From: Peter Korsgaard <peter@korsgaard.com>
To: buildroot@busybox.net
Subject: [Buildroot] Suggestion to "support/scripts/apply-patches.sh: do not apply patches with renames"
Date: Sat, 15 May 2021 08:44:04 +0200 [thread overview]
Message-ID: <87fsyopkcr.fsf@dell.be.48ers.dk> (raw)
In-Reply-To: <CAALxPJ0qprnhxWtHGg7oeY=Ws4+w3Wt-_hdM1WvbL6B42jQO5g@mail.gmail.com> (Ryota Kinukawa's message of "Thu, 13 May 2021 08:27:05 +0900")
>>>>> "Ryota" == Ryota Kinukawa <pojiro.jp@gmail.com> writes:
Hi,
> Hi. Thank you for your great work, Buildroot team.
> I'm writing this email for suggestions to support/scripts/apply-patches.sh.
> Recently, I was working with buildroot to apply patches-4.14.164-rt73.
> It included a patch((patches-4.14.164-rt73/0462-..)) which renames a file.
> This caused the build to terminate with an error.
> The reason for this is that apply-patches.sh exits when it finds a patch
> that contains a rename.
> This feature is kind for users who use legacy `patch` commands,
> but not kind for current users.
> So I'd like to suggest adding a kind of this line
> `grep -q "^BR2_*_LEGACY_PATCH_CMD=y" "$BR2_CONFIG" && \`
> to below if statement.
> ```
> if ${uncomp} "${path}/$patch" | grep -q "^rename from" && \
> ${uncomp} "${path}/$patch" | grep -q "^rename to" ; then
> ```
> Parsing the results of the `patch` is dependent on the implementation of it,
> so I think it would be better to get them who use `legacy patch` to set it
> explicitly.
> (For considering backward compatibility,
> (it might be better if the environment variable had the opposite meaning.
> How about this idea?
The problem with such an option is that it describes a property of the
build host, not of the Buildroot configuration - And we still wouldn't
be able to use rename patches as long as that option was supported.
So we want apply-patches.sh to work the same for everyone.
But we're now 4 years later, so perhaps it is time to unconditionally
drop support for GNU patch < 2.7?
GNU patch 2.7 was released in September 2012, so 8.5 years ago:
https://savannah.gnu.org/forum/forum.php?forum_id=7361
Thomas, Yann - Any ideas how common such old patch version are nowadays?
--
Bye, Peter Korsgaard
next prev parent reply other threads:[~2021-05-15 6:44 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-05-12 23:27 [Buildroot] Suggestion to "support/scripts/apply-patches.sh: do not apply patches with renames" Ryota Kinukawa
2021-05-15 6:44 ` Peter Korsgaard [this message]
2021-05-15 9:40 ` Yann E. MORIN
2021-05-15 10:57 ` Romain Naour
2021-05-15 11:19 ` Yann E. MORIN
2021-05-15 11:46 ` Romain Naour
2021-05-15 12:01 ` Yann E. MORIN
2021-05-16 0:51 ` Ryota Kinukawa
2021-05-16 8:18 ` Ryota Kinukawa
2021-05-17 12:46 ` Yann E. MORIN
2021-05-17 18:55 ` Arnout Vandecappelle
2021-05-19 3:28 ` [Buildroot] [PATCH] support/dependencies, scripts: accept patches with renames pojiro.jp at gmail.com
2021-05-19 6:43 ` Yann E. MORIN
2021-05-19 8:34 ` [Buildroot] [PATCH v2] " pojiro.jp at gmail.com
2021-05-19 8:46 ` Yann E. MORIN
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=87fsyopkcr.fsf@dell.be.48ers.dk \
--to=peter@korsgaard.com \
--cc=buildroot@busybox.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