Vladimir Kotal

I don't think so. That would be a global (or perhaps per project) tunable.

@ahornace recently mentioned having per user configuration for #3555 so perhaps one day it will be possible.

Then keep it open.

The foundation for per user settings is now in place via PR #3642.

Not at the moment. PRs are definitely welcome.

I can still see this problem on our production instance running 1.7.23 in bigger Mercurial repository that has renamed file handling enabled. The sync happens every 4 hours and usually...

Looking at the code in `get()`: https://github.com/oracle/opengrok/blob/110674a804dc9faa9a665dbac513689aaad3eca2/opengrok-indexer/src/main/java/org/opengrok/indexer/history/FileHistoryCache.java#L674-L686 makes me wonder whether the following situation could cause the problem (assuming Git or Mercurial repository): 1. repository is updated on disk (assuming...

Comparing the number of history entries for each cached file in a repo with some 42k files: a checkout that was recently reindexed from scratch had generally more history entries...

Looking at the actual history and history cache of some of the files where there is a discrepancy (i.e. they got reported by the above shell one-liner), I think the...

Another potentially notable issue is that the `RepositoryInfo` copy constructor was not copying the renamed file handling field. This was fixed via #3826.