Summary: | Disabling history doesn't disable memory of last-typed search | ||
---|---|---|---|
Product: | [Plasma] krunner | Reporter: | Andrew Brouwers <abrouwers> |
Component: | general | Assignee: | Plasma Bugs List <plasma-bugs-null> |
Status: | CONFIRMED --- | ||
Severity: | minor | CC: | alexander.lohnau, dashonwwIII, natalie_clarius, nate, nicolas.fella, xnaxdy |
Priority: | NOR | Keywords: | usability |
Version First Reported In: | 6.0.0 | ||
Target Milestone: | --- | ||
Platform: | Other | ||
OS: | Linux | ||
Latest Commit: | Version Fixed In: | ||
Sentry Crash Report: | |||
Attachments: |
krunner
konsole launched, while "kons" retained in a future krunner instance |
Description
Andrew Brouwers
2024-03-01 21:37:35 UTC
Hmm, I am not sure how to handle this. It is not any persistent history, but just your last typed query. In case a window opens and KRunner closes, you might otherwise miss your previously typed text. Though you should be able to override the text with just one keystroke. A sensible differentiation might be the cause for closing KRunner. Like the escape key could be handled separately and clear the text when a match is run. Though I am not sure if this behavior seems consistent to the average user. (In reply to Alexander Lohnau from comment #1) > Hmm, I am not sure how to handle this. It is not any persistent history, but > just your last typed query. In case a window opens and KRunner closes, you > might otherwise miss your previously typed text. Though you should be able > to override the text with just one keystroke. > > A sensible differentiation might be the cause for closing KRunner. Like the > escape key could be handled separately and clear the text when a match is > run. Though I am not sure if this behavior seems consistent to the average > user. I should clarify, even if I type "kons," hit "enter" once konsole is highlighted (ie, it didn't get interrupted, but rather successfully launched my searched-for app), it *still* shows in my history. The history setting just seems broken in plasma 6. Created attachment 166311 [details]
konsole launched, while "kons" retained in a future krunner instance
Currently this behavior is intentional, but I can see how it's somewhat confusing: disabling history doesn't disable *recent* history, only long term history. Maybe we need another setting for this:
> History: ( ) Disable
> ( ) Only last-typed search
> ( ) Suggestions from all past searches
> (o) Auto-complete from all past searches
(In reply to Nate Graham from comment #4) > Currently this behavior is intentional, but I can see how it's somewhat > confusing: disabling history doesn't disable *recent* history, only long > term history. Maybe we need another setting for this: > > > History: ( ) Disable > > ( ) Only last-typed search > > ( ) Suggestions from all past searches > > (o) Auto-complete from all past searches Yeah, I would expect disable history to start with a blank input, every time. (In reply to Nate Graham from comment #4) > Currently this behavior is intentional, but I can see how it's somewhat > confusing: disabling history doesn't disable *recent* history, only long > term history. Maybe we need another setting for this: > > > History: ( ) Disable > > ( ) Only last-typed search > > ( ) Suggestions from all past searches > > (o) Auto-complete from all past searches Retaining history and retaining prior search are actually two different pieces of functionality. See the implementation of the former here: https://invent.kde.org/plasma/plasma-workspace/-/merge_requests/226 If memory serves right, there used to be a checkbox that allowed you to toggle just that specific setting. In fact, if you manually add [General] RetainPriorSearch=false to your ~/.config/krunnerrc , KRunner will no longer retain your previous searches, as the author of this bug report desires. So, for me, the two takeaways are: 1. The current behavior for retaining search history is indeed intentional 2. The actual bug lies within removing the checkbox that allowed users to toggle this functionality Correction on my previous comment: I meant "implementation of the latter", of course... |