-
Notifications
You must be signed in to change notification settings - Fork 37
"Issue has already been reported" is not always true #82
Comments
Currently, issue equality is based on the title. If the titles are the same, then the issue is assumed to be the same. The problem is that |
One option could be to try and match stack traces, but that may be hard with different paths. Honestly though, this probably wont change anytime soon. |
That didn't seem to be the case with the |
Were there path differences in the title? |
Not in the truncated title. And it does use the truncated title in |
There seems to be a similar thing happening on this issue: file-icons/atom#476 As can be seen there, it's already closed, and from the other comments people seem to be redirected to an issue that's actually not the same and/or irrelevant. Just wanted to point it, if it can help to improve the behavior of the link-error-to-issue behavior. |
I've noticed this recently, and I can't find any related open issues (but forgive me if there is one). Sometimes while I'm developing something brand new on quantum-shell I'll run into an Uncaught TypeError and Atom will tell me that the issue has already been reported at sedabull/quantum-shell/issues and ask me if I would like to view it. This is obviously ridiculous, but not really a major issue. The reason I'm reporting it now is because I recently had a user open an issue (hoosierhobbyist/quantum-shell#17) that he believed was a repeat, but was in fact completely unrelated to the first two issues he mentioned. I understand this feature is meant to deter many people from reopening the same issue over and over, but as it stands now it may actually deter users from reporting an issue they believe is already being worked on. Let me know if there's anything I can do to help resolve this.
NOTE: was atom/atom#7475
The text was updated successfully, but these errors were encountered: