Mailman 3 June 2004 - Python-Dev

RE: [Python-Dev] Comparing heterogeneous types
by Chermside, Michael June 23, 2004

June 23, 2004

> If "a <= b" is to mean we convert a and b to either float or > long depending on > their magnitude, would it make any sense at all if other > operators like "a + > b" would do the same, to maximize precision? IMHO, probably not. We are NOT suggesting that "a <= b" mean we convert a and b to float or long depending on magnitude, we are suggesting that is just means to compare the values... and that if one value is float than it should be interpreted as if it had the precise value it represents rather than some range of values. Sure, we talked about converting things, but that was an *implementation* detail. I suspect that the reason Guido prefers "a + b" to always return a float when one is a float and the other is a long is because of his (recently noted in another thread) preference for operations to be consistent in what type they return. Of course, it COULD be done more accurately using your approach. The only good reason I can come up with for NOT doing so is to have the rule for what type results from arithmatic operations on mixed types be a simple one. -- Michael Chermside This email may contain confidential or privileged information. If you believe you have received the message in error, please notify the sender and delete the message without copying or disclosing it.

1 0

Re: [Python-Dev] VC 7.1 maintenance?
by M.-A. Lemburg June 23, 2004

June 23, 2004

Tim Peters wrote: > [M.-A. Lemburg] > >>Are we going to have a VC7.1 binary of Python 2.4 for Windows ? > > > That's indeed my understanding of The Plan today. More, we will not have a > VC6 binary for Python 2.4. That may be controversial, but nobody has > volunteered to produce an installer for a VC6 version. Perhaps ActiveState can fill this gap ?! >>(Just asking whether it's time to finally open the VC7 shrink- >> wrap box ;-) > > Yup. IIRC, it took about 3GB of disk space to do a full install (including > copying all the docs to disk). After it's installed, you won't recognize > it, and will feel completely lost. This is a learning experience <wink>. Hmm, looks like we'll have to get yet another box contributing to the global heating then... this Wintel alliance sure works well ;-) Thanks, -- Marc-Andre Lemburg eGenix.com Professional Python Services directly from the Source (#1, Jun 16 2004) >>> Python/Zope Consulting and Support ... http://www.egenix.com/ >>> mxODBC.Zope.Database.Adapter ... http://zope.egenix.com/ >>> mxODBC, mxDateTime, mxTextTools ... http://python.egenix.com/ ________________________________________________________________________ ::: Try mxODBC.Zope.DA for Windows,Linux,Solaris,FreeBSD for free ! ::::

2 1

Re: [Python-Dev] mimetypes and _winreg
by Garth June 22, 2004

June 22, 2004

Guido van Rossum wrote: >>I could file a patch if no one else is looking at it. The solution would >>be to use >> >>RegEnumKeyEx and remove RegQueryInfoKey. This loses compatability >>with win16 which I guess is ok. >> >> > >Indeed. Would you mind filing a patch? (If you've already done so, >my apologies -- do you know the patch #?) > >--Guido van Rossum (home page: http://www.python.org/~guido/) > > > Ok done it here. patch number 977553. https://sourceforge.net/tracker/index.php?func=detail&aid=977553&group_id=5… Some bench mark values for the script below:- python 2.3 5531 89.7130000591 python 2.4 +patch 5531 0.0469999313354 start = time.time() i = 0 try: while 1: _winreg.EnumKey(_winreg.HKEY_CLASSES_ROOT, i) i += 1 except WindowsError: pass print i, time.time() - start The max size of a key was taken from the MS Platform SDK. I hope you don't mind the 255 byte array on the stack? Not a recursive function so it shouldn't matter too much. I've noticed a few things while looking at the winreg module. * No unicode support * This patch concept could be applied to other functions in this module to speed up stuff. May sort out the rest of this but not soon as I'm off to Glastonbury tomorrow so if there's any problems I may not respond for a while. Garth

1 0

Re: [Python-Dev] VC 7.1 maintenance?
by Christian Tismer June 22, 2004

June 22, 2004

[for unknown reasons, this message didn't make it to the list] Tim Peters wrote: > [M.-A. Lemburg] >>(Just asking whether it's time to finally open the VC7 shrink- >> wrap box ;-) > > > Yup. IIRC, it took about 3GB of disk space to do a full install (including > copying all the docs to disk). After it's installed, you won't recognize > it, and will feel completely lost. This is a learning experience <wink>. I had to do this, primarily to try and compile Prothon, and after half a day of playing with the thing, I think it is better to use than VC 6. Some keyboard shortcuts are gone, unfortunately, but those for the debugger are still there (sigh). One tremendous improvement in the GUI: You can now decide which windows are docked and which not, and the docking concept is now so universal, that I begin to like it. Every window can split itself, becoming a container for more windows, it can become tabbed, you can drag windows out, and on and on. This is really much mre handy. Before, I always had trouble with my variable inspector windows. Touching them wrong, they would snap into a place where I didn't want them. And, now you can have as many VC instances as you like, which is *very* handy when testing different versions against each other. Does anybody have experience with the newly exposed COM model for debugging? I'd like to write an add-on that allows to inspect PyObjects more naturally, likewise using __repr__ calls to show PyObjects while debugging the interpreter. ciao - chris -- Christian Tismer :^) <mailto:tismer@stackless.com> Mission Impossible 5oftware : 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 89 09 53 34 home +49 30 802 86 56 mobile +49 173 24 18 776 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

