* On Windows, set name of Window Class, closes#7498
allow to customize it instead of current value hard coded "Window Class"
* feat(windows): add window_classname, closes#7498
allow to customize the window class name instead of current value hard coded "Window Class"
* feat: add window_classname, closes#7498
* add changes file
* Update core/tauri-config-schema/schema.json
* Update tooling/cli/schema.json
* missing pieces after merge
* clippy
---------
Co-authored-by: Géraud-Loup <47665233+geraudloup@users.noreply.github.com>
Co-authored-by: Lucas Nogueira <lucas@tauri.app>
* Fix mixup of `env_tauri_app_path()` and `env_tauri_frontend_path()` in tauri's path resolutions
* Rename functions in `app_paths` to match their corresponding, publicly exposed env var keys
* Rename `app_dir`/`app_path` variables that deal with the frontend app's directory to `frontend_dir
* Rename `APP_DIR` to `FRONTEND_DIR`
* Improve comment on meaning of tauri path env vars
* chore(deps) Update Rust crate image to v0.25.4
* Also bump json-patch and resvg
* Just json-patch for now
---------
Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>
Co-authored-by: Tony <legendmastertony@gmail.com>
* fix(runtime-wry): run `cursor_position` getter on main thread
closes#10340
* clippy
* clippy
---------
Co-authored-by: Lucas Nogueira <lucas@tauri.app>
* feat(core): add WebviewWindow::resolve_command_scope
This new functionality exposes the `CommandScope` resolution as a function (currently only commands can resolve them as a dependency injection via CommandItem)
This function is useful to validate the configuration at runtime (do some asserts at setup phase to ensure capabilities are properly configured) and to resolve scopes in a separate thread or context
* adjust return type
* feat(cli): add support for providing custom app/src paths to tauri's CLI via optional env vars
* fix tests
* rename env vars (app vs src is confusing)
* add change file
---------
Co-authored-by: Lucas Nogueira <lucas@tauri.app>
the audit failed, so the 2.0.2 release is failing. I'm also adding the latest merged change to the 2.0.2 release in this change so we're in sync in the changelog.
* chore: promote to v2 stable
- deletes all RC change files
- adds a new change file to promote all packages to v2 stable
- manually fix the tauri-driver, tauri-macos-sign, tauri-bundler versions so the next covector bump will move them to 2.0.0
- manually patch the metadata-v2.json file so the next covector update will mark all packages as 2.0.0
* ignore audit vuln without fixes
* bump msrv to 1.78
* run covector version
* fix sync lockfile covector
* #[allow(clippy::manual_inspect)]
* feat(cli): enhance Android dev port forwarding, closes#11137
this changes the `android dev` port forwarding (that is actually handled by the `android-studio-script` command - triggered by our Gradle plugin) with some enhancements:
- make the whole process more resilient by checking if the port was actually forwarded and rerunning the `adb reverse` command until it tells us the forward is ready
- if the `adb devices` list is empty, retry a few times (waiting a few seconds) to tolerate devices being booted - slows down "raw builds" (Build Project Android Studio menu for instance) that shouldn't happen often anyway - if you're running `android dev` you're usually running the app on a device instead of simply testing builds
* use host IP to run on android physical device
* fix(cli): iOS app signature not retaining entitlements, closes#11089
The IPA does not retain the entitlements as a regression from #10854 which removed the signing step from the build() and archive(), deferring to the export() call
To retain the entitlements we need to force sign one of the files in the app bundle. The most reliable way to do this is to use a self signed certificate as a dummy signature - it is replaced by the export() call so we do not rely on any user provided certificate
Additionally the export options are incorrectly configuring a manual signing, preventing Xcode from properly managing provisioning profiles, which is also part of the fix
* fix header
