desc?: string;
/** The routes come before the tracks, in the file and in the statistics. */
routes: RouteNode[];
tracks: TrackNode[];
waypoints: WaypointNode[];
/** Changes when the waypoints of the file change: refetch their coordinates. */
@@ -146,14 +151,22 @@ export interface GlobalStatistics {
/** Absent when there are no trackpoints. */
bounds?: { west: number; south: number; east: number; north: number };
}
export interface TrackNode {
id: string;
/** The name, description and style of a route or a track. */
export interface PathInfo {
name?: string;
desc?: string;
/** Style of the track, only present when the file defines it. */
/** Style, only present when the file defines it. */
color?: string;
opacity?: number;
width?: number;
}
/** A route: like a track, with a single segment. */
export interface RouteNode extends PathInfo {
id: string;
segment: SegmentNode;
}
export interface TrackNode extends PathInfo {
id: string;
segments: SegmentNode[];
}
export interface SegmentNode {
+2
View File
@@ -15,6 +15,7 @@ pub enum MergeType {
#[derive(Clone, Copy)]
pub enum SplitType {
Files,
Routes,
Tracks,
Segments,
}
@@ -23,6 +24,7 @@ impl From<SplitType> for engine::SplitType {
fn from(t: SplitType) -> Self {
match t {
SplitType::Files => Self::Files,
SplitType::Routes => Self::Routes,
SplitType::Tracks => Self::Tracks,
SplitType::Segments => Self::Segments,
}
-13
View File
@@ -1,13 +0,0 @@
.DS_Store
node_modules
/build
/.svelte-kit
/package
.env
.env.*
!.env.example
# Ignore files for PNPM, NPM and YARN
pnpm-lock.yaml
package-lock.json
yarn.lock
-31
View File
@@ -1,31 +0,0 @@
/** @type { import("eslint").Linter.Config } */
module.exports = {
root: true,
extends: [
'eslint:recommended',
'plugin:@typescript-eslint/recommended',
'plugin:svelte/recommended',
'prettier',
],
parser: '@typescript-eslint/parser',
plugins: ['@typescript-eslint'],
parserOptions: {
sourceType: 'module',
ecmaVersion: 2020,
extraFileExtensions: ['.svelte'],
},
env: {
browser: true,
es2017: true,
node: true,
},
overrides: [
{
files: ['*.svelte'],
parser: 'svelte-eslint-parser',
parserOptions: {
parser: '@typescript-eslint/parser',
},
},
],
};
+18 -18
View File
@@ -1,20 +1,20 @@
{
"$schema": "https://shadcn-svelte.com/schema.json",
"tailwind": {
"css": "src/app.css",
"baseColor": "neutral"
},
"aliases": {
"components": "$lib/components",
"utils": "$lib/utils",
"ui": "$lib/components/ui",
"hooks": "$lib/hooks",
"lib": "$lib"
},
"typescript": true,
"registry": "https://shadcn-svelte.com/registry",
"style": "nova",
"iconLibrary": "lucide",
"menuColor": "default",
"menuAccent": "subtle"
"$schema": "https://shadcn-svelte.com/schema.json",
"tailwind": {
"css": "src/app.css",
"baseColor": "neutral"
},
"aliases": {
"components": "$lib/components",
"utils": "$lib/utils",
"ui": "$lib/components/ui",
"hooks": "$lib/hooks",
"lib": "$lib"
},
"typescript": true,
"registry": "https://shadcn-svelte.com/registry",
"style": "nova",
"iconLibrary": "lucide",
"menuColor": "default",
"menuAccent": "subtle"
}
+47
View File
@@ -0,0 +1,47 @@
import js from '@eslint/js';
import ts from '@typescript-eslint/eslint-plugin';
import tsParser from '@typescript-eslint/parser';
import svelte from 'eslint-plugin-svelte';
import prettier from 'eslint-config-prettier';
import globals from 'globals';
/** @type {import('eslint').Linter.Config[]} */
export default [
{
ignores: ['build/', '.svelte-kit/', 'package/'],
},
js.configs.recommended,
...ts.configs['flat/recommended'],
...svelte.configs['flat/recommended'],
prettier,
...svelte.configs['flat/prettier'],
{
languageOptions: {
globals: { ...globals.browser, ...globals.node },
},
},
{
rules: {
// TypeScript already reports undefined names
'no-undef': 'off',
'no-empty': ['error', { allowEmptyCatch: true }],
'@typescript-eslint/no-unused-vars': [
'error',
{ argsIgnorePattern: '^_', varsIgnorePattern: '^_' },
],
},
},
{
files: ['**/*.svelte', '**/*.svelte.ts', '**/*.svelte.js'],
languageOptions: {
parserOptions: {
parser: tsParser,
extraFileExtensions: ['.svelte'],
},
},
rules: {
// $effect dependencies are tracked by reading them
'@typescript-eslint/no-unused-expressions': 'off',
},
},
];
+32 -2
View File
@@ -20,7 +20,7 @@
"mapillary-js": "^4.1.2",
"maplibre-gl": "^6.15.0",
"sanitize-html": "^2.17.0",
"sortablejs": "^1.15.6"
"sortablejs": "^1.15.7"
},
"devDependencies": {
"@fontsource-variable/inter": "^5.2.8",
@@ -39,11 +39,12 @@
"@types/mapbox__tilebelt": "^1.0.4",
"@types/node": "^22.15.30",
"@types/sanitize-html": "^2.16.0",
"@types/sortablejs": "^1.15.8",
"@types/sortablejs": "^1.15.9",
"@typescript-eslint/eslint-plugin": "^8.33.1",
"@typescript-eslint/parser": "^8.33.1",
"bits-ui": "^2.17.2",
"clsx": "^2.1.1",
"cross-env": "^10.1.0",
"eslint": "^9.28.0",
"eslint-config-prettier": "^10.1.5",
"eslint-plugin-svelte": "^3.9.1",
@@ -354,6 +355,13 @@
"tslib": "^2.4.0"
}
},
"node_modules/@epic-web/invariant": {
"version": "1.0.0",
"resolved": "https://registry.npmjs.org/@epic-web/invariant/-/invariant-1.0.0.tgz",
"integrity": "sha512-lrTPqgvfFQtR/eY/qkIzp98OGdNJu0m5ji3q/nJI8v3SXkRKEnWiOxMmbvcSoAIzv/cGiuvRy57k4suKQSAdwA==",
"dev": true,
"license": "MIT"
},
"node_modules/@esbuild/aix-ppc64": {
"version": "0.27.7",
"resolved": "https://registry.npmjs.org/@esbuild/aix-ppc64/-/aix-ppc64-0.27.7.tgz",
@@ -2762,6 +2770,8 @@
},
"node_modules/@types/sortablejs": {
"version": "1.15.9",
"resolved": "https://registry.npmjs.org/@types/sortablejs/-/sortablejs-1.15.9.tgz",
"integrity": "sha512-7HP+rZGE2p886PKV9c9OJzLBI6BBJu1O7lJGYnPyG3fS4/duUCcngkNCjsLwIMV+WMqANe3tt4irrXHSIe68OQ==",
"dev": true,
"license": "MIT"
},
@@ -3273,6 +3283,24 @@
"version": "1.0.3",
"license": "MIT"
},
"node_modules/cross-env": {
"version": "10.1.0",
"resolved": "https://registry.npmjs.org/cross-env/-/cross-env-10.1.0.tgz",
"integrity": "sha512-GsYosgnACZTADcmEyJctkJIoqAhHjttw7RsFrVoJNXbsWWqaq6Ym+7kZjq6mS45O0jij6vtiReppKQEtqWy6Dw==",
"dev": true,
"license": "MIT",
"dependencies": {
"@epic-web/invariant": "^1.0.0",
"cross-spawn": "^7.0.6"
},
"bin": {
"cross-env": "dist/bin/cross-env.js",
"cross-env-shell": "dist/bin/cross-env-shell.js"
},
"engines": {
"node": ">=20"
}
},
"node_modules/cross-spawn": {
"version": "7.0.6",
"dev": true,
@@ -5599,6 +5627,8 @@
},
"node_modules/sortablejs": {
"version": "1.15.7",
"resolved": "https://registry.npmjs.org/sortablejs/-/sortablejs-1.15.7.tgz",
"integrity": "sha512-Kk8wLQPlS+yi1ZEf48a4+fzHa4yxjC30M/Sr2AnQu+f/MPwvvX9XjZ6OWejiz8crBsLwSq8GHqaxaET7u6ux0A==",
