[pypy-dev] Student project ideas

2011-09-23 Thread Nick Wilson
I'm interested in volunteering my time to mentor a small group of senior Computer Science students at Oregon State University on a project relevant to the Python community. PyPy definitely qualifies, and I'm looking for project ideas. The project would be for their senior capstone class. Groups o

Re: [pypy-dev] Stacklets

2011-09-23 Thread Armin Rigo
Hi, 2011/9/23 Александр Седов : > I'm interested in porting _stackless to stacklets (and also probably > making it inter-thread). Thanks! Work in this direction is already well advanced. More precisely, the directory pypy/module/_stackless is obsolete and gone, and the pure Python module lib_py

Re: [pypy-dev] Stacklets

2011-09-23 Thread Александр Седов
2011/9/1 Armin Rigo : > Hi, > > The "stacklet" branch has been merged now.  The "_continuation" module > is available on all PyPys with or without the JIT on x86 and x86-64 > since a few days, and it will of course be part of release 1.6.1. > There is an almost-complete wrapper "greenlet.py".  For

Re: [pypy-dev] rlist and ll_delitem_nonneg index

2011-09-23 Thread Armin Rigo
Hi Zariko, On Fri, Sep 23, 2011 at 2:37 PM, Zariko Taba wrote: > I hit an assert in pypy/annotation/annrpython.py in addpendingblock line 231 > : assert annmodel.unionof(s_oldarg, s_newarg) == s_oldarg This is an assert that we keep hitting from time to time. Your explanation is wrong, though,

Re: [pypy-dev] Question about byte-code hacking

2011-09-23 Thread Armin Rigo
Hi, On Fri, Sep 23, 2011 at 11:27 AM, Steven D'Aprano wrote: > So the question is: would it be a burden for PyPy to make any guarantees > about the stability of bytecode? The answer is: Feel free to do anything or nothing with CPython's bytecode. As Fijal says it has little to do with PyPy. It

[pypy-dev] rlist and ll_delitem_nonneg index

2011-09-23 Thread Zariko Taba
Hi pypy ! I hit an assert in pypy/annotation/annrpython.py in addpendingblock line 231 : assert annmodel.unionof(s_oldarg, s_newarg) == s_oldarg I think I found an explanation by digging in code : The treated block is in the "ll_delitem_nonneg" function, s_newarg is a "SomeInteger(nonneg=True)"

Re: [pypy-dev] Question about byte-code hacking

2011-09-23 Thread Maciej Fijalkowski
On Fri, Sep 23, 2011 at 2:06 PM, Benjamin Peterson wrote: > 2011/9/23 Steven D'Aprano : >> >> So the question is: would it be a burden for PyPy to make any guarantees >> about the stability of bytecode? > > I would say not without great benefit. If you're doing something that > requires changing b

Re: [pypy-dev] Question about byte-code hacking

2011-09-23 Thread Benjamin Peterson
2011/9/23 Steven D'Aprano : > > So the question is: would it be a burden for PyPy to make any guarantees > about the stability of bytecode? I would say not without great benefit. If you're doing something that requires changing bytecode, the obvious answer is to add some syntax instead. -- Reg

Re: [pypy-dev] Question about byte-code hacking

2011-09-23 Thread Steven D'Aprano
Benjamin Peterson wrote: 2011/9/23 Steven D'Aprano : Hi guys, Over on the python-ideas mailing list, there is a long thread about the default argument hack in functions, used for micro-optimizations, early-binding, and monkey-patching. Various alternatives are being argued about. One proposal p

Re: [pypy-dev] Question about byte-code hacking

2011-09-23 Thread Armin Rigo
Hi, On Fri, Sep 23, 2011 at 6:12 AM, Benjamin Peterson wrote: >> What's the PyPy position on bytecode hacking? Good, bad, evil, don't mind >> either way? > > (...) > Secondly, it's useless for speed when you have a JIT. Indeed, although it is not 100% true, because we also have an interpreter.