From: Taylor Blau <me@ttaylorr.com>
To: Jeff King <peff@peff.net>
Cc: git@vger.kernel.org, Junio C Hamano <gitster@pobox.com>
Subject: Re: [PATCH 1/5] string-list: introduce `string_list_split_in_place_multi()`
Date: Tue, 18 Apr 2023 13:08:37 -0400 [thread overview]
Message-ID: <ZD7OlWjApwvE8Aih@nand.local> (raw)
In-Reply-To: <20230418101058.GA508219@coredump.intra.peff.net>
On Tue, Apr 18, 2023 at 06:10:58AM -0400, Jeff King wrote:
> On Thu, Apr 13, 2023 at 07:31:43PM -0400, Taylor Blau wrote:
>
> > Instead of using `strchr(2)` to locate the first occurrence of the given
> > delimiter character, `string_list_split_in_place_multi()` uses
> > `strpbrk(2)` to find the first occurrence of *any* character in the given
> > delimiter string.
> >
> > Since the `_multi` variant is a generalization of the original
> > implementation, reimplement `string_list_split_in_place()` in terms of
> > the more general function by providing a single-character string for the
> > list of accepted delimiters.
>
> I'd imagine that strpbrk() is potentially a lot slower than strchr().
> But I kind of doubt that splitting is such a hot path that it will
> matter. If we want to care, I think we could do something like:
>
> end = delim[1] ? strpbrk(p, delim) : strchr(p, delim[0]);
>
> It's entirely possible that a half-decent strpbrk() implementation does
> this already.
I did a little research here, and it looks like both glibc[1] and
musl[2] already do this. Those links are for their implementations of
strcspn(), but which both implement[^1] strpbrk() as "p += strcspn(p,
d)".
> So I mention it mostly in case we need to revisit this later. I think
> it's OK to ignore for now.
In addition to being OK from a performance standpoint I think solving
the semantics concern you noted lower down in the thread[3] is fairly
straightforward.
When in "_multi" mode, I think this boils down to moving past any number
of characters in `delim` before each iteration of the loop. And that's
doable with "p += strspn(p, delim)".
That avoids any behavior change, and more closely matches what strtok()
does, so I think it's a better path.
Thanks,
Taylor
[1]: https://sourceware.org/git/?p=glibc.git;a=blob;f=string/strcspn.c;hb=glibc-2.37#l35
[2]: https://git.musl-libc.org/cgit/musl/tree/src/string/strcspn.c?h=v1.2.3#n11
[3]: https://lore.kernel.org/git/20230418102320.GB508219@coredump.intra.peff.net/
[^1]: Not entirely. If strcspn() takes you to the end of the string,
strpbrk() will return NULL, instead of a pointer to the end of the
string.
next prev parent reply other threads:[~2023-04-18 17:09 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-04-13 23:31 [PATCH 0/5] banned: mark `strok()`, `strtok_r()` as banned Taylor Blau
2023-04-13 23:31 ` [PATCH 1/5] string-list: introduce `string_list_split_in_place_multi()` Taylor Blau
2023-04-18 10:10 ` Jeff King
2023-04-18 17:08 ` Taylor Blau [this message]
2023-04-13 23:31 ` [PATCH 2/5] t/helper/test-hashmap.c: avoid using `strtok()` Taylor Blau
2023-04-18 10:23 ` Jeff King
2023-04-18 18:06 ` Taylor Blau
2023-04-13 23:31 ` [PATCH 3/5] t/helper/test-oidmap.c: " Taylor Blau
2023-04-13 23:31 ` [PATCH 4/5] t/helper/test-json-writer.c: " Taylor Blau
2023-04-13 23:31 ` [PATCH 5/5] banned.h: mark `strtok()`, `strtok_r()` as banned Taylor Blau
2023-04-14 1:39 ` Junio C Hamano
2023-04-14 2:08 ` Chris Torek
2023-04-14 13:41 ` Taylor Blau
2023-04-18 19:18 ` [PATCH v2 0/6] banned: mark `strok()` " Taylor Blau
2023-04-18 19:18 ` [PATCH v2 1/6] string-list: introduce `string_list_split_in_place_multi()` Taylor Blau
2023-04-18 19:39 ` Junio C Hamano
2023-04-18 20:54 ` Taylor Blau
2023-04-22 11:12 ` Jeff King
2023-04-22 15:53 ` René Scharfe
2023-04-23 0:35 ` Jeff King
2023-04-24 16:24 ` Junio C Hamano
2023-04-23 2:38 ` [PATCH v2 1/6] string-list: introduce `string_list_split_in_place_multi()`t Taylor Blau
2023-04-23 2:40 ` Taylor Blau
2023-04-18 19:18 ` [PATCH v2 2/6] string-list: introduce `string_list_setlen()` Taylor Blau
2023-04-22 11:14 ` Jeff King
2023-04-18 19:18 ` [PATCH v2 3/6] t/helper/test-hashmap.c: avoid using `strtok()` Taylor Blau
2023-04-22 11:16 ` Jeff King
2023-04-24 21:19 ` Taylor Blau
2023-04-18 19:18 ` [PATCH v2 4/6] t/helper/test-oidmap.c: " Taylor Blau
2023-04-18 19:18 ` [PATCH v2 5/6] t/helper/test-json-writer.c: " Taylor Blau
2023-04-18 19:18 ` [PATCH v2 6/6] banned.h: mark `strtok()` as banned Taylor Blau
2023-04-24 22:20 ` [PATCH v3 0/6] banned: mark `strok()`, `strtok_r()` " Taylor Blau
2023-04-24 22:20 ` [PATCH v3 1/6] string-list: multi-delimiter `string_list_split_in_place()` Taylor Blau
2023-04-24 22:20 ` [PATCH v3 2/6] string-list: introduce `string_list_setlen()` Taylor Blau
2023-04-25 6:21 ` Jeff King
2023-04-25 21:00 ` Taylor Blau
2023-04-24 22:20 ` [PATCH v3 3/6] t/helper/test-hashmap.c: avoid using `strtok()` Taylor Blau
2023-04-24 22:20 ` [PATCH v3 4/6] t/helper/test-oidmap.c: " Taylor Blau
2023-04-24 22:20 ` [PATCH v3 5/6] t/helper/test-json-writer.c: " Taylor Blau
2023-04-25 13:57 ` Jeff Hostetler
2023-04-24 22:20 ` [PATCH v3 6/6] banned.h: mark `strtok()` and `strtok_r()` as banned Taylor Blau
2023-04-24 22:25 ` Chris Torek
2023-04-24 23:00 ` Taylor Blau
2023-04-25 6:26 ` Jeff King
2023-04-25 21:02 ` Taylor Blau
2023-04-25 6:27 ` [PATCH v3 0/6] banned: mark `strok()`, " Jeff King
2023-04-25 21:03 ` Taylor Blau
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=ZD7OlWjApwvE8Aih@nand.local \
--to=me@ttaylorr.com \
--cc=git@vger.kernel.org \
--cc=gitster@pobox.com \
--cc=peff@peff.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).