New submission from John Malmberg:
The _PyDumpTraceback and _Py_DumpTracebackThreads routines in traceback.h are
tagged with PyAPI_DATA attributes when they should be tagged with PyAPI_FUNC.
For platforms that use those attributes, this can cause run-time issues if
those methods are called
John Malmberg added the comment:
These files are identical to the files in the expat git repository.
So it looks like I need to submit the fix and get it accepted to that
repository first. And then at some point cPython will pick up the change.
I can still provide diffs if desired, but based
John Malmberg added the comment:
OpenVMS needs %lld or % PY_FORMAT_LONG_LONG d in order to build the _ctypes
module.
--
nosy: +John.Malmberg
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue10320
Changes by John Malmberg wb8...@gmail.com:
--
nosy: +John.Malmberg
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue17095
___
___
Python-bugs-list
Changes by John Malmberg wb8...@gmail.com:
--
nosy: +John.Malmberg
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue22038
___
___
Python-bugs-list
John Malmberg added the comment:
The existing Python text is uppercase D for Ctrl-D, so I maintained it.
stty documentation also uses upper case for control characters.
The EOF character can be viewed or set via the stty utility or the tcsetattr()
or equivalent routine. Some platforms may
John Malmberg added the comment:
Note that Python directly supports Unixes which supply fork() and Microsoft
Windows that supplies spawn*() methods.
So as long as a platform supports either the fork() or the the Microsoft
Windows spawn() methods, there should not be barriers implementing
New submission from John Malmberg:
Distutils currently can not handle a Posix platform that does not implement
fork().
This patch retries with the _spawn_nt to use the spawn() methods if fork() is
not implemented.
A platform that does not implement fork() can provide spawn*() methods
New submission from John Malmberg:
Have setquit() use the actual EOF character where available instead of assuming
Ctrl-D.
--
components: Library (Lib)
files: lib_site_py.gdiff
messages: 226154
nosy: John.Malmberg
priority: normal
severity: normal
status: open
title: Report actual EOF
John Malmberg added the comment:
There are multiple degrees of Posix compliance.
While X/Open documents the Posix requirements for implementing fork(), so far I
have not found anything that requires that fork() be present.
Configure tests for c-python also test for the presence of fork
New submission from John Malmberg:
The Modules/expat/expat.h and Modules/expat/xmlrole.h contain VMS specific code
that actually breaks building python 3.5 on VMS.
Per http://bugs.python.org/issue16136, such VMS specific code should be removed
for Python 3.5.
This code may still be needed
New submission from John Malmberg:
If the zlib.so fails to build while building python, subsequent runs of
setup.py fail, which prevents a trying again to build zlib.so after the issue
is fixed unless all the .so modules built are deleted.
File /PRJ_ROOT/CPYTHON/Lib/zipfile.py, line 20
New submission from John Malmberg:
Minor issue, the #define TAG line incorrectly is terminated with a semicolon.
This results in a double semicolon on the following line where the macro is
expanded.
As this is an informational message, this is just cosmetic.
Listing fragment
New submission from John Malmberg:
unicodeobject.c is using %zd instead of the PY_FORMAT_SIZE_T as set by
configure.
Comments in pymacconfig.h state that this will cause problems on OS-X 10.4 with
binaries build on OS-X 10.5.
The attached patch unicodeobject.giff changes unicodeobject.c
New submission from John Malmberg:
hashtable.c is using %zd instead of the PY_FORMAT_SIZE_T as set by configure.
Comments in pymacconfig.h state that this will cause problems on OS-X 10.4 with
binaries build on OS-X 10.5.
The attached patch hashtable.giff changes hashtable.c to be like
New submission from John Malmberg:
With issue 16136 VMS support was removed for Python V3
A test build of the in-development branch using the UNIX instruction and the
current GNV product with a few minor tweaks produced a Python.exe interpreter
that is somewhat functional.
Most of the issues
John Malmberg added the comment:
This is an informational ticket.
At this point, it is a spare time activity for me. Hopefully there are other
interested parties to help. Time will tell.
--
___
Python tracker rep...@bugs.python.org
http
John Malmberg added the comment:
Most of the issues needed to build python3 properly on VMS involve either bugs
in the VMS C API that need to be worked around or missing libraries that
Python3 expects to have.
As none of the Python developers should need to care about this, and these
issues
John Malmberg added the comment:
Does not look like anything vital to VMS got removed.
Configure basically worked under current GNV.
And a few tweaks later with out changing any files checked out of the Mercurial
repository, make is producing a functional python binary before it quits.
bash
John Malmberg added the comment:
Encompasserve.org is now back up. It took a little longer than expected.
Should be no issues now in getting free hobby license keys for VMS on VAX,
Alpha, and Itanium by using the web forms.
--
___
Python tracker
John Malmberg added the comment:
Access to VMS licenses and media:
1. Hobby - non-commercial applications - Free with 1 year time-bombed license
keys with free media download. A self-service mostly web based system.
http://www.openvms.org/pages.php?page=Hobbyist
Easiest way to get
21 matches
Mail list logo
