Wait, if we don't need to handle leading whitespaces (which we don't need AFAIK, although it might be nice), the logic is merely if (*line && *line != '[' && *line != '#') comment();, isn't it?

Yes, but I also think we shouldn't add # on empty lines or lines containing only whitespaces.

I could modify the code to use sci_get_line() and work on lines instead of individual characters, it just mens allocating/deallocating strings for each line (not that it matters much performance-wise though). Not sure which option is better here.

It doesn't matter, you don't have to switch to fetching lines to implement either, my "code" was merely laying out my thoughts, not really a suggestion of implementation per se.

And it's be easy enough to support all-whitespace lines if you really want to, just do something like (again, this has not to be the actual implementation):

for (line in document) { // see, pseudo-code :)
    const char *p = line;
    while (*p && isspace(*p))
        p++;
    if (! *p) continue; // empty line
    if (*p == '[') continue; // section, leave it alone
    if (*line == '#') continue; // already commented at column 0, leave it alone
    comment(&line);
}

which, given you have all the info, simplifies to if (*p && *p != '[' && *line != '#') comment().

—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.