Re: Google App Engine
On 08-04-2008, Duncan Booth wrote: > Google have announced a new service called 'Google App Engine' which may > be of interest to some of the people here (although if you want to sign > up you'll have to join the queue behind me): > > From the introduction: > >> What Is Google App Engine? ... It's also interesting to see that we can find django, webob and pyyaml in their sdk (license apache 2) -- William Dodé - http://flibuste.net Informaticien indépendant -- http://mail.python.org/mailman/listinfo/python-list
subprocess freeze with rsync on windows
Hi,
I try to call rsync (with ssh) with subprocess but _some times_ it
freeze without traceback, without any reason.
My code look like that :
def sh(self,cmd):
log.debug("S cmd: " + " ".join(cmd))
p = Popen(cmd, stdout = PIPE, stderr = PIPE)
stdout, stderr = p.communicate()
if stdout:
log.debug("S ret: %s" % stdout)
if stderr:
log.error("S err: %s" % stderr)
return p.returncode
When it freeze i need to kill the ms-dos console by hand.
I've no problem on linux with the same script.
thanks
the complet script (doing rsync and cp -al on the remote side) is
available freely here : http://flibuste.net/libre/snapy (but
documentation is in french)
--
William Dodé - http://flibuste.net
--
http://mail.python.org/mailman/listinfo/python-list
Re: ANN: Shed Skin 0.2, an experimental (restricted) Python-to-C++ compiler
On 19-07-2009, Mark Dufour wrote: > Hi all, > > I have just released version 0.2 of Shed Skin, an experimental > (restricted) Python-to-C++ compiler (http://shedskin.googlecode.com). I just tested it with a litle game, to find the places of horse on a board 5x5. The result is : c 5s gcj 7s java 7s shedskin 8s python + psyco 18s cython avec malloc *int 18s cython 55s avec [] python python 303s (5m3s) -- William Dodé - http://flibuste.net Informaticien Indépendant -- http://mail.python.org/mailman/listinfo/python-list
Re: ANN: Shed Skin 0.2, an experimental (restricted) Python-to-C++ compiler
On 20-07-2009, srepmub wrote: > >> Nice timings, can you please show me the Python, Java and C code >> versions? I may do more tests. Of course, the codes are here : http://hg.flibuste.net/libre/games/cheval Like you'll see, i tried to use exactly the same code for each langage. > > also, which shedskin options did you use? did you use -bw, to disable > bounds and wrap-around checking? this can make quite a difference for > code that does a lot of indexing. if the code uses random numbers, > then -r can also make a big difference, to use C rand(), instead of > Python compatible random numbers. > > and which C++ compiler flags did you use? the default -O2, or > something like -O3 -fomit-frame-pointer -msse2..? I used the default, shedksin cheval.py; make shedskin 0.2 With -bw and -O3 -fomit-frame-pointer -msse2 i have 5.5s (instead of 8) Let me know if you find better. -- William Dodé - http://flibuste.net Informaticien Indépendant -- http://mail.python.org/mailman/listinfo/python-list
Re: ANN: Shed Skin 0.2, an experimental (restricted) Python-to-C++ compiler
On 20-07-2009, Bearophile wrote: > Skip Montanaro: >> I read just enough French to know that "avec" means "with", but I don't >> understand the difference between "avec malloc *int" and "avec []". Can you >> explain please? > > Maybe it's the time difference between using a Python list from Cython > and using a C "array" allocated with a malloc from Cython. yes, it's this -- William Dodé - http://flibuste.net Informaticien Indépendant -- http://mail.python.org/mailman/listinfo/python-list
Re: ANN: Shed Skin 0.2, an experimental (restricted) Python-to-C++ compiler
On 21-07-2009, srepmub wrote: > >> With -bw and -O3 -fomit-frame-pointer -msse2 i have 5.5s (instead of 8) >> >> Let me know if you find better. > > thanks. now I'm wondering how fast does the C version become with > these flags..? :-) I don't see any difference... -- William Dodé - http://flibuste.net Informaticien Indépendant -- http://mail.python.org/mailman/listinfo/python-list
Re: ANN: Shed Skin 0.2, an experimental (restricted) Python-to-C++ compiler
I updated the script (python, c and java) with your unrolled version + somes litle thinks. I also tried with python3.1, unladen Q2, ironpython1.1.1 Unfortunately it doesn't work more with shedskin, i'll see on the shedskin group... c 1.85s gcj 2.15s java 2.8s python2.5 + psyco 3.1s unladen-2009Q2 145s (2m45) python2.5 254s (4m14s) python3.1 300s (5m) ironpython1.1.1 680s (11m20) -- William Dodé - http://flibuste.net Informaticien Indépendant -- http://mail.python.org/mailman/listinfo/python-list
