Mailman 3 November 2005 - pypy-dev

Nice floats
by Christian Tismer 22 Nov '05

22 Nov '05

Hi pypy-dev, I just stumbled oversome code from Gustavo. I think we were considering doing exact float formatting. I wrote something that works. But if we like to have it, this implementation gives the shortest possible representation. At some cost, of course... http://labix.org/python-nicefloat -- Christian Tismer :^) <mailto:tismer@stackless.com> tismerysoft GmbH : Have a break! Take a ride on Python's Johannes-Niemeyer-Weg 9A : *Starship* http://starship.python.net/ 14109 Berlin : PGP key -> http://wwwkeys.pgp.net/ work +49 30 802 86 56 mobile +49 173 24 18 776 fax +49 30 80 90 57 05 PGP 0x57F3BF04 9064 F4E1 D754 C2FF 1619 305B C09C 5A3B 57F3 BF04 whom do you want to sponsor today? http://www.stackless.com/

2 1

Re: [pypy-dev] Re: [pypy-svn] r20092 - pypy/dist/pypy/rpython
by Laura Creighton 21 Nov '05

21 Nov '05

In a message of Mon, 21 Nov 2005 01:31:06 PST, Christian Tismer writes: >arigo(a)codespeak.net wrote: > >> Author: arigo >> Date: Sat Nov 19 20:42:12 2005 >> New Revision: 20092 >> >> Modified: >> pypy/dist/pypy/rpython/robject.py >> Log: >> Bug fix (for Christian :-) to compile code like >> >> d = {} >> d[x] = y >> >> where x and y are SomeObjects. The reason is that operations manipulat >ing >> PyObjects must not have a 'void' return value, even operations like 'se >titem', >> but a PyObject one. > >Thank you very much! >I worked around that, but in the end, it is a bug. >And btw. I discovereed a bulk of more bugs or inconsistencies, >which might be tackled after I'm back from my "mission". > >ATM, I need to make a small addition to allow float values as dict >keys. There is a lot more considerable, but I can work around all >other stuff and don't want to spoil our project without discussion. >The need for float comes from the lack of a primitive long type. >Alternatively, I'm thinking to map the long type to some internal >64 bit type. This would make some sense, also because we also have >file related code that would need this type. > >I will start discusions when I'm back. Adding a bit of completeness >(hash on float) will probably not hurt anybody. > >Btw., I have meanwhile turned all their application into RPython. >Although this took many new little classes, and the lack of >properties was as painful as the inability to define any __xx__ >methods, the result is still rather pythonic and runs at blinding speed- >I'm now only 30% slower than their highly optimized Java >implementation, and I guess I could get the rest for free by finishing >the needed optimizations of refcounting and some other tiny things. > >all the best - chris > >-- >Christian Tismer :^) <mailto:tismer@stackless.com> >tismerysoft GmbH : Have a break! Take a ride on Python's >Johannes-Niemeyer-Weg 9A : *Starship* http://starship.python.net/ >14109 Berlin : PGP key -> http://wwwkeys.pgp.net/ >work +49 30 802 86 56 mobile +49 173 24 18 776 fax +49 30 80 90 57 05 >PGP 0x57F3BF04 9064 F4E1 D754 C2FF 1619 305B C09C 5A3B 57F3 BF04 > whom do you want to sponsor today? http://www.stackless.com/ >_______________________________________________ >pypy-dev(a)codespeak.net >http://codespeak.net/mailman/listinfo/pypy-dev Correct me if I am wrong, but if we allow float keys, then we make code that is not portable between machines that have different float representations? Also, given that your new American friends are dealing with money, what is wrong with using the decimal type as keys for them, or possibly better still Facundo's money module? Laura

3 2

Why does python's range function only take numbers?
by Laura Creighton 21 Nov '05

21 Nov '05

I just wrote: def chrange(start, end): return [chr(i) for i in range(ord(start), ord(end)+1)] for about the thousandth time in my life. Which got me thinking. Due to the miracle of signature based polymorphism, we could make our range function take chars as well. Anybody know why CPython doesn't do this? Laura, now musing about a range function that takes an arbitrary sequence ...

