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
Potentially optimize the ChangeSetFileElement how it applies and reverts changes, taking the monaco text model into account. I aimed at reusing the functionality to modify the monaco model if an editor is open without saving, and saving to disk directly if no editor is opened. But this should be tested for user experience and consistency.
The Coder Agent should be switched to use the monaco model, if it exists, to capture potentially dirty states -- otherwise the AI sees a different version than the user.
Show better progress while changes are generated. For example we could count the lines as they come in and do some vague guessing for the progress, e.g. assuming that the new file with be 150% the size of the old one and increasing the progress accordingly while making sure we slow down to not pass 100%.
The first time opening the diff editor can take a long time without any visual feedback
It proposed a change plan but then only did part of it, e.g. it suggested step 1, 2 and 3 and then only performed step 2
Newlines in strings must be especially handled. For example I had a string containing the literal \n and then the LLM suggestion wanted to replace this with an actual newline.
Could not reproduce this in a deterministic fashion: Often when I invoke the @coder multiple times in the same chat thread, it get's stuck randomly in the later calls, for example the change set tool functions were never called or never complete. Workaround was to open a new thread.
Close diff-editors / file-editors of discarded changes?
Feature Description:
@planger FYI, extracted from the PR.
The text was updated successfully, but these errors were encountered: