From: Johannes Schindelin <Johannes.Schindelin@gmx.de>
To: "Ævar Arnfjörð Bjarmason" <avarab@gmail.com>
Cc: git@vger.kernel.org, git-packagers@googlegroups.com,
gitgitgadget@gmail.com, gitster@pobox.com, peff@peff.net,
sandals@crustytoothpaste.net, szeder.dev@gmail.com
Subject: Re: [RFC/PATCH 7/7] grep: use PCRE v2 for optimized fixed-string search
Date: Wed, 26 Jun 2019 16:13:45 +0200 (CEST) [thread overview]
Message-ID: <nycvar.QRO.7.76.6.1906261610310.44@tvgsbejvaqbjf.bet> (raw)
In-Reply-To: <20190626000329.32475-8-avarab@gmail.com>
[-- Attachment #1: Type: text/plain, Size: 4070 bytes --]
Hi Ævar,
On Wed, 26 Jun 2019, Ævar Arnfjörð Bjarmason wrote:
> Bring back optimized fixed-string search for "grep", this time with
> PCRE v2 as an optional backend. As noted in [1] with kwset we were
> slower than PCRE v1 and v2 JIT with the kwset backend, so that
> optimization was counterproductive.
>
> This brings back the optimization for "-F", without changing the
> semantics of "\0" in patterns. As seen in previous commits in this
> series we could support it now, but I'd rather just leave that
> edge-case aside so the tests don't need to do one thing or the other
> depending on what --fixed-strings backend we're using.
Nice. Very, very nice.
> I could also support the v1 backend here, but that would make the code
> more complex, and I'd rather aim for simplicity here and in future
> changes to the diffcore. We're not going to have someone who
> absolutely must have faster search, but for whom building PCRE v2
> isn't acceptable.
I could not agree more.
> diff --git a/grep.c b/grep.c
> index 4716217837..6b75d5be68 100644
> --- a/grep.c
> +++ b/grep.c
> @@ -356,6 +356,18 @@ static NORETURN void compile_regexp_failed(const struct grep_pat *p,
> die("%s'%s': %s", where, p->pattern, error);
> }
>
> +static int is_fixed(const char *s, size_t len)
> +{
> + size_t i;
> +
> + for (i = 0; i < len; i++) {
> + if (is_regex_special(s[i]))
> + return 0;
> + }
> +
> + return 1;
> +}
> +
> #ifdef USE_LIBPCRE1
> static void compile_pcre1_regexp(struct grep_pat *p, const struct grep_opt *opt)
> {
> @@ -602,7 +614,6 @@ static int pcre2match(struct grep_pat *p, const char *line, const char *eol,
> static void free_pcre2_pattern(struct grep_pat *p)
> {
> }
> -#endif /* !USE_LIBPCRE2 */
Huh? Removing an `#endif` without removing the corresponding `#if`?
... but...
> static void compile_fixed_regexp(struct grep_pat *p, struct grep_opt *opt)
> {
> @@ -623,11 +634,13 @@ static void compile_fixed_regexp(struct grep_pat *p, struct grep_opt *opt)
> compile_regexp_failed(p, errbuf);
> }
> }
> +#endif /* !USE_LIBPCRE2 */
Ah hah!
If we would not have plenty of exercise for the PCRE2 build options, I
would be worried. But AFAICT the CI build includes this all the time, so
we're fine.
