https://sourceware.org/bugzilla/show_bug.cgi?id=28762
--- Comment #7 from Pete Moore <petemoore at gmx dot net> --- This is great, many thanks Sergey! Now I'm curious, I noticed if I introduced changes on a completely different line, the problem would go away, e.g. if I made a change around line 1000, the problem report on line 10290 would disappear. But similarly, if I made a change around line 500, and left line 1000 as it is, it would also solve the issue. So now I'm wondering how the insertion of a line in a completely different (arbitrary) location in the parse tree caused the preprocessing at another location to be affected, since the sequences involving "a", "f", "'", EOL, etc may have been disturbed by one of the changes I made, but not the other. Does my question make sense? I'm really happy you found the solution, but I'm curious exactly how almost any arbitrary change anywhere in the first 10,290 lines resolved the issue. :-) Also many thanks for such a quick turnaround! -- You are receiving this mail because: You are on the CC list for the bug.
