1 |
W dniu sob, 02.09.2017 o godzinie 23∶05 +0200, użytkownik Michał Górny |
2 |
napisał: |
3 |
> W dniu sob, 02.09.2017 o godzinie 12∶19 -0700, użytkownik Zac Medico |
4 |
> napisał: |
5 |
> > On 09/02/2017 10:46 AM, Michał Górny wrote: |
6 |
> > > dev-python/pycparser-2.18+ exposes a design flaw in dev-python/ply that |
7 |
> > > makes it unable to work with -OO code. Remove the optimizations from |
8 |
> > > Portage shebangs to prevent triggering the issue until we find a proper |
9 |
> > > solution for it. |
10 |
> > > |
11 |
> > > Bug: https://bugs.gentoo.org/628386 |
12 |
> > > --- |
13 |
> > > bin/clean_locks | 2 +- |
14 |
> > > bin/dispatch-conf | 2 +- |
15 |
> > > bin/ebuild | 2 +- |
16 |
> > > bin/emaint | 2 +- |
17 |
> > > bin/env-update | 2 +- |
18 |
> > > bin/portageq | 2 +- |
19 |
> > > tabcheck.py | 2 +- |
20 |
> > > 7 files changed, 7 insertions(+), 7 deletions(-) |
21 |
> > > |
22 |
> > > diff --git a/bin/clean_locks b/bin/clean_locks |
23 |
> > > index 13af06197..fb245972f 100755 |
24 |
> > > --- a/bin/clean_locks |
25 |
> > > +++ b/bin/clean_locks |
26 |
> > > @@ -1,4 +1,4 @@ |
27 |
> > > -#!/usr/bin/python -bO |
28 |
> > > +#!/usr/bin/python -b |
29 |
> > |
30 |
> > The diff shows -O, but the commit messages says -OO, so which one is it really? |
31 |
> |
32 |
> Yes, it's a curious problem. -OO is the one breaking it but py<3.5 seems |
33 |
> to be happy to load -OO files with -O: |
34 |
> |
35 |
> $ python2.7 -O -c 'import pycparser; print(pycparser.__file__)' |
36 |
> /usr/lib64/python2.7/site-packages/pycparser/c_parser.py:20: RuntimeWarning: parsing methods must have __doc__ for pycparser to work properly |
37 |
> class CParser(PLYParser): |
38 |
> /usr/lib64/python2.7/site-packages/pycparser/__init__.pyo |
39 |
> |
40 |
> $ python3.4 -O -c 'import pycparser; print(pycparser.__cached__)' |
41 |
> /usr/lib64/python3.4/site-packages/pycparser/c_parser.py:20: RuntimeWarning: parsing methods must have __doc__ for pycparser to work properly |
42 |
> class CParser(PLYParser): |
43 |
> /usr/lib64/python3.4/site-packages/pycparser/__pycache__/__init__.cpython-34.pyo |
44 |
> |
45 |
> This doesn't seem to be the case anymore for py3.5+: |
46 |
> |
47 |
> $ python3.5 -O -c 'import pycparser; print(pycparser.__cached__)' |
48 |
> /usr/lib64/python3.5/site-packages/pycparser/__pycache__/__init__.cpython-35.opt-1.pyc |
49 |
|
50 |
My roundabout thinking here was really silly. I was mistaking -O/-OO for |
51 |
.pyc/.pyo. For the record, the correct explanation: |
52 |
|
53 |
Python < 3.5 supports only either -O or -OO, and compiles both levels to |
54 |
the same .pyo files. So if we compiled the relevant packages with -OO, |
55 |
then -O implicitly meant -OO. |
56 |
|
57 |
-- |
58 |
Best regards, |
59 |
Michał Górny |