From: Marco Felsch <m.felsch@pengutronix.de>
To: Sascha Hauer <sha@pengutronix.de>
Cc: Ahmad Fatoum <a.fatoum@pengutronix.de>, barebox@lists.infradead.org
Subject: Re: [PATCH 2/2] commands: test: add based support for bash-test style
Date: Tue, 4 Jul 2023 12:07:55 +0200 [thread overview]
Message-ID: <20230704100755.g7vf7rkhmexz4a3k@pengutronix.de> (raw)
In-Reply-To: <20230704084646.GW18491@pengutronix.de>
On 23-07-04, Sascha Hauer wrote:
> On Tue, Jul 04, 2023 at 10:32:22AM +0200, Marco Felsch wrote:
> > Hi Ahmad,
> >
> > On 23-07-04, Ahmad Fatoum wrote:
> > > On 04.07.23 00:58, Marco Felsch wrote:
> > > > Add [[ expression ]] support which allow pattern matching like:
> > > >
> > > > - [[ "foo1" == "foo*" ]]
> > > > - [[ "foo" != "bar" ]]
> > > >
> > > > Signed-off-by: Marco Felsch <m.felsch@pengutronix.de>
> > > > ---
> > > > commands/Kconfig | 13 +++++++++++++
> > > > commands/test.c | 39 ++++++++++++++++++++++++++++++++++-----
> > > > 2 files changed, 47 insertions(+), 5 deletions(-)
> > > >
> > > > diff --git a/commands/Kconfig b/commands/Kconfig
> > > > index 4d3ff631a8..615e96aa9d 100644
> > > > --- a/commands/Kconfig
> > > > +++ b/commands/Kconfig
> > > > @@ -1219,6 +1219,19 @@ config CMD_TEST
> > > > !, =, !=, -eq, -ne, -ge, -gt, -le, -lt, -o, -a, -z, -n, -d, -e,
> > > > -f, -L; see 'man test' on your PC for more information.
> > > >
> > > > +if CMD_TEST
> > > > +
> > > > +config CMD_TEST_BASH_COMP
> > > > + tristate
> > > > + select CONFIG_FNMATCH
> > >
> > > select FNMATCH, no CONFIG_
> >
> > Right.
> >
> > > Does this really need its own symbol? Can't you just use CONFIG_GLOB
> > > and amend the help text?
> >
> > The implementation was inspired by busybox and they use fnmatch.
> > Checking glob.c I don't see any difference, therefore I would keep the
> > fnmatch.
>
> glob() is the wrong function anyway. glob() expands patterns to existing
> files. You want to do pattern matching only not looking into the VFS.
> fnmatch() is the right function here.
>
> Besides, glob() uses fnmatch(), so by using glob() you would add even
> more code.
>
> >
> > I wasn't sure if every user need the bash(ism) support and sometimes
> > bootloader partitions are really small due to legacy reasons. Therefore
> > I went this way to let the user the choice to enable/disable the
> > support.
>
> You can't do much in barebox without globalvar support which already
> uses fnmatch(), so it's likely enabled anyway.
Ah right, I will drop it then.
Thanks,
Marco
>
> Sascha
>
> > > > + return fnmatch(right_op, left_op, 0);
> > > > +#endif
> > >
> > > Scripts that execute differently depending on a barebox config option
> > > is not good user experience. I think it would be better if [[ is an
> > > error if support is missing.
> >
> > I was thinking about a warning which is printed once you use [[ and
> > didn't enabled it before. Then I read man bash again and [[ supports
> > '=' as well which equals the [ function. Therefore I dropped the
> > warning.
> >
> > > Also can't the #ifdef be replaced with a IS_ENABLED()
> >
> > I don't think so, since the fnmatch is added conditional to keep the
> > code base the same if not enabled.
>
> Use IS_ENABLED(). The linker will discard unused functions for you. We
> use this all over the place.
>
> Sascha
>
> --
> Pengutronix e.K. | |
> Steuerwalder Str. 21 | http://www.pengutronix.de/ |
> 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
> Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
>
next prev parent reply other threads:[~2023-07-04 10:09 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-07-03 22:58 [PATCH 1/2] commands: test: simplify argv handling Marco Felsch
2023-07-03 22:58 ` [PATCH 2/2] commands: test: add based support for bash-test style Marco Felsch
2023-07-04 7:33 ` Ahmad Fatoum
2023-07-04 8:32 ` Marco Felsch
2023-07-04 8:46 ` Sascha Hauer
2023-07-04 10:07 ` Marco Felsch [this message]
2023-07-04 8:51 ` Sascha Hauer
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=20230704100755.g7vf7rkhmexz4a3k@pengutronix.de \
--to=m.felsch@pengutronix.de \
--cc=a.fatoum@pengutronix.de \
--cc=barebox@lists.infradead.org \
--cc=sha@pengutronix.de \
/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