From mboxrd@z Thu Jan  1 00:00:00 1970
From: whydoubt at gmail.com (Jeffrey Smith)
Date: Sat, 21 Oct 2017 16:53:15 -0500
Subject: [PATCH 4/4] ui-blame: Allow syntax highlighting
In-Reply-To: <20171021144325.GF2393@john.keeping.me.uk>
References: <20171018041735.31592-1-whydoubt@gmail.com>
 <20171018041735.31592-5-whydoubt@gmail.com>
 <20171021144325.GF2393@john.keeping.me.uk>
Message-ID: <CAPX7N=7MRMpiUchLb2Qxq2kG3L_02GvPDnjjxrO_fTfXaXEkag@mail.gmail.com>

On Sat, Oct 21, 2017 at 9:43 AM, John Keeping <john at keeping.me.uk> wrote:
>
> This doesn't account for tabs, which is noticable in long lines (I
> happened to test with cgit.c where line 615 is quite a bit longer than
> average).
>
> It is fixed by using:
>
>         const char *start = blame_nth_line(sb, line);
>         const char *end = blame_nth_line(sb, line + 1);
>
>         html("\n");
>         len = end - start;
>         while (start < end)
>                 if (*(start++) == '\t')
>                         len += 7;
>

Nice catch.  Since '\t' is 'advance to next tab-stop' not 'advance # spaces',
that calculation could give a len larger than necessary.  I will add the code
to handle tabs properly.

>> +             if (len > maxlen)
>> +                     maxlen = len;
>> +     }
>>
>> -     html_ntxt(cp, cpend - cp);
>> +     for (len = 0; len < maxlen - 1; len++)
>> +             html(" ");
>
> Should this use &nbsp; or is a plain space guaranteed to be okay here?

This is inside of a 'pre' block, so spaces will do the right thing.