Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.
I think this assumption easily breaks down on larger (maybe proprietary) code bases.
Yes, though the probability you are editing a file affected by this is probably low.
Perhaps we can improve by looking for common tags (tags for function declarations must match with the source file in question right?) but not necessarily in this PR.
That would be one possibility (though maybe a bit an overkill). I was thinking we could use a path-based heuristic assuming that related headers are stored closer to the corresponding sources on the file system. For instance in
/A/B/C1/D/foo.h
/A/B/C/D/foo.h
/A/B/C/D2/foo.c
the header for foo.c would be /A/B/C/D/foo.h because the identical path prefix (3 directories) is longer than for /A/B/C1/D/foo.h (2 directories).
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.