2 1

Re: [pypy-svn] r20092 - pypy/dist/pypy/rpython
by Christian Tismer 21 Nov '05

21 Nov '05

arigo(a)codespeak.net wrote: > Author: arigo > Date: Sat Nov 19 20:42:12 2005 > New Revision: 20092 > > Modified: > pypy/dist/pypy/rpython/robject.py > Log: > Bug fix (for Christian :-) to compile code like > > d = {} > d[x] = y > > where x and y are SomeObjects. The reason is that operations manipulating > PyObjects must not have a 'void' return value, even operations like 'setitem', > but a PyObject one. Thank you very much! I worked around that, but in the end, it is a bug. And btw. I discovereed a bulk of more bugs or inconsistencies, which might be tackled after I'm back from my "mission". ATM, I need to make a small addition to allow float values as dict keys. There is a lot more considerable, but I can work around all other stuff and don't want to spoil our project without discussion. The need for float comes from the lack of a primitive long type. Alternatively, I'm thinking to map the long type to some internal 64 bit type. This would make some sense, also because we also have file related code that would need this type. I will start discusions when I'm back. Adding a bit of completeness (hash on float) will probably not hurt anybody. Btw., I have meanwhile turned all their application into RPython. Although this took many new little classes, and the lack of properties was as painful as the inability to define any __xx__ methods, the result is still rather pythonic and runs at blinding speed- I'm now only 30% slower than their highly optimized Java implementation, and I guess I could get the rest for free by finishing the needed optimizations of refcounting and some other tiny things. all the best - chris -- Christian Tismer :^) <mailto:tismer@stackless.com> tismerysoft GmbH : Have a break! Take a ride on Python's Johannes-Niemeyer-Weg 9A : *Starship* http://starship.python.net/ 14109 Berlin : PGP key -> http://wwwkeys.pgp.net/ work +49 30 802 86 56 mobile +49 173 24 18 776 fax +49 30 80 90 57 05 PGP 0x57F3BF04 9064 F4E1 D754 C2FF 1619 305B C09C 5A3B 57F3 BF04 whom do you want to sponsor today? http://www.stackless.com/

1 0

PyPy newbie question: How to play with types
by Noam Raphael 19 Nov '05

19 Nov '05

Hello, I want to use PyPy as a way of trying a change I will propose to Python. It includes a bit of messing with types. For example, I want to make a new type called frozendict, and another new type, basedict. frozendict and dict will be subclasses of basedict. basedict will not be instantiable, but will implement all the dict methods that do not change the dict. frozendict will add a single operation to basedict - it will be hashable. dict will be the same as the dict today, but since many operations will be inherited from basedict, it will be shorter. I also want to add a new built-in, frozen(x), that will simply call x.__frozen__(). I have looked into the directories. I have found pypy/objspace/dict{object,type}.py. My questions are: 1. How do I make new built-in types, play with their inheritence, make them uninstantiable? Should I register it somewhere? 2. Where do I add a new built-in like frozen()? How should I make it callable from interpreter-level code? Thanks a lot, Noam

2 4

Re: [pypy-dev] Re: [pypy-svn] r20071 - pypy/dist/pypy/doc
by Laura Creighton 19 Nov '05

19 Nov '05

In a message of Sat, 19 Nov 2005 14:20:10 +0100, Carl Friedrich Bolz writes: > > >lac(a)codespeak.net wrote: >> Author: lac >> Date: Sat Nov 19 10:21:38 2005 >> New Revision: 20071 >> >> Modified: >> pypy/dist/pypy/doc/news.txt >> Log: >> I now know the perfect test for Germans. You ask them to >> spell 'GÃ¶teborg'. Depending on how they misspell it, you can >> determine exactly which part of Germany they call home. :-) >> TODO: test on Nik, see if it is language or a country problem. :-) >> >> >> Modified: pypy/dist/pypy/doc/news.txt >> ======================================================================= >======= >> --- pypy/dist/pypy/doc/news.txt (original) >> +++ pypy/dist/pypy/doc/news.txt Sat Nov 19 10:21:38 2005 >> @@ -12,7 +12,7 @@ >> PyPy Sprint in GÃ¶teborg 7th - 11th December 2005 >> ================================================= >> >> -The next PyPy sprint is scheduled to be in December 2005 in GÃ¶theborg. >> +The next PyPy sprint is scheduled to be in December 2005 in GÃ¶teborg. >> Its main focus is heading towards phase 2, which means JIT work, >> alternate threading models and logic programming (but there are also >> other possible topics). We'll give newcomer-friendly introductions. > >well, I think in this case the part of Germany the author called home is > "Bristol" :-) But it happened WHILE HE WAS IN DUSSELDORF! I smell a paper for the JIR! http://www.jir.com/ Laura >Cheers, > >Carl Friedrich >_______________________________________________ >pypy-dev(a)codespeak.net >http://codespeak.net/mailman/listinfo/pypy-dev

