Still a WIP, but here's the deets so far:
## No more throwing canceled tokens
Throwing exceptions is expensive and since we essentially cancel tokens
anytime someone is typing beyond the debounce, we could be throwing
exceptions a ton during search. Since we don't care about those past
executions, now they just `return`.
## Reduced number of apps returned in search
While users can specify how many apps (no limit, 1, 5), if they specify
no limit, we hard limit it at 10. For a few reasons, fuzzy search gets
_really_ fuzzy sometimes and gives answers that users would think is
just plain wrong and they make the response list longer than it needs to
be.
## Fuzzy search: still fuzzy, but faster
Replaced `StringMatcher` class with `FuzzyStringMatcher`.
`FuzzyStringMatcher` is a C# port by @zadjii-msft of the Rust port by
@lhecker for [microsoft/edit](https://github.com/microsoft/edit), which
I believe originally came from [VS
Code](https://github.com/microsoft/vscode). It's a whole fuzzy rabbit
hole. But it's faster than the `StringMatcher` class it replaced.
## Fallbacks, you need to fall back
"In the beginning, fallbacks were created. This had made many people
very angry and has been widely regarded as a bad move."
Hitchhiker's Guide to the Galaxy jokes aside, fallbacks are one cause of
slower search results. A few modifications have been made to get them
out of the way without reverting their ability to do things dynamically.
1. Fallbacks are no longer scored and will always* appear at the bottom
of the search results
2. In updating their search text, we now use a cancellation token to
stop processing previous searches when a new keypress is recorded.
## * But Calculator & Run are special
So, remember when I said that all fallbacks will not be ranked and
always display at the bottom of the results? Surprise, some will be
ranked and displayed based on that score. Specifically, Calculator and
Run are fallbacks that are whitelisted from the restrictions mentioned
above. They will continue to act as they do today.
We do have the ability to add future fallbacks to that whitelist as
well.
---
## Current preview
Updated: 2025-09-24
https://github.com/user-attachments/assets/c74c9a8e-e438-4101-840b-1408d2acaefd
---
Closes#39763Closes#39239Closes#39948Closes#38594Closes#40330
## Summary of the Pull Request
Handles duplicate keybindings by using the first occurrence and ignoring
the rest (in `ContextMenuViewModel.Keybindings` and
`IContextMenuContext.Keybindings`).
Replaces LINQ with direct iteration for clarity.
Simplifies `CheckKeybinding` by removing redundant null checks and
clarifying the key-to-binding matching logic, improving both readability
and efficiency.
Add a new method to `KeyChordHelpers.FormatForDebug` that formats
KeyChord as string to help debugging.
Makes `KeyChordHelpers` class a static class.
<!-- Please review the items on the PR checklist before submitting-->
## PR Checklist
- [x] Closes: #41712
- [ ] **Communication:** I've discussed this with core contributors
already. If the work hasn't been agreed, this work might be rejected
- [x] **Tests:** all pass
- [x] **Localization:** All end-user-facing strings can be localized
- [x] **Dev docs:** Added/updated
- [x] **New binaries:** Added on the required places
- [ ] [JSON for
signing](https://github.com/microsoft/PowerToys/blob/main/.pipelines/ESRPSigning_core.json)
for new binaries
- [ ] [WXS for
installer](https://github.com/microsoft/PowerToys/blob/main/installer/PowerToysSetup/Product.wxs)
for new binaries and localization folder
- [ ] [YML for CI
pipeline](https://github.com/microsoft/PowerToys/blob/main/.pipelines/ci/templates/build-powertoys-steps.yml)
for new test projects
- [ ] [YML for signed
pipeline](https://github.com/microsoft/PowerToys/blob/main/.pipelines/release.yml)
- [x] **Documentation updated:** If checked, please file a pull request
on [our docs
repo](https://github.com/MicrosoftDocs/windows-uwp/tree/docs/hub/powertoys)
and link it here: #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
<!-- Describe how you validated the behavior. Add automated tests
wherever possible, but list manual validation steps taken as well -->
## Validation Steps Performed
Validated using a custom extension that has a duplicate item in the
context menu.
What the title says. 😄
Rather than relying on the potentially overloaded `!=` or `==` operators
when checking for null, now we'll use the `is` expression (possibly
combined with the `not` operator) to ensure correct checking. Probably
overkill for many of these classes, but decided to err on the side of
consistency. Would matter more on classes that may be inherited or
extended.
Using `is` and `is not` will provide us a guarantee that no
user-overloaded equality operators (`==`/`!=`) is invoked when a
`expression is null` is evaluated.
In code form, changed all instances of:
```c#
something != null
something == null
```
to:
```c#
something is not null
something is null
```
The one exception was checking null on a `KeyChord`. `KeyChord` is a
struct which is never null so VS will raise an error when trying this
versus just providing a warning when using `keyChord != null`. In
reality, we shouldn't do this check because it can't ever be null. In
the case of a `KeyChord` it **would** be a `KeyChord` equivalent to:
```c#
KeyChord keyChord = new ()
{
Modifiers = 0,
Vkey = 0,
ScanCode = 0
};
```
I just blindly moved all the messages. But _this_ one really makes more
sense as a UI message. It's got framework elements. It us used to
actually open a UI element. The whole thing is very UI specific.
re: #40113
_targets #40504_
Major refactoring for #40113
This moves a large swath of the codebase to a `.Core` project. "Core"
doesn't have any explicit dependencies on "extensions", settings or the
current `MainListPage`. It's just a filterable list of stuff. This
should let us make this component a bit more reusable.
This is half of a PR. As I did this, I noticed a particular bit of code
for TopLevelVViewModels and CommandPaletteHost that was _very rough_.
Solving it in this PR would make "move everything to a new project" much
harder to review. So I'm submitting two PRs simultaneously, so we can
see the changes separately, then merge together.
