You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The original partial arg code constructed a partial arg where each field
of the top-level command arg was either completely filled out or
completely missing. Before #2169, that was fine, because the top-level
commands were more specific, and thus their args had more detail and the
command id itself told you more. For example, in
`modifyTargetContainingScopeType`, you had a specific scope type that
was missing, but you knew from command id that it was "containing".
As of #2169, we do a lot more in the lower level captures. For example
`modifyTargetContainingScopeType` and `modifyTargetEveryScopeType` are
now just `modifyTarget`, which only has a `modifier` key at the top
level, so you don't know until the very end what you're dealing with,
because every modifier just looks like `{command: "modifyTarget",
partialArg: {modifier: undefined}}`.
This PR deeply reconstructs the partial target, so that you can get much
more information. See the new test cases for examples of these partial
arguments
- Depends on #2169
## Checklist
- [x] Switch to `CURRENT` for default arg
- [x] Remove unique queue file
- [x] I have added
[tests](https://www.cursorless.org/docs/contributing/test-case-recorder/)
- [-] I have updated the
[docs](https://github.com/cursorless-dev/cursorless/tree/main/docs) and
[cheatsheet](https://github.com/cursorless-dev/cursorless/tree/main/cursorless-talon/src/cheatsheet)
- [-] I have not broken the cheatsheet
0 commit comments