1 0

Re: [pypy-svn] r20071 - pypy/dist/pypy/doc
by Carl Friedrich Bolz 19 Nov '05

19 Nov '05

lac(a)codespeak.net wrote: > Author: lac > Date: Sat Nov 19 10:21:38 2005 > New Revision: 20071 > > Modified: > pypy/dist/pypy/doc/news.txt > Log: > I now know the perfect test for Germans. You ask them to > spell 'Göteborg'. Depending on how they misspell it, you can > determine exactly which part of Germany they call home. :-) > TODO: test on Nik, see if it is language or a country problem. :-) > > > Modified: pypy/dist/pypy/doc/news.txt > ============================================================================== > --- pypy/dist/pypy/doc/news.txt (original) > +++ pypy/dist/pypy/doc/news.txt Sat Nov 19 10:21:38 2005 > @@ -12,7 +12,7 @@ > PyPy Sprint in Göteborg 7th - 11th December 2005 > ================================================= > > -The next PyPy sprint is scheduled to be in December 2005 in Götheborg. > +The next PyPy sprint is scheduled to be in December 2005 in Göteborg. > Its main focus is heading towards phase 2, which means JIT work, > alternate threading models and logic programming (but there are also > other possible topics). We'll give newcomer-friendly introductions. well, I think in this case the part of Germany the author called home is "Bristol" :-) Cheers, Carl Friedrich

1 0

testing testing
by Michael Hudson 18 Nov '05

18 Nov '05

1 0

Re: [pypy-svn] r19981 - in pypy/branch/somepbc-refactoring/pypy: rpython rpython/test translator
by Michael Hudson 17 Nov '05

17 Nov '05

pedronis(a)codespeak.net writes: > Author: pedronis > Date: Thu Nov 17 16:35:10 2005 > New Revision: 19981 > > Modified: > pypy/branch/somepbc-refactoring/pypy/rpython/annlowlevel.py > pypy/branch/somepbc-refactoring/pypy/rpython/test/test_llann.py > pypy/branch/somepbc-refactoring/pypy/translator/annrpython.py > Log: > (arigo, pedronis) > > fixed annlowlevel.py and its basic tests in test_llann.py > > now annotator.build_types delegate to the function desc to possibly > apply a required specialisation Hey, that's what I did on the train :) Oh well, now I know that bit of the code much better :) Cheers, mwh -- Hmmm... its Sunday afternoon: I could do my work, or I could do a Fourier analysis of my computer's fan noise. -- Amit Muthu, ucam.chat (from Owen Dunn's summary of the year)

1 0

pypy-sync vs trains
by Michael Hudson 17 Nov '05

17 Nov '05

Assuming there's going to be a pypy-sync today (?), I'm not going to be at it as I'll be on a train. My activity report goes something like: LAST: somepbc-refactoring, admin NEXT: somepbc-refactoring BLOCKERS: failing to understand our current code :) I think we really need to get a move on with somepbc-refactoring or merging will become increasingly difficult. Cheers, mwh -- I sense much distrust in you. Distrust leads to cynicism, cynicism leads to bitterness, bitterness leads to the Awareness Of True Reality which is referred to by those-who-lack-enlightenment as "paranoia". I approve. -- David P. Murphy, alt.sysadmin.recovery

1 0