Stack overflow while generating tags completion table in emacs -

I am using GNU MX 23.3 on Windows. I work in a very large codebase for which I generate a TAGS file (using the atada binary provided with amatus). The tags file is too large (usually around 100 MB). I rarely have to use any functionality beyond find-tag , but there are times when I want to be able to complete the TAGS table.

Calling Full-Tag Emacs is the reason for creating the entire table automatically, the process takes a little while, but my problem is not in the amount of time, rather it The fact that at the end (about 100% complete), I have a stack overflow (sorry about countless characters)):

  entered the debugger - Lisp error: (error "in regexp matcher Stack overflow ") again search-forward (" ^ \\ (\\ ([^] + [^ - one-zA-Z0-9 _ + * $?] \\) \\ ([- -ZA-Z0 -9 _ + * $ ?:] \\) [^ - a-zA-Z0-9_ + * $?:] * \\) \\ (\\ ([^ \ n] \\) \\)? \\ ([0-9] + \\) ?, \\ ([0-9] + \ \)? \ N "zero t) - Tag-perfection-table () byte-code (...) Tag-Completion-Table ()   
 someone else has been run in this matter?  
  Edit:  Stack output after  debug-on-error      turned on  Edit:  Deleted Stack, because now I know how unsuccessful entries appear:  
  ^ lc: \ path \ to \ some \ header.h, 0 ^ LC: \ path \ to \ some \ otherheader.h, 0   
 There are a few entries in this format in my tag file. Looking at the included header, it is clear that they can not be parsed correctly by etags. It's ok, but I'm surprised that  tag-completions table  does not have an account in its regex for reference, for reference, here looks like a real entry:  
  ^ LC: \ path \ to \ some \ validheader.h, 115 squares class = CSomeClass? 12,345 bool SomeMethod (^? CSomeClass :: is used to match a tag entry inside regexp in the query; some method ^ A67,890    
 
 
  TAGS  file. I think the error could be if the file is formatted incorrectly (for example non-root Using line-ending), or if any entry is actually in fact, really big (an entry is usually a line or two, which is There should be no problem for gexp matures.)  
 One way to track the problem goes to the tag buffer and see where the point (cursor) is the error. Once you know that Which function is this, and you can live without tags without it, you can generate TAGS entries for it.  
 If the problem is due to a very complex entry, then I suggest that You want to send bug report to the ImaC team