forum.vmspython.org Forum Index forum.vmspython.org
Forum system
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

2.7.3 sources?

 
Post new topic   Reply to topic    forum.vmspython.org Forum Index -> Python for OpenVMS
View previous topic :: View next topic  
Author Message
douglasrupp



Joined: 24 Jul 2012
Posts: 15

PostPosted: Sat Jul 28, 2012 5:59 pm    Post subject: 2.7.3 sources? Reply with quote

Curious to know if the 2.7.3 sources are available for download? The latest I can find in the source directory are labeled 2.5.4.
Back to top
View user's profile Send private message
jfp



Joined: 12 Jul 2004
Posts: 632

PostPosted: Mon Jul 30, 2012 10:31 am    Post subject: Reply with quote

You can download the source code from python.org and the necessary patches from vmspython.org:

First you need a VMS system with Python 2.7 already installed.
Then run the following commands:
Code:
$ @python_vms:setup
$ @mercurial_root:[vms]setup


Clone the source repository :

prepare a directory where you will put the repository
Code:
$ create/dir [.python]/vers=1
$ set default [.python]


clone the repository
Code:
$ hg clone http://hg.python.org/cpython/


clone the 2.7 patches
Code:
$ set def [.cpython.^.hg]
$ hg clone http://hg.vmspython.org/cpython/.hg/patches-27/
$ set def [-]
$ hg qqueue --create 27


Verify that you are able to compile 2.7 :
Code:
$ hg update -C 2.7
$ hg qqueue -l
$ hg qqueue 27
$ hg qpush -a -v
$ hg summary
$ set def [.vms]
$ mms


Clean the 2.7 environment:
Code:
$ mms clean
$ set def [-]
$ hg qpop -a -v
$ hg update -C default
$ hg summary

JF
Back to top
View user's profile Send private message Send e-mail Visit poster's website
douglasrupp



Joined: 24 Jul 2012
Posts: 15

PostPosted: Tue Jul 31, 2012 12:09 am    Post subject: Reply with quote

I got down to: "hg qpush -a -v" and after awhile it failed with the following output. Yes it really does say "doneing Python/thread_pthread.h"

...
getting Modules/readline.c
getting Modules/socketmodule.c
getting Modules/socketmodule.h
getting Objects/fileobject.c
getting Objects/object.c
getting Python/dynload_shlib.c
getting Python/import.c
getting Python/random.c
getting Python/sysmodule.c
doneing Python/thread_pthread.h
** unknown exception encountered, please report by visiting
** http://mercurial.selenic.com/wiki/BugTracker
** Python 2.7.3+ (default, Apr 20 2012, 20:06:26) [DECC]
** Mercurial Distributed SCM (version bfbf64)
** Extensions loaded: mq
Traceback (most recent call last):
File "/MERCURIAL_ROOT/HG", line 14, in <module>
mercurial.dispatch.run()
File "/MERCURIAL_ROOT/mercurial/dispatch.py", line 27, in run
sys.exit((dispatch(request(sys.argv[1:])) or 0) & 255)
File "/MERCURIAL_ROOT/mercurial/dispatch.py", line 64, in dispatch
return _runcatch(req)
File "/MERCURIAL_ROOT/mercurial/dispatch.py", line 87, in _runcatch
return _dispatch(req)
File "/MERCURIAL_ROOT/mercurial/dispatch.py", line 685, in _dispatch
cmdpats, cmdoptions)
File "/MERCURIAL_ROOT/mercurial/dispatch.py", line 467, in runcommand
ret = _runcommand(ui, options, cmd, d)
File "/MERCURIAL_ROOT/mercurial/dispatch.py", line 739, in _runcommand
return checkargs()
File "/MERCURIAL_ROOT/mercurial/dispatch.py", line 693, in checkargs
return cmdfunc()
File "/MERCURIAL_ROOT/mercurial/dispatch.py", line 682, in <lambda>
d = lambda: util.checksignature(func)(ui, *args, **cmdoptions)
File "/MERCURIAL_ROOT/mercurial/util.py", line 513, in check
return func(*args, **kwargs)
File "/MERCURIAL_ROOT/hgext/mq.py", line 2619, in push
exact=opts.get('exact'))
File "/MERCURIAL_ROOT/hgext/mq.py", line 1234, in push
ret = self.apply(repo, s, list, all_files=all_files)
SystemError: error return without exception set
Back to top
View user's profile Send private message
douglasrupp



