🌐 Mailman 3 May 2000 - Python-list - python.org

Python, Glade, & GTK
by Mafee June 5, 2000

June 5, 2000

Hi all, Can anyone give me some clues of linking python code to a GUI interface generate by Glade or GTK or whatever please? Thanks in advance =) mailto:mathew_chan@hotmail.com

4 3

Pointrel Data Repository System for Python
by Paul Fernhout June 4, 2000

June 4, 2000

This is the very first (rough!) version of the Pointrel Data Repository System. It is available for download here under an MIT/X11 (Python-ish/Software Carpentry-ish) type license from this web page: http://www.kurtz-fernhout.com/pointrel/ One example (versioned file system viewer) requires wxPython, but otherwise the code just uses plain Python. See the web page for more details. The Pointrel Data Repository System is a loosely coupled set of concepts for data representation. It can be thought of as an information storage architecture. The code presented here implements the simplest concepts in this system. William Kent's ROSE/STAR system described in his 1979 book "Data and Reality" is similar to the Pointrel Data Repository System in intent in many ways. The Pointrel Data Repository System includes a triadal data storage system. Using triads, one can build arbitrary complex networks of relationships. These dynamic relationships can define the equivalent of records or objects in a database. A primary concept in the Pointrel Data Repository System is a triad. A triad is an object with three links and an optional string. These links can point to other triads. This allows one to build arbitrarily complex structures, as well as add to these structures at any time. What is distributed here is one possible implementation in Python of this "idea", which I have been playing around with in various languages for about twenty years. I've been going for an elegant compromise between flexibility, speed, and simplicity in storing and retrieving information. The Pointrel system aims at having a small number of unifying concepts that can be arranged in endless ways to build layers of complexity above them. I'm also trying to avoid the bloat where it takes 100MB and endless DLLs just to install a database system, let alone use it for anything. The core Pointrel source is around 20K. That makes it nicer for embedded systems and stand-alone programs. To do (for anyone who wants to help): * Integrate with IDLE for version control / version differencing. * Integrate with Zope for general repository use (bug tracking, source code tracking, file sharing, etc.). * Make into a HTTP server version both using some sort of binary Python query object and using plain text CGI like calls. * Make wxPython interface to file system work under TK. * General graphical interface for manipulating triads and exploring the repository. * Lots of improvements to the versioning structure in the file system (wxPython example shows the content of each version of a python script, not a version name). * Make a CVS look-alike interface. * Use as a caching proxy for web page downloading. * Use to store and retrieve mail and newsgroups. * Use as part of a distributed file/knowledge system. * Create a suite of export / import Python scripts and related viewers. * Make a simple front-end that makes it look like a familiar "relational" database with tables. * Flesh out the transaction support. * Use it for implementing tuple-spaces. * Make optimized C version w/DLL. * Build optimized C version into base Python.exe so *.py libraries, *.pyc, and maybe DLLs can all be in one common file. A FAQ on deletion: you can only add to the Pointrel repository (not delete). Deletion needs to be handled at a meta level, by tagging items as deleted (hidden) and looking for the tag in an appllication using the data, by creating new indexes without referencing deleted items, or by cloning the entire repository and leaving behind data you don't want in the new copy. The system is still in flux, so think of it as more of an experiment in this version and a way to get useability & conceptual feedback. Comments appreciated (including pointers to similar work). -Paul Fernhout Kurtz-Fernhout Software ========================================================= Developers of custom software and educational simulations Creators of the Garden with Insight(TM) garden simulator http://www.kurtz-fernhout.com

3 4

PyXML-0.5.4 and Site-Packages
by thomas＠cintra.no June 4, 2000

June 4, 2000

Hi, I`ve tried to install PyXML-0.5.4 ( actually PyXML-0.5.3 and python-xml-0.5.1.rpm too with same result ) on my Mandrake 7.0 Linux-box without luck. No erros during python setup.py build or setup.py install. The xml-folder has a __init__.py file with the __all__ : .... -line. When I try to use anything but the xml-folder, like "from xml.sax import *" I get a "no module named sax"-error. Then I tried to add a xml.pth in the site-packages-folder, looking like : xml xml/parsers xml/sax xml/unicode ... No change. I`ve tried to install this package before without luck. I just gave up in the end. But now it`s a vital part of a project, not some private messing around, so I need it to work. Is there a way around the site-package method of doing it? Any tips, hints? Thanks. Thomas

2 1

ANNOUNCEMENT: PythonWorks 1.0 now available (may 31, 2000)
by PythonWare June 4, 2000

June 4, 2000

it's time for a major release from the labs: PythonWorks is a rapid application development tool, designed for efficient development in Python. PythonWorks includes editors, a debugger, a user-interface design tool, project browser, deploy- ment tools, integrated on-line help, and a scriptable environment. To learn more, visit the PythonWorks product site: http://www.pythonware.com/products/works If you order now, you get: - PythonWorks 1.0 for Windows 9X/NT/2000 (more platforms 3Q00) - a personal license for PythonWorks 1.X on all available platforms - free upgrades throughout the entire 1.X series - support via mail For more information, please visit the product site, or mail us at info(a)pythonware.com. regards, the pythonworks team "Secret Labs -- makers of fine pythonware since 1997."