"license": "MIT"
},
"node_modules/source-map-js": {
+5 -3
View File
@@ -5,11 +5,12 @@
"scripts": {
"dev": "vite dev",
"build": "vite build",
"build:desktop": "BUILD_TARGET=desktop BASE_PATH='' vite build",
"build:desktop": "cross-env BUILD_TARGET=desktop BASE_PATH= vite build",
"prebuild": "npx tsx src/lib/scripts/pwa-manifest.ts",
"postbuild": "npx tsx src/lib/scripts/sitemap.ts",
"preview": "vite preview",
"check": "svelte-kit sync && svelte-check --tsconfig ./tsconfig.json",
"pretest": "svelte-kit sync",
"test": "node --import tsx --import ./src/test/register.mjs --experimental-wasm-modules --disable-warning=ExperimentalWarning --test \"src/**/*.test.ts\"",
"generate:example": "node --import tsx src/lib/scripts/example-statistics.ts",
"check:watch": "svelte-kit sync && svelte-check --tsconfig ./tsconfig.json --watch",
@@ -33,11 +34,12 @@
"@types/mapbox__tilebelt": "^1.0.4",
"@types/node": "^22.15.30",
"@types/sanitize-html": "^2.16.0",
"@types/sortablejs": "^1.15.8",
"@types/sortablejs": "^1.15.9",
"@typescript-eslint/eslint-plugin": "^8.33.1",
"@typescript-eslint/parser": "^8.33.1",
"bits-ui": "^2.17.2",
"clsx": "^2.1.1",
"cross-env": "^10.1.0",
"eslint": "^9.28.0",
"eslint-config-prettier": "^10.1.5",
"eslint-plugin-svelte": "^3.9.1",
@@ -80,6 +82,6 @@
"mapillary-js": "^4.1.2",
"maplibre-gl": "^6.15.0",
"sanitize-html": "^2.17.0",
"sortablejs": "^1.15.6"
"sortablejs": "^1.15.7"
}
}
+4 -4
View File
@@ -1,7 +1,7 @@
@import 'tailwindcss';
@import 'tw-animate-css';
@import "shadcn-svelte/tailwind.css";
@import "@fontsource-variable/inter";
@import 'shadcn-svelte/tailwind.css';
@import '@fontsource-variable/inter';
@custom-variant dark (&:is(.dark *));
@@ -158,7 +158,7 @@
body {
@apply bg-background text-foreground;
}
html {
@apply font-sans;
html {
@apply font-sans;
}
}
+5
View File
@@ -8,6 +8,11 @@ declare global {
// interface PageState {}
// interface Platform {}
}
interface Window {
/** The map, for extensions (set once it has loaded). */
_map?: import('maplibre-gl').Map;
}
}
export {};
+5 -5
View File
@@ -1,4 +1,4 @@
import { base } from '$app/paths';
import { asset, resolve as resolvePath } from '$app/paths';
import { languages } from '$lib/i18n/languages';
import { getURLForLanguage } from '$lib/utils';
@@ -37,16 +37,16 @@ export async function handle({ event, resolve }) {
<meta property="og:description" content="${description}" />
<meta name="twitter:title" content="gpx.studio — ${title}" />
<meta name="twitter:description" content="${description}" />
<meta property="og:image" content="https://gpx.studio${base}/og_logo.png" />
<meta property="og:image" content="https://gpx.studio${asset('/og_logo.png')}" />
<meta property="og:url" content="https://gpx.studio/" />
<meta property="og:type" content="website" />
<meta property="og:site_name" content="gpx.studio" />
<meta name="twitter:card" content="summary_large_image" />
<meta name="twitter:image" content="https://gpx.studio${base}/og_logo.png" />
<meta name="twitter:image" content="https://gpx.studio${asset('/og_logo.png')}" />
<meta name="twitter:url" content="https://gpx.studio/" />
<meta name="twitter:site" content="@gpxstudio" />
<meta name="twitter:creator" content="@gpxstudio" />
