This commit reduces the code surface that interacts with raw JSON data,
reducing code complexity and improving maintainability.
Files that needed to be changed drastically were additionally
cleaned up to remove any code cruft that has accrued over time.
In order to facility this the following changes were made:
* Move JSON handling from `CascadiaSettings` into `SettingsLoader`
This allows us to use STL containers for data model instances.
For instance profiles are now added to a hashmap for O(1) lookup.
* JSON parsing within `SettingsLoader` doesn't differentiate between user,
inbox and fragment JSON data, reducing code complexity and size.
It also centralizes common concerns, like profile deduplication and
ensuring that all profiles are assigned a GUID.
* Direct JSON modification, like the insertion of dynamic profiles into
settings.json were removed. This vastly reduces code complexity,
but unfortunately removes support for comments in JSON on first start.
* `ColorScheme`s cannot be layered. As such its `LayerJson` API was replaced
with `FromJson`, allowing us to remove JSON-based color scheme validation.
* `Profile`s used to test their wish to layer using `ShouldBeLayered`, which
was replaced with a GUID-based hashmap lookup on previously parsed profiles.
Further changes were made as improvements upon the previous changes:
* Compact the JSON files embedded binary, saving 28kB
* Prevent double-initialization of the color table in `ColorScheme`
* Making `til::color` getters `constexpr`, allow better optimizations
The result is a reduction of:
* 48kB binary size for the Settings.Model.dll
* 5-10% startup duration
* 26% code for the `CascadiaSettings` class
* 1% overall code in this project
Furthermore this results in the following breaking changes:
* The long deprecated "globals" settings object will not be detected and no
warning will be created during load.
* The initial creation of a new settings.json will not produce helpful comments.
Both cases are caused by the removal of manual JSON handling and the
move to representing the settings file with model objects instead
## PR Checklist
* [x] Closes#5276
* [x] Closes#7421
* [x] I work here
* [x] Tests added/passed
## Validation Steps Performed
* Out-of-box-experience is identical to before ✔️
(Except for the settings.json file lacking comments.)
* Existing user settings load correctly ✔️
* New WSL instances are added to user settings ✔️
* New fragments are added to user settings ✔️
* All profiles are assigned GUIDs ✔️
<!-- Enter a brief description/summary of your PR here. What does it fix/what does it change/how was it tested (even manually, if necessary)? -->
## Summary of the Pull Request
Adds directional modifiers for SplitState and convert those to the appropriate horizontal/vertical when splitting a pane.
<!-- Other than the issue solved, is this relevant to any other issues/existing PRs? -->
## References
<!-- Please review the items on the PR checklist before submitting-->
## PR Checklist
* [x] Closes#4340
* [x] CLA signed. If not, go over [here](https://cla.opensource.microsoft.com/microsoft/Terminal) and sign the CLA
* [ ] Tests added/passed
* [x] Documentation updated. If checked, please file a pull request on [our docs repo](https://github.com/MicrosoftDocs/terminal) and link it here: #xxx
* [x] Schema updated.
* [ ] I've discussed this with core contributors already. If not checked, I'm ready to accept this work might be rejected in favor of a different grand plan. Issue number where discussion took place: #xxx
<!-- Provide a more detailed description of the PR, other things fixed or any additional comments/features here -->
## Detailed Description of the Pull Request / Additional comments
"vertical" and "horizontal" splits were removed from `defaults.json`, but code was added to parse those as `right` and `down` respectively. It is also the case that if a user has a custom hotkey for `split: vertical` it will override the default for `split: right`.
<!-- Describe how you validated the behavior. Add automated tests wherever possible, but list manual validation steps taken as well -->
## Validation Steps Performed
Split the pane using each of the new directional movements
This commit introduces an alternative to specifying key bindings as a combination of key modifiers and a character. It allows you to specify an explicit virtual key as `vk(nnn)`.
Additionally this commit makes it possible to bind actions to scan codes. As scan code 41 appears to be the button below the Escape key on virtually all keyboards, we'll be able to bind the quake mode hotkey to `win+sc(41)` and have it work consistently across most if not all keyboard layouts.
## PR Checklist
* [x] Closes#7539, Closes#10203
* [x] I work here
* [x] Tests added/passed
## Validation Steps Performed
The following was tested both on US and DE keyboard layouts:
* Ctrl+, opens settings ✔️
* Win+` opens quake mode window ✔️
* Ctrl+plus/minus increase/decrease font size ✔️
Replaces `KeyModifiers` with the pretty much equivalent
`VirtualKeyModifiers` enum in winrt.