> static void compile_regexp(struct grep_pat *p, struct grep_opt *opt)
> {
> int err;
> int regflags = REG_NEWLINE;
> + int pat_is_fixed;
>
> p->word_regexp = opt->word_regexp;
> p->ignore_case = opt->ignore_case;
> @@ -636,8 +649,38 @@ static void compile_regexp(struct grep_pat *p, struct grep_opt *opt)
> if (memchr(p->pattern, 0, p->patternlen) && !opt->pcre2)
> die(_("given pattern contains NULL byte (via -f <file>). This is only supported with -P under PCRE v2"));
>
> - if (opt->fixed) {
> + pat_is_fixed = is_fixed(p->pattern, p->patternlen);
> + if (opt->fixed || pat_is_fixed) {
> +#ifdef USE_LIBPCRE2
> + opt->pcre2 = 1;
> + if (pat_is_fixed) {
> + compile_pcre2_pattern(p, opt);
> + } else {
> + /*
> + * E.g. t7811-grep-open.sh relies on the
> + * pattern being restored, and unfortunately
> + * there's no PCRE compile flag for "this is
> + * fixed", so we need to munge it to
> + * "\Q<pat>\E".
> + */
> + char *old_pattern = p->pattern;
> + size_t old_patternlen = p->patternlen;
> + struct strbuf sb = STRBUF_INIT;
> +
> + strbuf_add(&sb, "\\Q", 2);
> + strbuf_add(&sb, p->pattern, p->patternlen);
> + strbuf_add(&sb, "\\E", 2);
> +
> + p->pattern = sb.buf;
> + p->patternlen = sb.len;
> + compile_pcre2_pattern(p, opt);
> + p->pattern = old_pattern;
> + p->patternlen = old_patternlen;
> + strbuf_release(&sb);
> + }
> +#else
> compile_fixed_regexp(p, opt);
> +#endif
It might be a bit easier to read if the shorter clause came first.
Other than that: what a nice read. I should save reviewing all your patch
series for just-before-bed time.
Thanks,
Dscho
> return;
> }
>
> --
> 2.22.0.455.g172b71a6c5
>
>
next prev parent reply other threads:[~2019-06-26 14:15 UTC|newest]
Thread overview: 90+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-06-13 11:49 [PATCH 0/4] Support building with GCC v8.x/v9.x Johannes Schindelin via GitGitGadget
2019-06-13 11:49 ` [PATCH 1/4] poll (mingw): allow compiling with GCC 8 and DEVELOPER=1 Johannes Schindelin via GitGitGadget
2019-06-13 11:49 ` [PATCH 2/4] kwset: allow building with GCC 8 Johannes Schindelin via GitGitGadget
2019-06-13 16:11 ` Junio C Hamano
2019-06-14 9:53 ` SZEDER Gábor
2019-06-14 10:00 ` [RFC/PATCH v1 0/4] compat/obstack: update from upstream SZEDER Gábor
2019-06-14 10:00 ` [PATCH v1 1/4] " SZEDER Gábor
2019-06-14 10:00 ` [PATCH v1 2/4] SQUASH??? compat/obstack: fix portability issues SZEDER Gábor
2019-06-14 10:00 ` [PATCH v1 3/4] SQUASH??? compat/obstack: fix build errors with Clang SZEDER Gábor
2019-06-14 10:00 ` [PATCH v1 4/4] compat/obstack: fix some sparse warnings SZEDER Gábor
2019-06-14 17:57 ` [RFC/PATCH v1 0/4] compat/obstack: update from upstream Jeff King
2019-06-14 18:19 ` Junio C Hamano
2019-06-14 20:30 ` Ramsay Jones
2019-06-14 21:24 ` Ramsay Jones
2019-06-17 18:36 ` SZEDER Gábor
2019-06-14 16:12 ` [PATCH 2/4] kwset: allow building with GCC 8 Junio C Hamano
2019-06-17 18:26 ` SZEDER Gábor
2019-06-14 22:09 ` Ævar Arnfjörð Bjarmason
