Message ID | cover.1657597513.git.paul.eggleton@linux.microsoft.com |
---|---|
State | Not Applicable, archived |
Headers | show |
Thanks Paul, this was one of the most annoying shortcomings of devtool. Alex On Tue, 12 Jul 2022 at 06:07, Paul Eggleton <paul.eggleton@linux.microsoft.com> wrote: > > A few fixes for devtool, mostly relating to recipes that fetch from git > where S points to a subdirectory of the repo. > > Note: I wasn't 100% sure if it was appropriate to be referencing WORKDIR > in patch.py; let me know if I should try to rework that. > > > The following changes since commit db28cd0e1540e44db963108430205c8c0c817774: > > gperf: Switch to upstream patch (2022-07-09 20:58:31 +0100) > > are available in the git repository at: > > git://git.openembedded.org/openembedded-core-contrib paule/devtool37-oe > http://cgit.openembedded.org/openembedded-core-contrib/log/?h=paule/devtool37-oe > > Paul Eggleton (4): > patch: handle if S points to a subdirectory of a git repo > devtool: finish: handle patching when S points to subdir of a git repo > devtool: ignore pn- overrides when determining SRC_URI overrides > oe-selftest: devtool: test modify git recipe building from a subdir > > meta/lib/oe/patch.py | 8 +++--- > meta/lib/oe/recipeutils.py | 9 ++++-- > meta/lib/oeqa/selftest/cases/devtool.py | 51 +++++++++++++++++++++++++++++++++ > scripts/lib/devtool/standard.py | 29 ++++++++++++++----- > 4 files changed, 84 insertions(+), 13 deletions(-) > > -- > 1.8.3.1 > > > -=-=-=-=-=-=-=-=-=-=-=- > Links: You receive all messages sent to this group. > View/Reply Online (#167889): https://lists.openembedded.org/g/openembedded-core/message/167889 > Mute This Topic: https://lists.openembedded.org/mt/92326549/1686489 > Group Owner: openembedded-core+owner@lists.openembedded.org > Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [alex.kanavin@gmail.com] > -=-=-=-=-=-=-=-=-=-=-=- >
On Mon, 2022-07-11 at 21:06 -0700, Paul Eggleton wrote: > A few fixes for devtool, mostly relating to recipes that fetch from git > where S points to a subdirectory of the repo. > > Note: I wasn't 100% sure if it was appropriate to be referencing WORKDIR > in patch.py; let me know if I should try to rework that. It is probably ok. The code won't pick up on the variable usage from a hash perspective there but it would be referenced elsewhere anyway. There was a failure which is probably from this series though: https://autobuilder.yoctoproject.org/typhoon/#/builders/87/builds/3817/steps/14/logs/stdio (I've not confirmed it was the series, I'm just guessing) Cheers, Richard
On Tue, Jul 12, 2022 at 04:59:38PM +0100, Richard Purdie wrote: > On Mon, 2022-07-11 at 21:06 -0700, Paul Eggleton wrote: > > A few fixes for devtool, mostly relating to recipes that fetch from git > > where S points to a subdirectory of the repo. > > > > Note: I wasn't 100% sure if it was appropriate to be referencing WORKDIR > > in patch.py; let me know if I should try to rework that. > > It is probably ok. The code won't pick up on the variable usage from a > hash perspective there but it would be referenced elsewhere anyway. > > There was a failure which is probably from this series though: > > https://autobuilder.yoctoproject.org/typhoon/#/builders/87/builds/3817/steps/14/logs/stdio > > > (I've not confirmed it was the series, I'm just guessing) Hmm, yes, it probably is. I did run the tests but somehow I didn't see this failure. I'll investigate further. Paul