After doing this I noticed #10593 which changes the KeyChords a lot, but
it seems these PRs are still compatible
The issue also mentions replacing Vkey with
`Windows::System::VirtualKey`, but I chose not to because that enum only
includes a subset of the keys terminal supports here (no VK_OEM_* keys)
## Validation Steps Performed
Changed key bind in config, and confirmed it still works after
restarting terminal
Closes#877
This entirely removes `KeyMapping` from the settings model, and builds on the work done in #9543 to consolidate all actions (key bindings and commands) into a unified data structure (`ActionMap`).
## References
#9428 - Spec
#6900 - Actions page
Closes#7441
## Detailed Description of the Pull Request / Additional comments
The important thing here is to remember that we're shifting our philosophy of how to interact/represent actions. Prior to this, the actions arrays in the JSON would be deserialized twice: once for key bindings, and again for commands. By thinking of every entry in the relevant JSON as a `Command`, we can remove a lot of the context switching between working with a key binding vs a command palette item.
#9543 allows us to make that shift. Given the work in that PR, we can now deserialize all of the relevant information from each JSON action item. This allows us to simplify `ActionMap::FromJson` to simply iterate over each JSON action item, deserialize it, and add it to our `ActionMap`.
Internally, our `ActionMap` operates as discussed in #9428 by maintaining a `_KeyMap` that points to an action ID, and using that action ID to retrieve the `Command` from the `_ActionMap`. Adding actions to the `ActionMap` automatically accounts for name/key-chord collisions. A `NameMap` can be constructed when requested; this is for the Command Palette.
Querying the `ActionMap` is fairly straightforward. Helper functions were needed to be able to distinguish an explicit unbinding vs the command not being found in the current layer. Internally, we store explicitly unbound names/key-chords as `ShortcutAction::Invalid` commands. However, we return `nullptr` when a query points to an unbound command. This is done to hide this complexity away from any caller.
The command palette still needs special handling for nested and iterable commands. Thankfully, the expansion of iterable commands is performed on an `IMapView`, so we can just expose `NameMap` as a consolidation of `ActionMap`'s `NameMap` with its parents. The same can be said for exposing key chords in nested commands.
## Validation Steps Performed
All local tests pass.
This pull request adds an appearance configuration object to our
settings model and app lib, allowing the control to be rendered
differently depending on its state, and then uses it to add support for
an "unfocused" appearance that the terminal will use when it's not in
focus.
To accomplish this, we isolated the appearance-related settings from
Profile (into AppearanceConfig) and TerminalSettings (into the
IControlAppearance and ICoreAppearance interfaces). A bunch of work was
done to make inheritance work.
The unfocused appearance inherits from the focused one _for that
profile_. This is important: If you define a
defaults.unfocusedAppearance, it will apply all of defaults' settings to
any leaf profile when a terminal in that profile is out of focus.
Specified in #8345Closes#3062Closes#2316
**BE NOT AFRAID**. I know that there's 107 files in this PR, but almost
all of it is just find/replacing `TerminalControl` with `Control`.
This is the start of the work to move TermControl into multiple pieces,
for #5000. The PR starts this work by:
* Splits `TerminalControl` into separate lib and dll projects. We'll
want control tests in the future, and for that, we'll need a lib.
* Moves `ICoreSettings` back into the `Microsoft.Terminal.Core`
namespace. We'll have other types in there soon too.
* I could not tell you why this works suddenly. New VS versions? New
cppwinrt version? Maybe we're just better at dealing with mdmerge
bugs these days.
* RENAMES `Microsoft.Terminal.TerminalControl` to
`Microsoft.Terminal.Control`. This touches pretty much every file in
the sln. Sorry about that (not sorry).
An upcoming PR will move much of the logic in TermControl into a new
`ControlCore` class that we'll add in `Microsoft.Terminal.Core`.
`ControlCore` will then be unittest-able in the
`UnitTests_TerminalCore`, which will help prevent regressions like #9455
## Detailed Description of the Pull Request / Additional comments
You're really gonna want to clean the sln first, then merge this into
your branch, then rebuild. It's very likely that old winmds will get
left behind. If you see something like
```
Error MDM2007 Cannot create type
Microsoft.Terminal.TerminalControl.KeyModifiers in read-only metadata
file Microsoft.Terminal.TerminalControl.
```
then that's what happened to you.