<link rel="alternate" hreflang="x-default" href="https://gpx.studio${getURLForLanguage('en', path)}" />`;
<link rel="alternate" hreflang="x-default" href="https://gpx.studio${resolvePath(getURLForLanguage('en', path))}" />`;
// the manifests are made for the website, not for the desktop app (`npm run build:desktop`)
if (process.env.BUILD_TARGET !== 'desktop') {
@@ -56,7 +56,7 @@ export async function handle({ event, resolve }) {
if (page !== '404') {
for (let lang of Object.keys(languages)) {
headTag += ` <link rel="alternate" hreflang="${lang}" href="https://gpx.studio${getURLForLanguage(lang, path)}" />
headTag += ` <link rel="alternate" hreflang="${lang}" href="https://gpx.studio${resolvePath(getURLForLanguage(lang, path))}" />
`;
}
}
+6
View File
@@ -20,6 +20,12 @@ export function createFile() {
currentTool.set(Tool.ROUTING);
}
/** Creates a file that starts with an empty route, which the routing tool then draws. */
export function createRouteFile() {
engine.newFile(newFileName(), undefined, true);
currentTool.set(Tool.ROUTING);
}
/** Asks for the files to open: with the file picker of the system in the desktop app. */
export async function openFiles() {
if (get(engine.workspace)) {
@@ -60,7 +60,7 @@
</script>
<svelte:head>
<link rel="preconnect" href="https://21XLD94PE3-dsn.algolia.net" crossorigin />
<link rel="preconnect" href="https://21XLD94PE3-dsn.algolia.net" crossorigin="anonymous" />
</svelte:head>
<div id="docsearch" class={props.class ?? ''}></div>
@@ -26,7 +26,7 @@
<Tooltip.Root>
<Tooltip.Trigger>
{#snippet child({ props })}
<Button {...props} {variant} class="bg-inherit {className}" {onclick}>
<Button {...props} {variant} {disabled} class="bg-inherit {className}" {onclick}>
{@render children()}
</Button>
{/snippet}
+4 -3
View File
@@ -1,4 +1,5 @@
<script lang="ts">
import { resolve } from '$app/paths';
import { Button } from '$lib/components/ui/button';
import LanguageSelect from '$lib/components/LanguageSelect.svelte';
import ModeSwitch from '$lib/components/ModeSwitch.svelte';
@@ -31,7 +32,7 @@
<Button
variant="link"
class="h-6 px-0 has-[>svg]:px-0 text-muted-foreground"
href={getURLForLanguage(i18n.lang, '/')}
href={resolve(getURLForLanguage(i18n.lang, '/'))}
>
<House size="16" />
{i18n._('homepage.home')}
@@ -40,7 +41,7 @@
data-sveltekit-reload
variant="link"
class="h-6 px-0 has-[>svg]:px-0 text-muted-foreground"
href={getURLForLanguage(i18n.lang, '/app')}
href={resolve(getURLForLanguage(i18n.lang, '/app'))}
>
<Map size="16" />
{i18n._('homepage.app')}
@@ -48,7 +49,7 @@
<Button
variant="link"
class="h-6 px-0 has-[>svg]:px-0 text-muted-foreground"
href={getURLForLanguage(i18n.lang, '/help')}
href={resolve(getURLForLanguage(i18n.lang, '/help'))}
>
<BookOpenText size="16" />
{i18n._('menu.help')}
+2 -1
View File
@@ -2,6 +2,7 @@
import { CircleQuestionMark } from '@lucide/svelte';
import { i18n } from '$lib/i18n/i18n.svelte';
import type { Snippet } from 'svelte';
import { resolve } from '$app/paths';
let {
link,
@@ -19,7 +20,7 @@
<div>
{@render children()}
{#if link}
<a href={link} target="_blank" class="text-[13px] text-link hover:underline">
<a href={resolve(link)} target="_blank" class="text-[13px] text-link hover:underline">
{i18n._('menu.more')}
</a>
{/if}