Re: ANN: Shed Skin 0.2, an experimental (restricted) Python-to-C++ compiler
On 22-07-2009, srepmub wrote: > > please send any program that doesn't work with shedskin (it still is > experimental after all) to me, or create an issue at > shedskin.googlecode.com, and I will have a look at the problem. I did it, on the discussion group http://groups.google.com/group/shedskin-discuss/browse_thread/thread/c1f47a7c21897b44 -- William Dodé - http://flibuste.net Informaticien Indépendant -- http://mail.python.org/mailman/listinfo/python-list
Re: ANN: Shed Skin 0.2, an experimental (restricted) Python-to-C++ compiler
On 22-07-2009, George Sakkis wrote: > On Jul 22, 7:38 am, William Dode wrote: > >> I updated the script (python, c and java) with your unrolled version >> + somes litle thinks. >> >> I also tried with python3.1, unladen Q2, ironpython1.1.1 >> >> Unfortunately it doesn't work more with shedskin, i'll see on the >> shedskin group... >> >> c 1.85s >> gcj 2.15s >> java 2.8s >> python2.5 + psyco 3.1s >> unladen-2009Q2 145s (2m45) >> python2.5 254s (4m14s) >> python3.1 300s (5m) >> ironpython1.1.1 680s (11m20) > > Cool; it would be interesting to see the numbers for Jython and Boo as > well if it's not too much effort. I just tried with jython, but oddly it's faster without array. Thanks to Mark, i could compile to shedskin again. And add somes little improvements... c 1.65s gcj 1.9s java 2.4s python2.5 + psyco 2.9s shedskin 3.4s unladen-2009Q2 125s (2m05) Jython 2.2.1 on java1.6.0_12 176s (without array, like shedskin) Jython 2.2.1 on java1.6.0_12 334s (with array) python2.5 215s (3m35s) python3.1 246s (4m06s) ironpython1.1.1 512 (8m32s) -- William Dodé - http://flibuste.net Informaticien Indépendant -- http://mail.python.org/mailman/listinfo/python-list
Re: ANN: Shed Skin 0.2, an experimental (restricted) Python-to-C++ compiler
On 22-07-2009, William Dode wrote: > c 1.65s > gcj 1.9s > java 2.4s > python2.5 + psyco 2.9s > shedskin 3.4s with -bw i have 2.6s > unladen-2009Q2 125s (2m05) > Jython 2.2.1 on java1.6.0_12 176s (without array, like shedskin) > Jython 2.2.1 on java1.6.0_12 334s (with array) > python2.5 215s (3m35s) > python3.1 246s (4m06s) > ironpython1.1.1 512 (8m32s) somebody can test with ironpython on windows ? Anyway, it's very impressive. I wonder if unladen will be so close in the futur. -- William Dodé - http://flibuste.net Informaticien Indépendant -- http://mail.python.org/mailman/listinfo/python-list
Re: ANN: Shed Skin 0.2, an experimental (restricted) Python-to-C++ compiler
On 23-07-2009, Kurt Smith wrote: > On Wed, Jul 22, 2009 at 2:48 AM, Bearophile wrote: >> greg: >>> Posting benchmark times for Pyrex or Cython is pretty >>> meaningless without showing the exact code that was >>> used, since times can vary enormously depending on >>> how much you C-ify things. >> >> Was this link, shown by William, not enough? >> http://hg.flibuste.net/libre/games/cheval/file/46797c3a5136/chevalx.pyx#l1 > > I took a stab at converting the recent psyco-optimized code to cython, > and got a speedup. thanks, i updated my repo with litle more tips. -- William Dodé - http://flibuste.net Informaticien Indépendant -- http://mail.python.org/mailman/listinfo/python-list
Re: ANN: psyco V2
On 23-07-2009, Christian Tismer wrote: > On 7/17/09 4:11 AM, Bearophile wrote: >> Very good, thank you. I'll try it when I can. >> >> Is Psyco3 going to borrow/steal some ideas/code from Unladen Swallow? > > Psyco3: nice typo! :-) > > Well, I haven't so far found a new idea there that I'd want > to borrow and did not know from PyPy, before. > Wasn't the project plan saying the opposite, borrowing > some ideas from psyco? :-) > http://code.google.com/p/unladen-swallow/wiki/ProjectPlan How do you see the futur of psyco when unladen-swallow will grab the best of psyco (if they can !) ? Wait and see ? Anyway, thanks a lot for your work that we can use NOW ! -- William Dodé - http://flibuste.net Informaticien Indépendant -- http://mail.python.org/mailman/listinfo/python-list
Re: ANN: psyco V2
