On Mon 2019-06-24 19:43:58 -0700, William Casarin wrote:
>> +static bool
>> +_mime_node_set_up_part (mime_node_t *node, GMimeObject *part, int numchild);
>> +
>
> nit: Instead of a forward declaration, could _mime_node_create be moved after
> _mime_node_set_up_part instead?

yep, we could definitely do that.  I did it this way because the diff
feels cleaner to me -- rather than moving a big chunk of code, i'm just
breaking one function in half.

If anyone feels strongly about it, i wouldn't object to doing it the
other way around, but i'd be unlikely to want to rerun the patch series
for this change on its own.

       --dkg