@@ -1,4 +1,5 @@
<script lang="ts">
import { resolve } from '$app/paths';
import { page } from '$app/state';
import * as Select from '$lib/components/ui/select';
import { languages } from '$lib/i18n/languages';
@@ -15,13 +16,13 @@
</span>
</Select.Trigger>
<Select.Content>
{#each Object.entries(languages) as [lang, label]}
{#each Object.entries(languages) as [lang, label] (lang)}
{#if page.url.pathname.includes('404')}
<a href={getURLForLanguage(lang, '/')}>
<a href={resolve(getURLForLanguage(lang, '/'))}>
<Select.Item value={lang}>{label}</Select.Item>
</a>
{:else}
<a href={getURLForLanguage(lang, page.url.pathname)}>
<a href={resolve(getURLForLanguage(lang, page.url.pathname))}>
<Select.Item value={lang}>{label}</Select.Item>
</a>
{/if}
+3 -3
View File
@@ -1,6 +1,6 @@
<script lang="ts">
import { mode } from 'mode-watcher';
import { base } from '$app/paths';
import { asset } from '$app/paths';
let {
iconOnly = false,
@@ -9,13 +9,13 @@
}: {
iconOnly?: boolean;
company?: 'gpx.studio' | 'github' | 'crowdin' | 'facebook' | 'reddit';
[key: string]: any;
[key: string]: unknown;
} = $props();
</script>
{#if company === 'gpx.studio'}
<img
src="{base}/{iconOnly ? 'icon' : 'logo'}{mode.current === 'dark' ? '-dark' : ''}.svg"
src={asset(`/${iconOnly ? 'icon' : 'logo'}${mode.current === 'dark' ? '-dark' : ''}.svg`)}
alt="Logo of gpx.studio."
{...others}
/>
+77 -32
View File
@@ -1,4 +1,5 @@
<script lang="ts">
import { resolve } from '$app/paths';
import * as Menubar from '$lib/components/ui/menubar/index.js';
import { Button } from '$lib/components/ui/button';
import Logo from '$lib/components/Logo.svelte';
@@ -48,6 +49,7 @@
Maximize2,
Minimize2,
FolderX,
ArrowRightLeft,
} from '@lucide/svelte';
import { map } from '$lib/components/map/map';
import { editMetadata } from '$lib/components/file-list/metadata/utils.svelte';
@@ -67,6 +69,7 @@
closeFiles,
closeFolder,
createFile,
createRouteFile,
loadFiles,
openFiles,
openFolder,
@@ -76,7 +79,7 @@
} from '$lib/actions/file-actions';
import UnsavedDialog from '$lib/components/file-list/UnsavedDialog.svelte';
import { engine } from '$lib/engine';
import { allHidden } from '$lib/engine/all-hidden';
import { allHidden, selectedFilesConversion } from '$lib/engine/selection-stores';
import { selectedFileIds, selectionSize } from '$lib/engine/selection-helpers';
import { boundsManager } from '$lib/state/bounds';
import { onMount } from 'svelte';
@@ -105,8 +108,10 @@
let desktop = $derived($workspace !== undefined);
let selectedIds = $derived(selectedFileIds($selection));
let anyUnsaved = $derived(Object.values($workspace?.files ?? {}).some((file) => file.dirty));
// metadata and style can be edited for files and tracks only
let filesOrTracksSelected = $derived($selection.type === 'file' || $selection.type === 'track');
// metadata and style can be edited for files, routes and tracks only
let filesOrTracksSelected = $derived(
$selection.type === 'file' || $selection.type === 'route' || $selection.type === 'track'
);
function switchBasemaps() {
[$currentBasemap, $previousBasemap] = [$previousBasemap, $currentBasemap];
@@ -144,7 +149,7 @@
<div
class="w-fit flex flex-row items-center justify-center p-1 bg-background rounded-b-md md:rounded-md pointer-events-auto shadow-md"
