Mailman 3 December 2010 - pypy-dev

funding/popularity?
by Dima Tisnek 20 Dec '10

20 Dec '10

First of all, congratulation on getting Eurostars Would be nice if companies that use Python extensively chipped in [more]. By [out of] my [@%%] calculations, each percent of perfromance improvement that pypy achieves potentially saves google many millions a year in data centre costs. What are other companies use/love python and how much? Popularity, I for one would like python to be used much more widely; it is of course a double-edges sword, on one hand more users and developers means more contributors, libraries and understanding that python is important, in terms of research, funding, etc; on the other more bad code and mess. Anyhow I see at least these problems for now: in general: 2.x/3.x dichtomy; some ext's are not ported, pypy doesn't yet run 3k (I'd be willing to work on this btw), lack of authoritative information on the subject; Since pypy is a forward-looking project, I think a push to include 3.2 (and perhaps 2.7 which would come naturally) would be very good here, at least clear statement that pypy is going there. More visibility for performance achievements would do good too. in classic apps: hard to bundle python with downloadable apps, e.g. win builds still recommend 2.5; perhaps stackless single-exe builds are the best here? A bit better on linux where dependency tracking is the responsibility of the disrto. on the web: bad support in apache, little advice for users what to use and how frameworks are structured; I reckon some pypy side projects could be great here, e.g. isolated programs within same interpreter or multiple interpreters in same process. in schools/unis: I really can't comprehend why! Anyhow I think 1.4 release is great! Though I must contend I am still to try it with most of my code, too many ext deps and very little of my python code is cpu-bound. Sidetracking... one day when pypy jit/gc/etc are all worked out, how hard would it be to use same techniques and most of backends for some unrelated language that doesn't have jit yet, e.g. php? And how hard would it be to marry two dynamic languages, so that modules from one could be used in the other? Or that modules written in rpython could be used in several langs?

2 1

playing with fast-forward
by Gary Robinson 20 Dec '10

20 Dec '10

I'm experimenting with the fast-forward branch. I'm actually not sure about the proper way to get it. (I have the main branch working fine.) I downloaded a nojit version from http://buildbot.pypy.org/nightly/branch/fast-forward/ since I didn't see any jit versions listed there. I was happy to see that the method I care most about, multiprocessing.Pool.imap_unordered, seemed to work. But I want to test it with the jit. The page (https://bitbucket.org/pypy/pypy/src/021b219e0aef) appears to be for the branch, but the mercurial URL shown on that page appears to be for the main project. I tried the downloads under the Source pop-up menu. For some reason, I only sporadically am able to get a complete .gz file.** But I did get one. I was able to run: python translate.py -Ojit successfully -- or at least it appeared so. But I couldn't find a bin/pypy to run!? Any suggestions how to run fast-forward with jit? ** I've used text browsers w3m and elinks on SuSE, and Safari on OS X (two different machines in different cities). Twice, the Safari download has stopped at < 500K. (It LOOKS like a successful download, it's just incomplete.) The SuSE browsers usually get 16 or 18MB and stop, and tar gets "gzip: stdin: unexpected end of file" when I try to uncompress them. -- Gary Robinson CTO Emergent Discovery, LLC personal email: garyrob(a)me.com work email: grobinson(a)emergentdiscovery.com Company: http://www.emergentdiscovery.com Blog: http://www.garyrobinson.net

3 3

ImportError: No module named _sqlite3
by Arnd Rechenburg 20 Dec '10

20 Dec '10

Hi, I tried to use sqlite3 with pypy but I got the following error: >>>> import sqlite Traceback (most recent call last): File "<console>", line 1, in <module> ImportError: No module named sqlite >>>> import sqlite3 Traceback (most recent call last): File "<console>", line 1, in <module> File "sqlite3/__init__.py", line 24, in <module> from dbapi2 import * File "sqlite3/dbapi2.py", line 27, in <module> from _sqlite3 import * ImportError: No module named _sqlite3 Someone knows where to put the _sqlite3.so library? Any help would be appreciated. Thx in advance, Arnd Rechenburg

1 0

Interpretor for vectorized langugage
by Leon Sit 18 Dec '10

18 Dec '10

Hi all Is pypy suitable for writing interpretor of vectorized language like Matlab, R etc which vector and matrix are first class objects? This includes matrix shape inference, and efficient linear algebra code generation. Any thoughts? Leon

