Mailman 3 July 2005 - pypy-dev

targetpypy
by Ben.Young＠risk.sungard.com July 15, 2005

July 15, 2005

Hi Everyone, I've been running translate_pypy -text targetpypy every couple of days recently, to see how things are progressing, but I always get the same error: ImportError: No module named _formatting Is this a known problem, or am I doing something wrong? Cheers, Ben

2 1

plans for os and math
by Samuele Pedroni July 14, 2005

July 14, 2005

This mail is an attempt to summarize a bit our current thinking and plans for starting implementing os and math modules functionality. The idea is to implement them as mixed modules (like sys and __builtin__ are now), so to have, for starting, a module/posix and module/math packages in module. E.g. open would be implemented by posix at interpreter level, for example like: def open(w_fname, w_flag, w_mode=0777): fname = space.str_w(w_fname) flag = space.int_w(w_flag) mode = … [View More]

1 0

Fw: [pypy-svn] r14617 - pypy/dist/pypy/module/unicodedata[POSSIBLE SPAM ]
by Ben.Young＠risk.sungard.com July 13, 2005

July 13, 2005

Hi, I was just looking over the subversion commit messages and had a couple of comments to make. Did you really want [0 * ... ] in some of these diffs, rather than [0] * ...? Also, don't lists resize to avoid this kind of behaviour? Sorry if you had already spotted them! Cheers, Ben pypy-svn-bounces(a)codespeak.net wrote on 13/07/2005 15:05:11: > Author: ac > Date: Wed Jul 13 16:05:10 2005 > New Revision: 14617 > > Modified: > pypy/dist/pypy/module/unicodedata/… [View More]function.py > Log: > Avoid frequent reallocations when normalizing unicode. > > Modified: pypy/dist/pypy/module/unicodedata/function.py > ============================================================================== > --- pypy/dist/pypy/module/unicodedata/function.py (original) > +++ pypy/dist/pypy/module/unicodedata/function.py Wed Jul 13 16:05:10 2005 > @@ -100,6 +100,7 @@ > NCount = (VCount*TCount) > SCount = (LCount*NCount) > > + > def normalize(space, w_form, w_unistr): > form = space.str_w(w_form) > if not space.is_true(space.isinstance(w_unistr, space.w_unicode)): > @@ -121,7 +122,9 @@ > space.wrap('invalid normalization form')) > > strlen = space.int_w(space.len(w_unistr)) > - result = [] > + result = [0] * (strlen + strlen / 10 + 10) > + j = 0 > + resultlen = len(result) > # Expand the character > for i in range(strlen): > ch = space.int_w(space.ord(space.getitem(w_unistr, space.wrap(i)))) > @@ -132,33 +135,57 @@ > V = VBase + (SIndex % NCount) / TCount; > T = TBase + SIndex % TCount; > if T == TBase: > - result.extend([L, V]) > + if j + 2 > resultlen: > + result.extend([0 * (j + 2 - resultlen + 10)]) > + resultlen = len(result) > + result[j] = L > + result[j + 1] = V > + j += 2 > else: > - result.extend([L, V, T]) > + if j + 3 > resultlen: > + result.extend([0 * (j + 3 - resultlen + 10)]) > + resultlen = len(result) > + result[j] = L > + result[j + 1] = V > + result[j + 2] = T > + j += 3 > continue > - > - result.extend(decomposition.get(ch, [ch])) > + decomp = decomposition.get(ch) > + if decomp: > + decomplen = len(decomp) > + if j + decomplen > resultlen: > + result.extend([0 * (j + decomplen - resultlen + 10)]) > + resultlen = len(result) > + for ch in decomp: > + result[j] = ch > + j += 1 > + else: > + if j + 1 > resultlen: > + result.extend([0 * (j + 1 - resultlen + 10)]) > + resultlen = len(result) > + result[j] = ch > + j += 1 > > # Sort all combining marks > - for i in range(len(result)): > + for i in range(j): > ch = result[i] > comb = unicodedb.combining(ch) > if comb == 0: > continue > - for j in range(i, 0, -1): > - if unicodedb.combining(result[j - 1]) <= comb: > - result[j] = ch > + for k in range(i, 0, -1): > + if unicodedb.combining(result[k - 1]) <= comb: > + result[k] = ch > break > > - result[j] = result[j - 1] > + result[k] = result[k - 1] > else: > result[0] = ch > > if not composed: # If decomposed normalization we are done > - return space.newunicode(result) > + return space.newunicode(result[:j]) > > - if len(result) <= 1: > - return space.newunicode(result) > + if j <= 1: > + return space.newunicode(result[:j]) > > current = result[0] > starter_pos = 0 > @@ -166,8 +193,8 @@ > prev_combining = 0 > if unicodedb.combining(current): > prev_combining = 256 > - for j in range(1, len(result)): > - next = result[j] > + for k in range(1, j): > + next = result[k] > next_combining = unicodedb.combining(next) > if next_insert == starter_pos + 1 or prev_combining < next_combining: > # Combine if not blocked > _______________________________________________ > pypy-svn mailing list > pypy-svn(a)codespeak.net > http://codespeak.net/mailman/listinfo/pypy-svn > [View Less]

