Page MenuHomeFreeBSD

Mk/Uses/python.mk: Flag unknown USE_PYTHON features
AbandonedPublic

Authored by tobik on Mar 13 2019, 10:04 AM.
Tags
None
Referenced Files
Unknown Object (File)
Jan 19 2024, 5:49 PM
Unknown Object (File)
Jan 13 2024, 12:09 AM
Unknown Object (File)
Jan 2 2024, 7:54 PM
Unknown Object (File)
Dec 30 2023, 8:39 AM
Unknown Object (File)
Dec 22 2023, 11:47 PM
Unknown Object (File)
Dec 2 2023, 11:41 PM
Unknown Object (File)
Nov 10 2023, 12:16 AM
Unknown Object (File)
Nov 7 2023, 5:51 AM
Subscribers

Details

Summary
Test Plan

Ran make -V DEV_ERROR over the tree. The following ports currently have problems:

# Port				unknown features
databases/py-asyncpg		setuptools
graphics/py-cairocffi		cffi
graphics/sk1			noflavor
science/dvc			noflavor

Diff Detail

Repository
rP FreeBSD ports repository
Lint
No Lint Coverage
Unit
No Test Coverage
Build Status
Buildable 23060
Build 22128: arc lint + arc unit

Event Timeline

All the USES that validate their arguments set IGNORE so that the mistake is found out early. I think this should also be setting IGNORE. Also, could you check that the arguments to USES=python:xxx are also validated?

  • Check USES=python arguments too

New failures:

# Port				bogus args
games/ski			3.2+:run
net-mgmt/librenms-agent		2.7:run
sysutils/py-python-consul	2.7-3.6+
sysutils/xen-guest-tools	configure
  • Simplify and use :N instead of :C
# Port				bogus args
games/ski			3.2+:run
net-mgmt/librenms-agent		2.7:run
sysutils/py-python-consul	2.7-3.6+
sysutils/xen-guest-tools	configure

Mmmm, nice, I "fixed" all four.

I don't see anything wrong with the code, I'll let the other portmgr ponder this for a bit.

sunpoet added inline comments.
Mk/Uses/python.mk
254–258

I would suggest merging _PYTHON_FEATURE_<FEATURENAME> setting (line 263-266) as follows:

.for var in ${USE_PYTHON}
.  if !empty(_VALID_PYTHON_FEATURES:M${var})
_PYTHON_FEATURE_${var:C/=.*$//:tu}=        ${var:C/.*=//:S/,/ /g}
.  else
_INVALID_PYTHON_FEATURES+=        ${var}
.  endif
.endfor
Mk/Uses/python.mk
254–258

Why? What is wrong with looping twice?

Mk/Uses/python.mk
254–258

I'm not saying it's wrong. It's just a suggestion.

Thanks. Committed in rP499257. I probably should have reclaimed this before committing for Phabricator to update this post-commit.