7 12

speed.pypy.org and mercurial
by Antonio Cuni 18 Dec '10

18 Dec '10

Hi Miquel, hi all, as you probably have noticed, we have recently migrated the main repo to mercurial. Now speed.pypy.org receives a revision number in the form "40046:2088ce763fc2", but of course it can no longer fetches the commit logs from the svn server. Would it be possible to fetch the commits from the bitbucket repo, please? ciao, Anto

8 13

Can't run PyPy due to missing libbz2.so.1.0
by Gary Robinson 17 Dec '10

17 Dec '10

When I try to start PyPy, I get: bin/pypy: error while loading shared libraries: libbz2.so.1.0: cannot open shared object file: No such file or directory This is on a SuSE Linux Enterprise Server 10.3 SP2. This system does have: /lib/libbz2.so.1.0.0 As a quick experiment, I tried temporarily renaming it to /lib/libbz2.so.1.0 but that didn't change the result. I also tried installing the bzip2 package from the rpm, but it said it was already installed. Can anyone suggest a solution? -- Gary Robinson CTO Emergent Discovery, LLC personal email: garyrob(a)me.com work email: grobinson(a)emergentdiscovery.com Company: http://www.emergentdiscovery.com Blog: http://www.garyrobinson.net

3 4

error from function.py
by Arnd Rechenburg 16 Dec '10

16 Dec '10

Hi, I got the following error message: File "/home/unixhome/rechenbu/output/pypy-1.4/lib-python/2.5.2/calendar.py", line 11, in <module> import locale as _locale File "/home/unixhome/rechenbu/output/pypy-1.4/lib-python/modified-2.5.2/local e.py", line 30, in <module> from _locale import * File "/home/unixhome/rechenbu/output/pypy-1.4/lib_pypy/_locale.py", line 8, in <module> from ctypes_support import standard_c_lib as libc File "/home/unixhome/rechenbu/output/pypy-1.4/lib_pypy/ctypes_support.py", line 24, in <module> standard_c_lib.__errno_location.restype = ctypes.POINTER(ctypes.c_int) File "/home/unixhome/rechenbu/output/pypy-1.4/lib-python/modified-2.5.2/ctype s/__init__.py", line 363, in __getattr__ func = self.__getitem__(name) File "/home/unixhome/rechenbu/output/pypy-1.4/lib-python/modified-2.5.2/ctype s/__init__.py", line 368, in __getitem__ func = self._FuncPtr((name_or_ordinal, self)) File "/home/unixhome/rechenbu/output/pypy-1.4/lib_pypy/_ctypes/function.py", line 112, in __init__ ptr = self._getfuncptr([], ctypes.c_int) File "/home/unixhome/rechenbu/output/pypy-1.4/lib_pypy/_ctypes/function.py", line 197, in _getfuncptr return cdll.ptr(self.name, argshapes, resshape, self._flags_) AttributeError: 'NoneType' object has no attribute 'ptr' Someone knows more about it? Regards, Arnd

2 1

import errors when calling pypy
by Arnd Rechenburg 16 Dec '10

16 Dec '10

Hi, I have problems to start pypy. I got the following errors: $ pypy-1.4/bin/pypy pypy-1.4/bin/pypy: /usr/lib64/libcrypto.so.0.9.8: no version information available (required by pypy-1.4/bin/pypy) pypy-1.4/bin/pypy: /usr/lib64/libssl.so.0.9.8: no version information available (required by pypy-1.4/bin/pypy) 'import site' failed Python 2.5.2 (79529, Nov 25 2010, 23:04:06) [PyPy 1.4.0] on linux2 Type "help", "copyright", "credits" or "license" for more information. ImportError: No module named atexit debug: OperationError: debug: operror-type: ImportError debug: operror-value: No module named _pypy_interact Could someone tell me how to set the environment for the libraries? I tried update PATH etc. but without success. On the system (linux SLES10, AMD 64-bit) itself python 2.5.5 is used. Any help would be appreciated. Thx in advance, Arnd Rechenburg Arnd Rechenburg * Senior Software Engineer * Tele Atlas Deutschland GmbH & Co. KG * Am Neuen Horizont 1, 31177 Harsum, Deutschland * HRA 200613 * VAT ID DE811953885 * Steuernummer 30/210/15702 * e-mail: arnd.rechenburg(a)teleatlas.com <mailto:arnd.rechenburg@teleatlas.com> * direct +49 5127 40 8364 * reception +49 5127 40 80 * fax +49 5127 40 8348 * www.teleatlas.com <http://www.teleatlas.com/> <http://www.teleatlas.com/> P Please consider the environment before printing this e-mail. E-mail Confidentiality: This e-mail and any attachments thereto may contain information which is confidential and/or protected by intellectual property rights and are intended for the sole use of the recipient(s) named above. Any use of the information contained herein (including, but not limited to, total or partial reproduction, communication or distribution in any form) by persons other than the designated recipient(s) is prohibited. TELE ATLAS DOES NOT AMEND EXISTING AGREEMENTS OR CREATE ANY BINDING AGREEMENTS OR OBLIGATIONS VIA E-MAIL. ANY STATEMENTS OR ATTACHMENTS WITHIN THIS MESSAGE WILL NOT BE CONSIDERED BINDING. If you have received this e-mail in error, please notify the sender either by telephone or by e-mail and delete the material from any computer. Thank you for your cooperation. Tele Atlas B.V. and its affiliates. www.teleatlas.com <http://www.teleatlas.com/>

