Re: What's cooking in git.git (Oct 2020, #03; Mon, 19)

On Fri, Oct 23, 2020 at 1:02 AM Jeff King [off-list ref] wrote:

On Mon, Oct 19, 2020 at 11:13:22AM -0700, Junio C Hamano wrote:

quoted

* en/strmap (2020-10-13) 10 commits
 - strmap: enable allocations to come from a mem_pool
 - strmap: add a strset sub-type
 - strmap: add functions facilitating use as a string->int map
 - strmap: enable faster clearing and reusing of strmaps
 - strmap: add more utility functions
 - strmap: new utility functions
 - hashmap: introduce a new hashmap_partial_clear()
 - hashmap: allow re-use after hashmap_free()
 - hashmap: adjust spacing to fix argument alignment
 - hashmap: add usage documentation explaining hashmap_free[_entries]()

 A speciailization of hashmap that uses a string as key has been
 introduced.  Hopefully it will see wider use over time.

 Will merge to 'next'.

I haven't gotten a chance to look at this carefully yet, but note that
the strset here sort-of conflicts with the one I added recently in
builtin/shortlog.c (already in master). I think the compiler won't
complain because they never show up in the same translation unit, but we
should avoid having two almost-the-same types in the codebase.

Makes sense.

I suspect my "check_and_add" function could be folded into Elijah's
implementation. The other big difference is that mine uses the
FLEX_ALLOC approach, and his doesn't. I haven't digested the code and
discussion around that from Elijah's series yet.

From a brief glance, it looks like the code in shortlog would probably
be easy to convert over.  But more important is making sure we have a
strmap/strset/strintmap API that we like.  You provided a lot of good
feedback and pointers on the first round of the series, so if you
could find some time to take a look at the second round, it'd be very
appreciated.  I'm also very interested in your opinion on whether the
changes I made were good enough, or if you'd like me to dig further
into whether going the FLEX_ALLOC route will work.