[OE-core] gpgme -- yet another python issue...

Mark Hatle mark.hatle at windriver.com
Wed Mar 8 20:18:03 UTC 2017


I don't have a fix for this, but if I enable multilib processing I see two
additional issues.

Maybe someone else can spot how to fix this?

My config (add to local.conf)

# Uncomment the following lines to enable multilib builds
require conf/multilib.conf
MULTILIBS = "multilib:lib32"
DEFAULTTUNE_virtclass-multilib-lib32 = "x86"


MACHINE=qemux86-64 bitbake gpgme


What I get is:

WARNING: gpgme-1.8.0-r0 do_package: QA Issue: gpgme: Files/directories were
installed but not shipped in any package:
  /usr/lib/python3.5/site-packages/gpg-1.8.0-py3.5.egg-info
  /usr/lib/python3.5/site-packages/gpg/_gpgme.cpython-35m-x86_64-linux-gnu.so
  /usr/lib/python3.5/site-packages/gpg/version.py
  /usr/lib/python3.5/site-packages/gpg/callbacks.py
  /usr/lib/python3.5/site-packages/gpg/core.py
  /usr/lib/python3.5/site-packages/gpg/errors.py
  /usr/lib/python3.5/site-packages/gpg/__init__.py
  /usr/lib/python3.5/site-packages/gpg/results.py
...
  /usr/lib/python3.5/site-packages/gpg/__pycache__/results.cpython-35.pyc
  /usr/lib/python3.5/site-packages/gpg/__pycache__/util.cpython-35.pyc
  /usr/lib/python3.5/site-packages/gpg/__pycache__/gpgme.cpython-35.pyc
Please set FILES such that these items are packaged. Alternatively if they are
unneeded, avoid installing them or delete them within do_install.
gpgme: 59 installed and not shipped files. [installed-vs-shipped]


So two problems:

1) When multilibs are enabled the path is incorrect.  The gpgme is using:
PYTHON_SITEPACKAGES_DIR

but this is pointing to /usr/lib64/python3.5, but the files installed into
/usr/lib/python3.5.. (presumably due to a bug in gpgme's makefiles..)

2) Why are the __pycache__ files being generated, I did not see this behavior on
other (non-x86) architectures.  So we may have some kind of contamination problem.


(I likely will not have time to fix these before next week, so if someone else
can figure this out please give it a try!)

--Mark



More information about the Openembedded-core mailing list