On 24-07-2009, Christian Tismer wrote: > On 7/24/09 1:04 AM, William Dode wrote: >> On 23-07-2009, Christian Tismer wrote: > ... > >>> Wasn't the project plan saying the opposite, borrowing >>> some ideas from psyco? :-) >>> http://code.google.com/p/unladen-swallow/wiki/ProjectPlan >> >> How do you see the future of psyco when unladen-swallow will grab the >> best of psyco (if they can !) ? > > I have no objections, but also no idea so far how that could work. > Sounded like an unreflected idea to me, without seriously > checking the possibilities and/or implications. The same > kind of research apparently did not happen concerning PyPy, > which IMHO is much more suitable to take advantage from. > > This is for the current status of psyco, of course. > It will change dramatically in the next months, slowly > adopting some of PyPy's ideas. Then it might start to > make more sense. I didn't understand how psyco could use somes PyPy's ideas, i thought it's only the opposite... So i was going to read the PyPy's documentation : it's fascinating !!! But how to test it ? The doc says : "When PyPy is translated into an executable like pypy-c, the executable contains a full virtual machine that can optionally include a Just-In-Time compiler." Where is this 'option', is there somethings to activate it ? thx -- William Dodé - http://flibuste.net Informaticien Indépendant -- http://mail.python.org/mailman/listinfo/python-list
Re: ANN: Shed Skin 0.2, an experimental (restricted) Python-to-C++ compiler
On 27-07-2009, Bearophile wrote: > William Dode': >> I updated the script (python, c and java) with your unrolled version >> + somes litle thinks. > [...] >> c 1.85s >> gcj 2.15s >> java 2.8s >> python2.5 + psyco 3.1s >> unladen-2009Q2 145s (2m45) >> python2.5 254s (4m14s) >> python3.1 300s (5m) >> ironpython1.1.1 680s (11m20) > > Sorry for being late, I was away. > > In your last C version this code is useless because the C compiler is > able to perform such simple optimization by itself (but probably > Python isn't able, so if you want the code to be the similar in all > versions it may be better to keep it): I wanted so, specialy if it doesn't change a lot of the result (the difference is so small that i cannot see it)... ... > I have tried your latest C version using your compiler options, my > MinGW based on GCC 4.3.2 produces a crash at runtime. Maybe because of -msse2 ? > Using LLVM-GCC > it runs in 1.31 seconds. The D version is a bit less optimized than > your last C versions, yet using DMD it runs in 1.08-1.10 seconds. > Let's see if someone is able to write a C version faster than that D > code :-) > > Have you have compiled/read my D version? In the D version you may > have missed that I did use an extra trick: unsigned integers, so it > needs just two tests to see if a number is in the 0-5, 0-5 square :-) I didn't see, fine ! But the difference is also too small to see... > Note that Pyd, the Python-D bridge, may work with the latest DMD > version still (and it works if you use a bit older DMD compiler): > http://pyd.dsource.org/ I completly don't know anything about D... When i see the result of psyco or shedskin, i'm affraid i'll not look somewhere else soon ! However, i'd like to see a lisp implementation of this... bye -- William Dodé - http://flibuste.net Informaticien Indépendant -- http://mail.python.org/mailman/listinfo/python-list
Re: New implementation of re module
On 27-07-2009, MRAB wrote: > Hi all, > > I've been working on a new implementation of the re module. The details > are at http://bugs.python.org/issue2636, specifically from > http://bugs.python.org/issue2636#msg90954. I've included a .pyd file for > Python 2.6 on Windows if you want to try it out. > Someone can remember me how to compile it (on debian lenny), if possible with python2.5. I've also python3.1 that i build alone... I could test it with pytextile, i've a bunch of texts to bench and compare. Did you announce it on the unladen-swallow list ? They wanted to hack on RE also... -- William Dodé - http://flibuste.net Informaticien Indépendant -- http://mail.python.org/mailman/listinfo/python-list
Re: New implementation of re module
On 28-07-2009, MRAB wrote: > With the official Python 2.6 distribution for Mac OS X it works. >> > The source code is intended to replace the current 're' module in Python > 2.7 (and I'll be porting it to Python 3.2), so I'm not that worried > about Python versions earlier than 2.6 for testing, although if there's > sufficient need then I could tweak the sources for 2.5. I understand now why i could'nt compile it ! So, i would like if it's not too much work for you. -- William Dodé - http://flibuste.net Informaticien Indépendant -- http://mail.python.org/mailman/listinfo/python-list
