Case insensitive match for completion items filtering #15949
Conversation
Missing a letter is allowed by matchSumOfSquares so it makes more sense to make filtering case insensitive.
|
Thanks for making a pull request to jupyterlab! |
|
Thanks for submitting your first pull request! You are awesome! 🤗 |
christophediprima
changed the title
There was a problem hiding this comment.
Thank you for the contribution @christophediprima.
It certainly does make sense to allow this as a setting. jupyterlab-lsp has a setting for it here. As whatever the default should be - opinions will vary.
Can you try to add the setting and unit tests for completion with this case sensitive and insensitive matchin on and off?
|
Sure but I am a bit busy right now. I will have a look when I come back from holidays! :) |
|
@christophediprima did you have a chance to take another look? It seems to have missed to 4.2 merge window, I will switch this PR to draft. |
| @@ -390,7 +390,7 @@ export class CompleterModel implements Completer.IModel { | |||
| index > -1 | |||
| ? originalItem.label.substring(0, index) | |||
| : originalItem.label; | |||
| const match = StringExt.matchSumOfSquares(escapeHTML(text), query); | |||
| const match = StringExt.matchSumOfSquares(escapeHTML(text.toLowerCase()), query.toLowerCase()); | |||
There was a problem hiding this comment.
I think using toLocaleLowerCase() instead of toLowerCase() might be a better call here. It tends to handle language-specific rules more gracefully, which could save us some headaches down the line. Plus, it aligns better with user expectations, especially in multilingual environments. Cheers!
Missing a letter is allowed by matchSumOfSquares so it makes more sense to make filtering case insensitive.