Re: algorithm of incremental updating

[Shigio Yamaguchi]

> This makes it impossible to process multiple files by one parser.
...
> Another thing to be consider is that parsers open GTAGS for reading
> while GTAGS is opened for modifying.
(derived from the next mail)

You are right. It's a good point.
OK. Let's change the verbose message according to your implementation
completely.

> I propose changing like this.
> 
> incremental(dbpath, cwd)
> {
>   if (!vflag
>       && ((comline = dbname(GTAGS)) != NULL
>           && locatestring(comline, "%S", MATCH_FIRST) != NULL)
>          || ((comline = dbname(GRTAGS)) != NULL
>           && locatestring(comline, "%S", MATCH_FIRST) != NULL)
>          || ((comline = dbname(GSYMS)) != NULL
>           && locatestring(comline, "%S", MATCH_FIRST) != NULL))
>     return incremental_optimized_for_speed(dbpath, cwd);

I don't agree. The -v(verbose) option should not change the behaveior.

The explanation to the users:
o The algorithm of incremental updating was changed to improve the
  performance. Along with it, the verbose message was greatly changed.

Is it right?

Thank you for your great hack.
--
Shigio YAMAGUCHI <address@hidden> - Tama Communications Corporation
PGP fingerprint: D1CB 0B89 B346 4AB6 5663  C4B6 3CA5 BBB3 57BE DDA3