Joined: 24 Jul 2012
Posts: 15

PostPosted: Tue Jul 31, 2012 12:56 am    Post subject: Reply with quote

Looks like the problem was "no space left on device".
Trying to recover and retry.
Back to top
View user's profile Send private message
douglasrupp



Joined: 24 Jul 2012
Posts: 15

PostPosted: Tue Jul 31, 2012 3:33 am    Post subject: Reply with quote

Executing "mms" calls a .com file which looks for sys$common:[openssl098g.vms]. AFAICT that version is no longer available from HP, the latest is 098w, and the previous 098h.

I'll try the latest.
Back to top
View user's profile Send private message
douglasrupp



Joined: 24 Jul 2012
Posts: 15

PostPosted: Tue Jul 31, 2012 5:33 am    Post subject: Reply with quote

There's no "openssl_startup.com" in the most recent sources from HP, so I don't know what's supposed to happen, whether this is a homegrown startup file, or it's been renamed (there's an openssl_systartup.com") or what.

Please advise.
Back to top
View user's profile Send private message
jfp



Joined: 12 Jul 2004
Posts: 632

PostPosted: Tue Jul 31, 2012 8:36 am    Post subject: Reply with quote

Well,

the building process was for my systems and never intended to be used outside my environment, so you may expect to have some problem.

For example you need the local source tree from the 2.5.4 source kits, some logicals like disk$tools, etc...

JF
Back to top
View user's profile Send private message Send e-mail Visit poster's website
douglasrupp



Joined: 24 Jul 2012
Posts: 15

PostPosted: Tue Jul 31, 2012 2:43 pm    Post subject: Reply with quote

No problem, procedure has been remarkably smooth so far. I was just asking what was in this openssl_startup.com file or where to find openssl098g if necessary.
Back to top
View user's profile Send private message
douglasrupp



Joined: 24 Jul 2012
Posts: 15

PostPosted: Tue Jul 31, 2012 10:17 pm    Post subject: Reply with quote

I had to upgrade mms, now I get a ways into the build process before CC complains about not finding <ffi.h> when compiling callbacks.c.

The /include directive looks correct
I have the latest C compiler installed.
The logical name LIBFFI is defined.
I can list the file ffi.h at the location specified, no problem.

Attempting to compile callbacks.c by hand from DCL results in an unrelated error about trigraphs, so I'm stuck for now.
Back to top
View user's profile Send private message
douglasrupp



Joined: 24 Jul 2012
Posts: 15

PostPosted: Tue Jul 31, 2012 11:41 pm    Post subject: Reply with quote

There was a process logical defined in the build scripts that was hiding the correct definition of libffi

FTR, Some of the library sources are also required, e.g. readline, in disk$toolsL[libsrc...]

Progress.
Back to top
View user's profile Send private message
douglasrupp



Joined: 24 Jul 2012
Posts: 15

PostPosted: Wed Aug 01, 2012 12:04 am    Post subject: Reply with quote

Success. Python.exe, Python_nothread.exe, and pythonshr.exe successfully linked.

The last bit needed was to run mms in the disk$tools[libsrc.readline*] directory and fiddle with some logical names specific to your setup, e.g. libraries that were expected to be found in sys$common vice lda1.

Smile
Back to top
View user's profile Send private message
jfp



Joined: 12 Jul 2004
Posts: 632

PostPosted: Thu Aug 02, 2012 8:35 am    Post subject: Reply with quote

Congratulations Very Happy

If I remember correctly the libffi provided in the ld image has some bugs.
If you want I can put online the latest port which has no known bugs...

JF
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Display posts from previous:   
Post new topic   Reply to topic    forum.vmspython.org Forum Index -> Python for OpenVMS All times are GMT + 2 Hours
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group