Re: [Python-Dev] VC 7.1 maintenance?
by Guido van Rossum June 22, 2004

June 22, 2004

Another question about the VS7.1 project files: these are XML text, but they seem to be checked in as binary, therefore CVS won't even attempt merges. Is there a reason they can't be marked as text? Surely the old VC 6.0 reason for this is no longer valid now it's XML? --Guido van Rossum (home page: http://www.python.org/~guido/)

5 8

RE: [Python-Dev] Allowing u.encode() to return non-strings
by Chermside, Michael June 21, 2004

June 21, 2004

> When looking for near-C speed, type inferencing is most important > for a relatively small set of particularly efficiently manipulable > types: most notably, smallish integers. Being able to prove that > something is a Unicode object just isn't all that useful for > efficiency, because most of the things you can do to Unicode > objects aren't all that cheap relative to the cost of finding out > what they are. Likewise, though perhaps a bit less so, for being > able to prove that something is a string. I think you may be missing an important special case here... machines for which "unicode" is a machine-native type. Like the JVM or .NET runtime. These machines are increasingly important, and one of the two major implementations of Python runs on such a machine. (I can't say "2 out of 3" because IronPython doesn't count until it's released publicly.) Of course, I *still* don't think that it makes all that much difference. I agree with the folks who are saying that the "functions generally return a known type" rule is useful not so much because of optimization as because it helps the readability of code. (And that's why obvious exceptions like construction factories don't bother anybody.) -- Michael Chermside This email may contain confidential or privileged information. If you believe you have received the message in error, please notify the sender and delete the message without copying or disclosing it.

1 0

Re: [Python-Dev] VC 7.1 maintenance?
by Christian Tismer June 21, 2004

June 21, 2004

Tim Peters wrote: > [M.-A. Lemburg] >>(Just asking whether it's time to finally open the VC7 shrink- >> wrap box ;-) > > > Yup. IIRC, it took about 3GB of disk space to do a full install (including > copying all the docs to disk). After it's installed, you won't recognize > it, and will feel completely lost. This is a learning experience <wink>. I had to do this, primarily to try and compile Prothon, and after half a day of playing with the thing, I think it is better to use than VC 6. Some keyboard shortcuts are gone, unfortunately, but those for the debugger are still there (sigh). One tremendous improvement in the GUI: You can now decide which windows are docked and which not, and the docking concept is now so universal, that I begin to like it. Every window can split itself, becoming a container for more windows, it can become tabbed, you can drag windows out, and on and on. This is really much mre handy. Before, I always had trouble with my variable inspector windows. Touching them wrong, they would snap into a place where I didn't want them. And, now you can have as many VC instances as you like, which is *very* handy when testing different versions against each other. Does anybody have experience with the newly exposed COM model for debugging? I'd like to write an add-on that allows to inspect PyObjects more naturally, likewise using __repr__ calls to show PyObjects while debugging the interpreter. ciao - chris -- Christian Tismer :^) <mailto:tismer@stackless.com> Mission Impossible 5oftware : 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 89 09 53 34 home +49 30 802 86 56 mobile +49 173 24 18 776 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

RE: [Python-Dev] Re: VC 7.1 maintenance?
by Moore, Paul June 21, 2004

June 21, 2004

[I never saw this message on python-dev - apologies if others see it twice...] From: Ivan R. Judson >> Is VC2003 a fancier name for VC7.1? or at least binary compatible? >> Will this allow people to compile extensions without $$ to MS? > The free download is the compiler and linker and basic tools to build apps. > The part that costs money is the Visual Studio Development Environment (the > IDE and higher level tools), however I can find the concrete details if you > want them. The problem for Python development is that the free download VC doesn't support linking to the DLL verstion of the CRT (msvcr71). This means that there will still be CRT incompatibilities if extensions are built with this compiler :-( Mingw is another option, as the latest version of mingw allows building DLLs which link to msvcr71, but it's pretty bleeding-edge, and I haven't been able to do much work yet on proving the relevant distutils support is there, as I don't have an up-to-date Python binary build. I'm planning on doing some testing when 2.4a1 comes out... Paul. __________________________________________________________________________ This e-mail and the documents attached are confidential and intended solely for the addressee; it may also be privileged. If you receive this e-mail in error, please notify the sender immediately and destroy it. As its integrity cannot be secured on the Internet, the Atos Origin group liability cannot be triggered for the message content. Although the sender endeavours to maintain a computer virus-free network, the sender does not warrant that this transmission is virus-free and will not be liable for any damages resulting from any virus transmitted. __________________________________________________________________________

1 0

Re: Patch #945642, was: Re: [Python-Dev] release23-maint tree closed for 2.3.4rc1
by Tino Lange June 20, 2004

June 20, 2004

Anthony Baxter wrote: > Well, I'd recommend you chase people up who've expressed an interest in > the patch and make sure one of them checks it in. I will try to get to > it in a couple of weeks, but my time is (as always) somewhat limited, > and at the moment I'm ramping up for the 2.4 release cycle. Hi all - may I remind you of patch #945642? It would be really nice to have it revised and checked in for 2.3.5 and 2.4. Thank you a lot! Tino

1 0

functions decorators in 2.4?
by Martin Zarate June 20, 2004

June 20, 2004

Why not use the underused "as" keyword for decorators? That is, def foo(bar, baz) as StaticMethod: It would be very handy for classes, as I've worked with custom class structures (such as an overcomplicated interface-based structure that I was experimenting with) that it would be useful for. Are there plans to extend this to fully custom blocks and statements like in Ruby? Or will it only be for modified def and class statements?

3 2