2 1

Re: [pypy-svn] r14625 - pypy/dist/pypy/objspace/std
by Michael Hudson July 13, 2005

July 13, 2005

tismer(a)codespeak.net writes: > Author: tismer > Date: Wed Jul 13 17:53:23 2005 > New Revision: 14625 > > Modified: > pypy/dist/pypy/objspace/std/strutil.py > Log: > another less urgent optimization from the train Kiel-Berlin: > > refactored string_to_float quite a lot. The issue was raised > by Python 2.4's test_long, which caused an overflow in strutil > instead of an 1.#inf. > I took the chance to rework this quite a little, with the result of > … [View More]

2 2

pirate list
by Michal Wallace July 13, 2005

July 13, 2005

Hello, PyPy list :) I've just started a new list for pirate, a python compiler for the parrot virtual machine. Pirate is written in python and currently uses the standard compiler package. However, as parrot supports native types, I would like to refactor it so that it can also work with PyPy's type annotation system (and possibly integrate with pypy in other ways) The pirate website is here: http://pirate.tangentcode.com/ Quite a bit of work has been done on pirate since I last … [View More]

2 3

Re: [pypy-dev] Refining terminology
by holger krekel July 12, 2005

July 12, 2005

Hi Bengt, On Tue, Jul 12, 2005 at 11:00 -0700, Bengt Richter wrote: > At 12:19 2005-07-12 +0200, you (holger krekel) wrote: > >Any opinions or feedback (also from people just following > >the project's progress)? > I think any project tends to accumulate informal docs re key concepts > and implementation details, but they tend to be hard to find until > someone collects good links and the good-link collection becomes > available and advertised via some standard … [View More]

1 0

Refining terminology
by holger krekel July 12, 2005

July 12, 2005

Hi folks, i have just started to write some documentation about PyPy's bytecode interpreter (only the overview chapter in fact see http://codespeak.net/pypy/index.cgi?doc/interpreter.html ). And i realized that i'd like to push to refine the terminology we are using throughout the PyPy documentation and our presentations. For starters, i think that it does not make sense to fight against the common notion of an "interpreter" refering to the whole thing (our bytecode interpreter + … [View More]

3 4

Please give advice for newbie reading pypy source......
by seberino＠spawar.navy.mil July 12, 2005

July 12, 2005

PyPy Developers: I'm really excited about learning PyPy source code really well and possibly helping the project in some capacity. Reading source code inevitably leads to tons of questions. **Any advice on how to progress quickly in PyPy source code understanding?** Any chance there are PyPy people close to San Diego, California?? Any better method than posting questions to this mailing list and slowly getting replies to all my questions? Any help would be greatly appreciated. Chris -- … [View More]

2 1

left in irc
by Laura Creighton July 9, 2005

July 9, 2005

amep (~amp(a)user-12lcgki.cable.mindspring.com) has joined channel #pypy <amep> I wanted to say how cool I think pypy is. And how much I'm looking forward to a complete release. I'm writing a program that would really be improved by some extra speed and I'm sure pypy will be able to provide that. (Machine code output and all) amep has left channel #pypy Laura

1 0

os.getuid
by Ben.Young＠risk.sungard.com July 5, 2005

July 5, 2005

Hello PyPyers! I was just wondering if anyone had noticed that on windows (and cygwin) os.getuid doesn't actaully exist, and so I am getting lots of failures. I see that great progress is being made anyhow! Is there a status report due out a the end of the sprint so us observers can see what has been done? Cheers, Ben

2 3