-
Notifications
You must be signed in to change notification settings - Fork 145
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Put libraries in $WHEELNAME.libs #90
Conversation
Oh, wow. That actually worked. Now it just needs a unit test. |
@njsmith, this works on all of the existing unit tests, and I can see that the libraries are getting placed where we expected. Do you think it is actually necessary to add a unit test of a wheel with multiple C extensions? |
If you have a package that contains multiple C extensions that depend on the same shared libraries, then `delocate-wheel` saves duplicate copies of all of the shared libraries. An example is shown below. This makes for packages that are several times larger than necessary. This patch causes `delocate-wheel` to sweep up all of the dependencies and put them in a single directory, rather than creating duplicates. The convention is based on the proposal in pypa/auditwheel#89 and pypa/auditwheel#90.
If you have a package that contains multiple C extensions that depend on the same shared libraries, then `delocate-wheel` saves duplicate copies of all of the shared libraries. This makes for packages that are several times larger than necessary. This patch causes `delocate-wheel` to sweep up all of the dependencies and put them in a single directory, rather than creating duplicates. The convention is based on the proposal in pypa/auditwheel#89 and pypa/auditwheel#90.
If you have a package that contains multiple C extensions that depend on the same shared libraries, then `delocate-wheel` saves duplicate copies of all of the shared libraries. This makes for packages that are several times larger than necessary. This patch causes `delocate-wheel` to sweep up all of the dependencies and put them in a single directory, rather than creating duplicates. The convention is based on the proposal in pypa/auditwheel#89 and pypa/auditwheel#90.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think it is necessary to add a test to ensure the behaviour works right now and in the future when changes are made. Please do add one.
@daa you can actually make a pull request relative to my branch. |
I haven't noticed at the time that |
And one more fix: https://github.com/lpsinger/auditwheel/pull/3 |
This is a pretty big semantic change so I'm going to hold off until after the 2.0 release to review. |
If you have a package that contains multiple C extensions that depend on the same shared libraries, then `delocate-wheel` saves duplicate copies of all of the shared libraries. This makes for packages that are several times larger than necessary. This patch causes `delocate-wheel` to sweep up all of the dependencies and put them in a single directory, rather than creating duplicates. The convention is based on the proposal in pypa/auditwheel#89 and pypa/auditwheel#90.
If you have a package that contains multiple C extensions that depend on the same shared libraries, then `delocate-wheel` saves duplicate copies of all of the shared libraries. This makes for packages that are several times larger than necessary. This patch causes `delocate-wheel` to sweep up all of the dependencies and put them in a single directory, rather than creating duplicates. The convention is based on the proposal in pypa/auditwheel#89 and pypa/auditwheel#90.
If you have a package that contains multiple C extensions that depend on the same shared libraries, then `delocate-wheel` saves duplicate copies of all of the shared libraries. This makes for packages that are several times larger than necessary. This patch causes `delocate-wheel` to sweep up all of the dependencies and put them in a single directory, rather than creating duplicates. The convention is based on the proposal in pypa/auditwheel#89 and pypa/auditwheel#90.
Codecov Report
@@ Coverage Diff @@
## master #90 +/- ##
==========================================
- Coverage 87.32% 87.19% -0.13%
==========================================
Files 19 19
Lines 978 976 -2
Branches 215 214 -1
==========================================
- Hits 854 851 -3
Misses 86 86
- Partials 38 39 +1
Continue to review full report at Codecov.
|
I've rebased the patch and it's passing the tests. |
I've rebased this PR again. Is there any further feedback, or is this ready to be merged? |
can you check te builds? |
Yes, one of the jobs failed here:
Would you please click retry in travis? |
I restarted the build but now it's failing because of another issue:
For some reason it's passing the wrong platform name. |
Oh, weird. They're all failing now, too. |
Looks like the |
Thanks @lpsinger! |
Thank you all! |
Should fix #89.