2019-06-14 22:55 ` Can we just get rid of kwset & obstack in favor of optimistically using PCRE v2 JIT? Ævar Arnfjörð Bjarmason
2019-06-14 23:19 ` Ævar Arnfjörð Bjarmason
2019-06-20 10:35 ` Jeff King
2019-06-15 9:01 ` Carlo Arenas
2019-06-15 19:15 ` brian m. carlson
2019-06-15 22:14 ` Ævar Arnfjörð Bjarmason
2019-06-26 0:03 ` [RFC/PATCH 0/7] grep: move from kwset to optional PCRE v2 Ævar Arnfjörð Bjarmason
2019-06-26 14:02 ` Johannes Schindelin
2019-06-27 9:16 ` Johannes Schindelin
2019-06-27 16:27 ` Ævar Arnfjörð Bjarmason
2019-06-27 18:21 ` Johannes Schindelin
2019-06-27 23:39 ` [PATCH v2 0/9] " Ævar Arnfjörð Bjarmason
2019-06-28 7:23 ` Ævar Arnfjörð Bjarmason
2019-06-28 16:10 ` Junio C Hamano
2019-07-01 21:20 ` [PATCH v3 00/10] " Ævar Arnfjörð Bjarmason
2019-07-01 21:31 ` Junio C Hamano
2019-07-02 11:10 ` Ævar Arnfjörð Bjarmason
2019-07-02 12:32 ` Johannes Schindelin
2019-07-02 19:57 ` Junio C Hamano
2019-07-03 10:08 ` Johannes Schindelin
2019-07-03 10:25 ` Johannes Schindelin
2019-07-03 11:27 ` Johannes Schindelin
2019-07-01 21:20 ` [PATCH v3 01/10] log tests: test regex backends in "--encode=<enc>" tests Ævar Arnfjörð Bjarmason
2019-07-01 21:20 ` [PATCH v3 02/10] grep: don't use PCRE2?_UTF8 with "log --encoding=<non-utf8>" Ævar Arnfjörð Bjarmason
2019-07-01 21:20 ` [PATCH v3 03/10] t4210: skip more command-line encoding tests on MinGW Ævar Arnfjörð Bjarmason
2019-07-01 21:20 ` [PATCH v3 04/10] grep: inline the return value of a function call used only once Ævar Arnfjörð Bjarmason
2019-07-01 21:20 ` [PATCH v3 05/10] grep tests: move "grep binary" alongside the rest Ævar Arnfjörð Bjarmason
2019-07-01 21:20 ` [PATCH v3 06/10] grep tests: move binary pattern tests into their own file Ævar Arnfjörð Bjarmason
2019-07-01 21:20 ` [PATCH v3 07/10] grep: make the behavior for NUL-byte in patterns sane Ævar Arnfjörð Bjarmason
2019-07-01 21:20 ` [PATCH v3 08/10] grep: drop support for \0 in --fixed-strings <pattern> Ævar Arnfjörð Bjarmason
2019-07-01 21:20 ` [PATCH v3 09/10] grep: remove the kwset optimization Ævar Arnfjörð Bjarmason
2019-07-01 21:21 ` [PATCH v3 10/10] grep: use PCRE v2 for optimized fixed-string search Ævar Arnfjörð Bjarmason
2019-06-27 23:39 ` [PATCH v2 1/9] log tests: test regex backends in "--encode=<enc>" tests Ævar Arnfjörð Bjarmason
2019-06-27 23:39 ` [PATCH v2 2/9] grep: don't use PCRE2?_UTF8 with "log --encoding=<non-utf8>" Ævar Arnfjörð Bjarmason
2019-06-27 23:39 ` [PATCH v2 3/9] grep: inline the return value of a function call used only once Ævar Arnfjörð Bjarmason
2019-06-27 23:39 ` [PATCH v2 4/9] grep tests: move "grep binary" alongside the rest Ævar Arnfjörð Bjarmason
2019-06-27 23:39 ` [PATCH v2 5/9] grep tests: move binary pattern tests into their own file Ævar Arnfjörð Bjarmason
2019-06-27 23:39 ` [PATCH v2 6/9] grep: make the behavior for NUL-byte in patterns sane Ævar Arnfjörð Bjarmason
2019-06-27 23:39 ` [PATCH v2 7/9] grep: drop support for \0 in --fixed-strings <pattern> Ævar Arnfjörð Bjarmason