2 1

Mercurial workflow
by Jacob Hallén 16 Dec '10

16 Dec '10

Hi folks, now that the switch to Mercurial has happened, people are discovering that their Subversion workflow habits don't quite work. This is because Mercurial has a distributed philosophy, inlike svn which has the concept of the holy central server where all operations take place. In Mercurial, you do all the important operations in your local repository. If you have a central node (which is not strictly necessary), you just push ready made change sets to it. Any conflict resolving has already been done on your local machine. This means that it is very convenient to have at least 2 repositories locally (this takes hadly any extra space, since unchanged files are hard linked to be the same file in each repository). You keep one repository as a staging area to which you pull changes from the central server (or other peoples repositories) and to which you pull changes from your other local repositories. It is here you resolve any merge conflicts and it is from here you do pushes of new versions to the central pypy server on bitbucket. For every idea you have, you clone your repository and do the work for the idea in the clone. If you need to collaborate with someone else you synchronize your repo with their repo without going through the central server. Branching is a concept that is separated from the repository clones and it is just as good an idea to make branches for new features as it was in svn. When you work on an idea you should keep in mind that it is an extremely fast operation to commit your work. You can do it for a number of various reasons - whenever you have made a set of changes that you feel belong together, if your machine is flakey and you want to have a backup copy somewhere else or if you want to continue the work on a different machine. I sometimes program in bed, and it is very simple to do a pull from my workstation to my laptop. The new workflow will make the detailed changes less visible to others, since you commit work to the central server in fewer and bigger chunks. This means that there will be fewer "oops, I misspelled this" and more of a higher level view of what is being developed. I hope this helps getting your minds around how to work with Mercurial. I hope you will discover that the way it is different from svn will improve your work environment and your work habits. It has done so for me. Jacob

6 5

Re: [pypy-dev] pypy-dev Digest, Vol 365, Issue 14
by Leon Sit 15 Dec '10

15 Dec '10

Yes I have. However I want to write an interpretor for matlab/R codebase, not using python to write numerical code. Leon Best Regards, Leon Sit Graduate Student in Computer Science at University of Chicago Message: 7 > Date: Wed, 15 Dec 2010 17:21:38 +0000 > From: Ren? Dudfield <renesd(a)gmail.com> > Subject: Re: [pypy-dev] Interpretor for vectorized langugage > To: PyPy Dev <pypy-dev(a)codespeak.net> > Message-ID: > <AANLkTin_Q8C2DgEm06tnFi-NWT_p+6kg665EbcG=YPKj(a)mail.gmail.com> > Content-Type: text/plain; charset="iso-8859-1" > > Hi, > > have you seen numpy/scipy? > > Also see these: > http://morepypy.blogspot.com/2009/07/pypy-numeric-experiments.html > http://ademan.wordpress.com/category/code/pypy/ > > > cheers, > > > On Wed, Dec 15, 2010 at 5:11 PM, Leon Sit <wing1127aishi(a)gmail.com> wrote: > > > Hi all > > > > Is pypy suitable for writing interpretor of vectorized language like > > Matlab, R etc which vector and matrix are first class objects? This > includes > > matrix shape inference, and efficient linear algebra code generation. > > > > Any thoughts? > > > > Leon > > > > _______________________________________________ > > pypy-dev(a)codespeak.net > > http://codespeak.net/mailman/listinfo/pypy-dev > > >

2 1