Re: Minor bug: git config ignores empty sections

On Mon, Aug 15, 2016 at 2:55 PM, Jeff King <peff@peff.net> wrote:

So it sounds like removing an empty header is problematic in most cases, but adding a new variable to an existing empty header should not be...?

I looked at the code, and had a rough sketch that works as follows:

* Make git_parse_source() call the callback in a special way to note
  that a section header is seen (I hacked it by passing a special value,
  a pointer to a global string, as the second argument)

* Add a new store.last_section_offset field

* In store_aux(), if it's getting the special value, and the section
  name matches, save the offset in store.last_section_offset

* In git_config_set_multivar_in_file_gently() right before the "write
  the first part of the config" comment, test that
      (store.seen == 1 && copy_begin == 0 && copy_end == contents_sz
       && store.last_section_offset > 0)
  and if so, write the contents up to that point, and set copy_begin;
  if the condition is false, do the same thing it does now

* A bit after that, add "&& store.last_section_offset == 0" to the
  condition that decides whether to call store_write_section()

It looks to me like something like this can work, but it's very hacky because I wanted to see if it can work quickly, and because I don't know if this kind of a solution will be wanted, and because I don't know enough about that code in general. Making it be an actual solution involves a better way to call the callback in a special way (my hack does the special thing only if `fn==store_aux`, but it shouldn't), calling it after the last variable in a section is seen so it's added after that.

So, will something like this be acceptable? If so, is there anyone who I can ask questions about the code?