Currently, when loading command with sub-commands that fail to parse,
we result with command that:
* Is not considered nested (has no sub-commands)
* Has no action of its own
The commit contains a few changes:
1. Protection in the dispatch that will prevent NPE
2. Change in the command parsing that will no load
a command if all its sub-commands failed to parse
3. We will add a warning in this case (the solution is somewhat
hacky, due to the hack that was there previously)
When such command is passed to a dispatch we crash with NPE.
Closes#9448
Show a validation warning when someone sets a `setColorScheme` action
with an invalid scheme
In the setting validation phase, scan all commands for all the "set
color scheme" actions, and check each of them has a valid scheme. If any
of them has an invalid scheme name, raise a warning. Do not check
iterable commands that will be expanded to valid color schemes.
## Validation Steps Performed
- Added tests to LocalTests_SettingsModel
- Manual tests, add commands to settings.json with invalid color scheme
and check the warning pops up. Try simple and nested commands.
Closes#7221
The Settings UI exposes the `profiles.defaults` (PD) object. Today, we
remove PD if there's nothing in it. However, that causes problems with
the Settings UI, because we have no `Profile` object to bind to
(resulting in a crash). Rather than making the Settings UI create a PD,
and link it in the inheritance tree, it's much easier to just _always_
create and link the PD object.
## References
#1564 - Settings UI (fixes a crash for this)
#7923 - Introduces inheritance
## PR Checklist
* [X] Tests added/passed
## Validation Steps Performed
* [x] repro steps for crash in Settings UI (copied changes over to that
branch for testing)
* [x] tests passed
## Summary of the Pull Request
This adds `ToJson` functions to `Profile`, `GlobalAppSettings`, and `ColorScheme`. They are used in `CascadiaSettings` to completely serialize an instance of the settings model. Thanks to #7923, all of the settings are `std::optional`, and `JsonUtils` only writes out values that are actually populated.
`CascadiaSettings::WriteSettingsToDisk` serializes the current settings and writes them to the settings.json. A backup file is created with your old contents.
#### Limitations:
- all of the color schemes are serialized regardless of them coming from defaults.json or settings.json
- keybindings/actions are copied/pasted
## References
#1564 - Settings UI
TSM Specs (#6904 and #7876)
## PR Checklist
* [x] Tests added/passed
## Summary of the Pull Request
When we get a serialization error, we "catch" it in `AppLogic` and only
`LoadDefaults()`. Since `LoadDefaults()` doesn't perform a full
validation of `CascadiaSettings`, we need to manually update our list of
active profiles (similar to how we manually resolve the default
profile).
## Validation Steps Performed
Repro steps fixed:
1. add deserialization error to settings.json (i.e. "fontWeight": "wumbo")
2. launch WT
3. verify that dropdown is populated with active profiles
Closes#8146
## Summary of the Pull Request
This PR replaces `CascadiaSettings::_profiles` with...
- `_allProfiles`: the list of all available profiles in the settings model (i.e. settings.json, dynamic profiles, etc...)
- `_activeProfiles`: the list of all non-hidden profiles (used for the new tab dropdown)
## References
#8018: maintaining a list of all profiles allows us to serialize hidden profiles
#1564: Settings UI can link to `AllProfiles()` instead of `ActiveProfiles()` to expose hidden profiles
## PR Checklist
* [x] Closes#4139
* [x] Tests added/passed
## Validation Steps Performed
Deploy and testing succeeded
## Summary of the Pull Request
Introduces `IInheritable` as an interface that helps move cascading settings into the Terminal Settings Model. `GlobalAppSettings` and `Profile` both are now `IInheritable`. `CascadiaSettings` was updated to `CreateChild()` for globals and each profile when we are loading the JSON data.
IInheritable does most of the heavy lifting. It introduces a two new macros and the interface. The macros help implement the fallback functionality for nullable and non-nullable settings.
## References
#7876 - Spec Addendum
#6904 - TSM Spec
#1564 - Settings UI
#7876 - `Copy()` needs to be updated to include _parent
## Summary of the Pull Request
This implements the `Copy` function for `CascadiaSettings`. Copy performs a deep copy of a `CascadiaSettings` object. This is needed for data binding in the Terminal Settings Editor.
The `Copy` function was basically implemented in every settings model object. This was mostly just repetitive work.
## References
#7667 - TSM
#1564 - Settings UI
## PR Checklist
* [X] Tests added/passed
## Summary of the Pull Request
Added watch on desktopImagePath to check when the path equals "DesktopWallpaper"
If it does equal "DesktopWallpaper" it replaces the path with a path to the desktop's wallpaper
*I am a student and this is my first pull request for Terminal so please give feedback no matter how small. It's the best way I can learn.
