Brian Gernhardt <brian@xxxxxxxxxxxxxxxxxxxxx> writes:
> The --color display code in git-grep is giving me a bus error in
> show_line at line 492:
>
>> printf("%.*s%s%.*s%s",
>> match.rm_so, bol,
>> opt->color_match,
>> match.rm_eo - match.rm_so, bol +
>> match.rm_so,
>> GIT_COLOR_RESET);
>
> The first problem is that %.*s does not appear to do on OS X what the
> author thinks it does.
Hmm, that means that printf on OSX does not dohat the POSIX thinks it
ought to.
http://www.opengroup.org/onlinepubs/009695399/functions/fprintf.html
says that "a negative precision is taken as if the precision were
omitted"; it does not say "a negative or zero" here.
Which is a bit sad, because we would need to apply a workaround like
yours. We shouldn't have to.
> To fix that, I changed it to the following:
>
>> if( match.rm_so > 0 )
>> printf( "%.*s", match.rm_so, bol );
>> if( match.rm_eo > match.rm_so )
>> printf("%s%.*s%s",
>> opt->color_match,
>> match.rm_eo - match.rm_so, bol + match.rm_so,
>> GIT_COLOR_RESET);
> This code does not fail, but instead gives lines like the following
> (showing the raw color codes):
>
> .gitignore:\033[31m\033[1m(nugit
Hmm, that is strange. Your above change issues color_match and COLOR_RESET
only when you have something between rm_eo and rm_so. I do not see
anything between "ESC [ 31 m" and "ESC [ m" above, and you have an extra "1"
between "ESC [" and terminating "m" in the reset sequence.
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
