Rls 0.23.0rc1 #41
Rls 0.23.0rc1 #41
Conversation
|
Hi! This is the friendly automated conda-forge-linting service. I wanted to let you know that I linted all conda-recipes in your PR ( Here's what I've got... For recipe:
|
|
Hi! This is the friendly automated conda-forge-linting service. I just wanted to let you know that I linted all conda-recipes in your PR ( |
| url: https://github.com/pandas-dev/pandas/releases/download/v{{ version }}/{{ name }}-{{ version }}.tar.gz | ||
| sha256: 4cdec5f000d2988cf791de590b94037fa52e198c31f07278dd889cc13cf21d6f | ||
|
|
||
| build: | ||
| number: 0 | ||
| script: python setup.py install --single-version-externally-managed --record=record.txt | ||
| script: python -m pip install --no-deps --ignore-installed . |
There was a problem hiding this comment.
recipe/meta.yaml
Outdated
| @@ -1,23 +1,22 @@ | |||
| {% set name = "pandas" %} | |||
| {% set version="0.23.0rc1" %} | |||
| {% set version = "0.23.0rc1" %} | |||
There was a problem hiding this comment.
We actually need this to the bot can parse the version and send PRs updating the package. I'll check if the master branch needs this change too.
There was a problem hiding this comment.
Thanks. If it’s easier I was just planning to merge this to master, even though it’s just an RC. Will that cause issues?
There was a problem hiding this comment.
This change will publish the master branch in the rc channel, to avoid confusing we send PRs like this to a "dev" branch instead to keep master only for stable releases. Is that OK for you?
There was a problem hiding this comment.
Yeah, I just didn't want to mess anything up by trying to do that myself :) To get a dev branch, should I
- Branch `dev off the current master
- Cherry pick 7bd7d85
- Push that branch to upstream
Is there any way to skip the CI builds when I push upstream? Will [ci skip] do it?
|
Linux is hitting conda-forge/conda-smithy#762 :-( |
|
Ok, I subscribed there for updates. Our build is also apparently has a hardcoded path in it somewhere that conda-build doesn't like. Trying to reproduce locally. |
Don't worry about it. We can merge with the failing status and the package will be uploaded, or I can remove the latest BTW, I created a branch for the dev releases: https://github.com/conda-forge/pandas-feedstock/tree/dev do you mind pointing this PR there? |
|
Thanks, I changed the base branch to |
|
BTW, if you re-render with |
|
OK, re-rendered w/ master. |
@ocefpaf to clarify, do you think the failure at https://ci.appveyor.com/project/conda-forge/pandas-feedstock/build/1.0.88/job/056twhw62n7wn3f6#L4381 is OK? That seems like it's halting the build, and will have to be addressed by a new sdist from pandas. |
Nope. I meant the CircleCI failure, The ApVeyor failure is real 😄 The previous released is passing in #42 and has virtually the change recipe changes as this PR. Do you think it may be something with the RC release? I'll take a closer look at the logs soon. |
|
This hard-coded path seems odd See https://ci.appveyor.com/project/conda-forge/pandas-feedstock/build/1.0.88/job/056twhw62n7wn3f6#L4381 |
|
Yep, definitely a problem with our build script :D (we call abspath in our setup.py) I'll tag fix that, tag and rc2, and then update the SHA here. |
This caused failures on the conda-forge build xref conda-forge/pandas-feedstock#41
|
Thanks for all your help @ocefpaf! |
|
Glad I could help. Thanks for the RC release, looking forward to trying it out! |
| - cython | ||
| - numpy 1.9.* # [py27 or py35] | ||
| - numpy 1.11.* # [py36] | ||
| run: | ||
| - python | ||
| - numpy >=1.9.* # [py27 or py35] | ||
| - numpy >=1.11.* # [py36] | ||
| - python-dateutil | ||
| - python-dateutil >=2.5.0 |
There was a problem hiding this comment.
thanks @TomAugspurger yeah meant to update this :>
No description provided.