## PR Checklist
* [X] Closes#7295
* [X] CLA signed. If not, go over [here](https://cla.opensource.microsoft.com/microsoft/Terminal) and sign the CLA
* [?] Tests added/passed
* [X] Documentation updated. If checked, please file a pull request on [our docs repo](https://github.com/MicrosoftDocs/terminal) and link it here: https://github.com/MicrosoftDocs/terminal/pull/155
* [?] Schema updated. (Not sure if this is needed, also not sure where this would be)
* [X] I've discussed this with core contributors already. If not checked, I'm ready to accept this work might be rejected in favor of a different grand plan. Issue number where discussion took place: #7295 (Have only talked with the people on the issue, which I don't think has any core contributors)
## Detailed Description of the Pull Request / Additional comments
I am using SystemParametersInfo for SPI_GETDESKWALLPAPER which puts the path into a WCHAR and that is then inserted as the BackgroundImagePath.
I do not think an additional test would add value. The SPI_GETDESKTOPWALLPAPER uses the computers local wallpaper path and puts it into a WCHAR, which then I feed into BackgroundImagePath() as it's new path. I don't think there adds value in making a static path of the desktop background and testing that, given that static tests are already done for "BackgroundImage()".
## Validation Steps Performed
(Manual Validation - Test False Value)
1. Ran Terminal
2. Set setting ["backgroundImage": "<some random img path>"] under profiles->defaults
3. Verified terminal's background is not the desktops wallpaper.
(Manual Validation - Test True Value)
1. Ran Terminal
2. Set setting ["backgroundImage": "DesktopWallpaper"] under profiles->defaults
3. Verified the background image matches the desktop background image.
(Manual Validation - Multiple Tabs True Value)
1. Ran Terminal
2. Set setting ["backgroundImage": "DesktopWallpaper"] under profiles->defaults
3. Verified the background image matches the desktop background image.
4. Opened new tabs
5. Verified the background image matches the desktop background image for each tab.
## Summary of the Pull Request
Introduce the `IconPathConverter` to `TerminalApp`. `Command` and `Profile` now both return the unexpanded icon path. `IconPathConverter` is responsible for expanding the icon path and retrieving the appropriate icon source.
This also removes `Profile`'s expanded icon path and uses the `IconPathConverter` when necessary. This allows users to set profile icons to emoji as well. However, emoji do not appear in the jumplist.
## References
Based on #7667
## PR Checklist
* [X] Closes#7784
* [x] Documentation updated. If checked, please file a pull request on [our docs repo](https://github.com/MicrosoftDocs/terminal) and link it here: #xxx
* [x] Schema updated.
## Validation Steps Performed
Deploy succeeded.
Introduces a new TerminalSettingsModel (TSM) project. This project is
responsible for (de)serializing and exposing Windows Terminal's settings
as WinRT objects.
## References
#885: TSM epic
#1564: Settings UI is dependent on this for data binding and settings access
#6904: TSM Spec
In the process of ripping out TSM from TerminalApp, a few other changes
were made to make this possible:
1. AppLogic's `ApplicationDisplayName` and `ApplicationVersion` was
moved to `CascadiaSettings`
- These are defined as static functions. They also no longer check if
`AppLogic::Current()` is nullptr.
2. `enum LaunchMode` was moved from TerminalApp to TSM
3. `AzureConnectionType` and `TelnetConnectionType` were moved from the
profile generators to their respective TerminalConnections
4. CascadiaSettings' `SettingsPath` and `DefaultSettingsPath` are
exposed as `hstring` instead of `std::filesystem::path`
5. `Command::ExpandCommands()` was exposed via the IDL
- This required some of the warnings to be saved to an `IVector`
instead of `std::vector`, among some other small changes.
6. The localization resources had to be split into two halves.
- Resource file linked in init.cpp. Verified at runtime thanks to the
StaticResourceLoader.
7. Added constructors to some `ActionArgs`
8. Utils.h/cpp were moved to `cascadia/inc`. `JsonKey()` was moved to
`JsonUtils`. Both TermApp and TSM need access to Utils.h/cpp.
A large amount of work includes moving to the new namespace
(`TerminalApp` --> `Microsoft::Terminal::Settings::Model`).
Fixing the tests had its own complications. Testing required us to split
up TSM into a DLL and LIB, similar to TermApp. Discussion on creating a
non-local test variant can be found in #7743.
Closes#885
