for the record, i narrowed this down to the maven element of
`compilation-error-regexp-alist'. if i remove that element,
`compilation-mode' is fast on the minimal test case in
Sounds like Bug#13369, although that report specifically mentions the
'omake' element as the major culprit.
I know this isn't directly related to the original issue raised, and not to sound awfully negative...
But if I understand this correctly this is a bug which was filed and debugged in 2013, half a decade ago, with a known fix, but the bug is still in the Emacs codebase.
Why? Because whoever filed and diagnosed the bug in the first place can't/won't signed copyright assignment papers.
Now that the problem is solved, can't someone else just make the edit, write the patch, and make this terrible bug go away? It's a one-line diff after all. We do provide exceptions for such small patches after all.
But seeing bugs like this, makes me ask some other questions:
- Are there any systems in place to follow up "solved" bugs which doesn't result in commits?
- Do we know how many such bugs there currently are?
- And what improvements could we make to Emacs if there was an initiative to get all those issues properly closed?
--
Regards
Jostein Kjønigsen