>
<a href={getURLForLanguage(i18n.lang, '/')} target="_blank" class="shrink-0">
<a href={resolve(getURLForLanguage(i18n.lang, '/'))} target="_blank" class="shrink-0">
<Logo class="h-5 mt-0.5 mx-2 md:hidden" iconOnly={true} width="16" />
<Logo class="h-5 mt-0.5 mx-2 hidden md:block" width="96" />
</a>
@@ -157,9 +162,13 @@
<Menubar.Content class="border-none">
<Menubar.Item onclick={createFile}>
<Plus size="16" />
{i18n._('menu.new')}
{i18n._('menu.new_file')}
<Shortcut key="+" ctrl={true} />
</Menubar.Item>
<Menubar.Item onclick={createRouteFile}>
<Plus size="16" />
{i18n._('menu.new_route_file')}
</Menubar.Item>
<Menubar.Separator />
<Menubar.Item onclick={openFiles}>
<FolderOpen size="16" />
@@ -285,19 +294,6 @@
<PaintBucket size="16" />
{i18n._('menu.style.button')}
</Menubar.Item>
<Menubar.Item
onclick={() => engine.setSelectionHidden(!$allHidden)}
disabled={selectionCount == 0}
>
{#if $allHidden}
<Eye size="16" />
{i18n._('menu.unhide')}
{:else}
<EyeOff size="16" />
{i18n._('menu.hide')}
{/if}
<Shortcut key="H" ctrl={true} />
</Menubar.Item>
{#if $treeFileView}
{#if $selection.type === 'file'}
<Menubar.Separator />
@@ -308,6 +304,13 @@
<Plus size="16" />
{i18n._('menu.new_track')}
</Menubar.Item>
<Menubar.Item
onclick={() => engine.newRoute()}
disabled={selectionCount !== 1}
>
<Plus size="16" />
{i18n._('menu.new_route')}
</Menubar.Item>
{:else if $selection.type === 'track'}
<Menubar.Separator />
<Menubar.Item
@@ -318,6 +321,32 @@
{i18n._('menu.new_segment')}
</Menubar.Item>
{/if}
{#if $selection.type === 'route' || $selection.type === 'track' || $selection.type === 'segment'}
{#if $selection.type !== 'track'}
<Menubar.Separator />
{/if}
<Menubar.Item onclick={() => engine.convert()}>
<ArrowRightLeft size="16" />
{i18n._(
$selection.type === 'route'
? 'menu.convert_to_track'
: 'menu.convert_to_route'
)}
</Menubar.Item>
{/if}
{/if}
{#if $selectedFilesConversion}
{#if !$treeFileView}
<Menubar.Separator />
{/if}
<Menubar.Item onclick={() => engine.convert()}>
<ArrowRightLeft size="16" />
{i18n._(
$selectedFilesConversion === 'route'
? 'menu.convert_to_routes'
: 'menu.convert_to_tracks'
)}
</Menubar.Item>
{/if}
<Menubar.Separator />
<Menubar.Item onclick={() => engine.selectAll()} disabled={noFiles}>
@@ -325,18 +354,6 @@
{i18n._('menu.select_all')}
<Shortcut key="A" ctrl={true} />
</Menubar.Item>
<Menubar.Item
onclick={() => {
if (selectionCount > 0) {
boundsManager.centerMapOnSelection();
}
}}
disabled={selectionCount == 0}
>
<Maximize size="16" />
{i18n._('menu.center')}
<Shortcut key="⏎" ctrl={true} />
</Menubar.Item>
{#if $treeFileView}
<Menubar.Separator />
<Menubar.Item onclick={() => engine.copy()} disabled={selectionCount === 0}>
@@ -401,6 +418,34 @@
/>
</Menubar.CheckboxItem>
<Menubar.Separator />
<Menubar.Item
inset
onclick={() => engine.setSelectionHidden(!$allHidden)}
disabled={selectionCount == 0}
>
{#if $allHidden}
<Eye size="16" />
{i18n._('menu.unhide')}
{:else}
<EyeOff size="16" />
{i18n._('menu.hide')}
{/if}
<Shortcut key="H" ctrl={true} />
</Menubar.Item>
<Menubar.Item
inset
onclick={() => {
if (selectionCount > 0) {
boundsManager.centerMapOnSelection();
}
}}
disabled={selectionCount == 0}
>
<Maximize size="16" />
{i18n._('menu.center')}
<Shortcut key="⏎" ctrl={true} />
</Menubar.Item>
<Menubar.Separator />
<Menubar.Item inset onclick={() => map.toggle3D()}>
<Box size="16" />
{i18n._('menu.toggle_3d')}
@@ -482,8 +527,8 @@
</Menubar.SubTrigger>
<Menubar.SubContent>
<Menubar.RadioGroup value={i18n.lang}>
{#each Object.entries(languages) as [lang, label]}
<a href={getURLForLanguage(lang, '/app')}>
{#each Object.entries(languages) as [lang, label] (lang)}
<a href={resolve(getURLForLanguage(lang, '/app'))}>
<Menubar.RadioItem value={lang}>{label}</Menubar.RadioItem>
</a>
{/each}
+5 -4
View File
@@ -1,4 +1,5 @@
<script lang="ts">
import { resolve } from '$app/paths';
import Logo from '$lib/components/Logo.svelte';
import { Button } from '$lib/components/ui/button';
import { BookOpenText, House, Map } from '@lucide/svelte';
@@ -9,7 +10,7 @@
<nav class="sticky top-0 w-full px-12 py-2 bg-background z-50 flex flex-col items-center border-b">
<div class="w-full max-w-5xl flex flex-row items-center gap-4 sm:gap-8">
<a
href={getURLForLanguage(i18n.lang, '/')}
href={resolve(getURLForLanguage(i18n.lang, '/'))}
class="shrink-0 translate-y-0.25 justify-self-start"
>
<Logo class="h-8 xs:hidden" iconOnly={true} width="26" />
@@ -18,7 +19,7 @@
<Button
variant="link"
class="text-base px-0 has-[>svg]:px-0 ml-auto"
href={getURLForLanguage(i18n.lang, '/')}
href={resolve(getURLForLanguage(i18n.lang, '/'))}
>
<House size="18" />
{i18n._('homepage.home')}
@@ -27,7 +28,7 @@
data-sveltekit-reload
variant="link"
class="text-base px-0 has-[>svg]:px-0"
href={getURLForLanguage(i18n.lang, '/app')}
href={resolve(getURLForLanguage(i18n.lang, '/app'))}
>
<Map size="18" />
{i18n._('homepage.app')}
@@ -35,7 +36,7 @@
<Button
variant="link"
class="text-base px-0 has-[>svg]:px-0"
href={getURLForLanguage(i18n.lang, '/help')}
href={resolve(getURLForLanguage(i18n.lang, '/help'))}
>
<BookOpenText size="18" />
{i18n._('menu.help')}
+3 -3
View File
@@ -39,7 +39,7 @@ export function getPreviousGuide(currentGuide: string): string | undefined {
return `${previousGuide}/${guides[previousGuide][guides[previousGuide].length - 1]}`;
}
} else {
if (guides.hasOwnProperty(subguides[0])) {
if (Object.hasOwn(guides, subguides[0])) {
const subguideIndex = guides[subguides[0]].indexOf(subguides[1]);
if (subguideIndex > 0) {
return `${subguides[0]}/${guides[subguides[0]][subguideIndex - 1]}`;
@@ -56,7 +56,7 @@ export function getNextGuide(currentGuide: string): string | undefined {
const subguides = currentGuide.split('/');
if (subguides.length === 1) {
if (guides.hasOwnProperty(currentGuide)) {
if (Object.hasOwn(guides, currentGuide)) {
if (guides[currentGuide].length === 0) {
const keys = Object.keys(guides);
const index = keys.indexOf(currentGuide);
@@ -68,7 +68,7 @@ export function getNextGuide(currentGuide: string): string | undefined {
return undefined;
}
} else {
if (guides.hasOwnProperty(subguides[0])) {
if (Object.hasOwn(guides, subguides[0])) {
const subguideIndex = guides[subguides[0]].indexOf(subguides[1]);
if (subguideIndex < guides[subguides[0]].length - 1) {
return `${subguides[0]}/${guides[subguides[0]][subguideIndex + 1]}`;
Loaded 100 of 518 files, more files were not shown because too many files have changed in this diff. Show more