2019-06-27 23:39 ` [PATCH v2 8/9] grep: remove the kwset optimization Ævar Arnfjörð Bjarmason
2019-06-27 23:39 ` [PATCH v2 9/9] grep: use PCRE v2 for optimized fixed-string search Ævar Arnfjörð Bjarmason
2019-06-26 0:03 ` [RFC/PATCH 1/7] grep: inline the return value of a function call used only once Ævar Arnfjörð Bjarmason
2019-06-26 0:03 ` [RFC/PATCH 2/7] grep tests: move "grep binary" alongside the rest Ævar Arnfjörð Bjarmason
2019-06-26 14:05 ` Johannes Schindelin
2019-06-26 18:13 ` Junio C Hamano
2019-06-26 0:03 ` [RFC/PATCH 3/7] grep tests: move binary pattern tests into their own file Ævar Arnfjörð Bjarmason
2019-06-26 0:03 ` [RFC/PATCH 4/7] grep: make the behavior for \0 in patterns sane Ævar Arnfjörð Bjarmason
2019-06-27 2:03 ` brian m. carlson
2019-06-26 0:03 ` [RFC/PATCH 5/7] grep: drop support for \0 in --fixed-strings <pattern> Ævar Arnfjörð Bjarmason
2019-06-26 16:14 ` Junio C Hamano
2019-06-26 0:03 ` [RFC/PATCH 6/7] grep: remove the kwset optimization Ævar Arnfjörð Bjarmason
2019-06-26 0:03 ` [RFC/PATCH 7/7] grep: use PCRE v2 for optimized fixed-string search Ævar Arnfjörð Bjarmason
2019-06-26 14:13 ` Johannes Schindelin [this message]
2019-06-26 18:45 ` Junio C Hamano
2019-06-27 9:31 ` Johannes Schindelin
2019-06-27 18:45 ` Johannes Schindelin
2019-06-27 19:06 ` Junio C Hamano
2019-06-28 10:56 ` Johannes Schindelin
2019-06-13 11:49 ` [PATCH 3/4] winansi: simplify loading the GetCurrentConsoleFontEx() function Johannes Schindelin via GitGitGadget
2019-06-13 11:49 ` [PATCH 4/4] config: avoid calling `labs()` on too-large data type Johannes Schindelin via GitGitGadget
2019-06-13 16:13 ` Junio C Hamano
2019-06-16 6:48 ` René Scharfe
2019-06-16 8:24 ` René Scharfe
2019-06-16 14:01 ` René Scharfe
2019-06-16 22:26 ` Junio C Hamano
2019-06-20 19:58 ` René Scharfe
2019-06-20 21:07 ` Junio C Hamano
2019-06-21 18:35 ` Johannes Schindelin
2019-06-22 10:03 ` René Scharfe
2019-06-22 10:03 ` [PATCH v2 1/3] config: use unsigned_mult_overflows to check for overflows René Scharfe
2019-06-22 10:03 ` [PATCH v2 2/3] config: don't multiply in parse_unit_factor() René Scharfe
2019-06-22 10:03 ` [PATCH v2 3/3] config: simplify parsing of unit factors René Scharfe
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
List information: http://vger.kernel.org/majordomo-info.html
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=nycvar.QRO.7.76.6.1906261610310.44@tvgsbejvaqbjf.bet \
--to=johannes.schindelin@gmx.de \
--cc=avarab@gmail.com \
--cc=git-packagers@googlegroups.com \
--cc=git@vger.kernel.org \
--cc=gitgitgadget@gmail.com \
--cc=gitster@pobox.com \
--cc=peff@peff.net \
--cc=sandals@crustytoothpaste.net \
--cc=szeder.dev@gmail.com \
/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.
Code repositories for project(s) associated with this public inbox
https://80x24.org/mirrors/git.git
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).