7 8

A Mountain of Perl Books + Python Advocacy
by lewst June 3, 2000

June 3, 2000

Before I launch into another question and gripe, I'd like to thank everyone who offered comments and suggestions on the advanced Python books that are out there. Now something I can't quite figure out is: why are there so many more books on Perl out there than on Python? Searching through Fatbrain.COM (which is where I order my books from), I found 68 books, 4 training manuals, and 2 eMatter documents on Perl. Compare this to Python's 13 books and 2 eMatter documents. What is it about Perl that makes it so much more popular and have such a huge grassroots swell? I personally find Perl an abomination and Python a breath of fresh air. Perl has that first mover advantage I suppose, but should that really make such a hugh difference? I'll admit that Perl is what led me to Python in the first place. After hearing about how great Perl was several years ago from the local sysadmin, I learned it and started using it for my scripting needs. If there is one thing that sums up my Perl experience, it is that it always took me too long write the programs that I needed to write. The syntax was always no unnatural for me that I could never get my head out of the reference manual, and errors were always so tough to track down. In the end was frustration albeit a working result. This frustration led me to look into Python and I'll never touch Perl again. Python was so natural for me I often found myself correctly "guessing" at the syntax as I learned it. I rewrote all my Perl scripts in one weekend and most of them worked on the first try. At this point I wondered if my brain was just different than all those Perl junkies out there. But now I really don't think so; I think it's a question of awareness. Perl is very publicized and well-known while the better language is sitting here a dark corner unnoticed. Sure there will always be some religious fanatics that won't even give Python a try, but I think Python's popularity could be vastly improved with some serious advocacy work. CNRI and/or PSA should seriously look into funding a Python "marketing" campaign of sorts. I think the result would be allot of converts and more understanding and respect for Python. With this in mind, let me include one of my favorite pro-Python quotes of all time. This is from a message to the fetchmail-announce mailing list by Eric S. Raymond <http://tuxedo.org/~esr>, the famous open-source advocate and author of many popular software programs. He is discussing "fetchmailconf" which is the Tkinter GUI for his fetchmail POP/IMAP mail client. I think it demonstrates the point I make above perfectly. "A note about fetchmailconf. It took me approximately six days to write this elaborate multi-paneled GUI -- that's counting the four days it took me to learn the implementation language in the process. This could easily have been a two-month project in C (with six weeks of that spent debugging and bugs still left). Or a week-long project in Perl, with working but ugly and unmaintainable results." "The verdict: Python is *waaaay* cool! I'm sold. It's clean, it's elegant, it's easy, and it's astonishingly powerful. I'm not going to program anything longer than one screenful of script in Perl anymore. I love Larry Wall dearly, but Guido van Rossum is the better designer -- I haven't had this much fun with a language since the glory days of LISP. Eric sez check it out." Regards. __________________________________________________ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com

54 84

Is it just Syntactic Sugar ?
by Eric Hagemann June 3, 2000

June 3, 2000

Based on some recent code I have been writing I am courious on describing the following as 'syntactic sugar' a common operation is to a=a+1 which in C can be written as a+=1 in its simplest form I agree there might not be much to this but if you do this with a complex expression as a[x][f][g] = a[x][f][g] + 1 does the python engine recompute the 'address' of the variable 'a[x][f][g]' twice ? Would the a+=1 for speed things up or is it just expanded to a=a+1 anyway ? I have found that p=a[x][f][g] p=p+1 is faster than above so I kinda believe the 'address' is computed twice (I am leaping here) Comments / Thoughts ? Cheers Eric P.S. The application, if you are wondering is to generate counts of events in log files the events have various dimensions such as date/time/orgin that I catalog them by

12 19

Timeout using tkinter?
by digitig＠cix.co.uk June 2, 2000

June 2, 2000

Using Python & tkinter, is it possible to build a (platform independent) time-out on a user response?

2 3

Using python on the web
by Charley Horse June 2, 2000

June 2, 2000

I've just started fooling with python. I've looked around the python site and a number of the other sites referenced there and am a bit puzzled. Umm... how does one go about using python for web applications, especially database driven apps? Almost everything deals with python just as a language, as C++ would be. Not much about the web specifically. I was a bit shocked to find the only direct refs on about this at python.org are entwined with CGI (unless I missed it). I have no exposure to CGI, but rather to php, CF, and ASP, so python has gone from looking very attractive to looking... well, like a puzzle. I know about Zope, but am interested in how one uses python on the web just by itself. Thanks

20 25

starship status?
by Michael Hudson June 2, 2000

June 2, 2000

Is starship now more or less functional? I tried to post to crew@starship, but didn't get it back... If it is mostly working, perhaps it's time to replace the current rather forbidding front page with something less likely to scare off the passer-by; some links to the rest of the starship would be a nice start! Also, are there any plans to get zope up again soon? (no particular reason for asking...). Cheers, M. -- There are two ways of constructing a software design: one way is to make it so simple that there are obviously no deficiencies and the other way is to make it so complicated that there are no obvious deficiencies. -- C. A. R. Hoare

1 1

Re: The REALLY bad thing about Python lists ..
by Glyph Lefkowitz June 2, 2000

June 2, 2000

"Tim Peters" <tim_one(a)email.msn.com> writes: > [Tim] > >> ISO/ANSI C doesn't fully define what happens when (roughly) > >> integer division or % are handed negative numbers. > > [Glyph Lefkowitz] > > The foibles of standards commitees will never cease to amuse me. > > Thank you for this little tidbit. > > The next iteration of C has threatened to "fix" this, by mandating Fortran's > original mistake (truncation, as opposed to Python's floor). By "threatened" you don't mean "already mandated but it hasn't been implemented yet", do you? Please say this is still open for discussion :-) > BTW, have you "served" on a stds committee? It's, umm, educational. [ comittee-nonsense snipped ] Wow. Come see the ANSI Standards Comittees perform "USENET, live and in concert!"... although the anecdotes I've heard about the Common Lisp and ANSI C standards committees jibe with this altogether too much :-\ > > Do you actually have a basis for that statistic? > > The only *hard* statistic Guido has is the download stats from > python.org, where the number of Windows downloads routinely beats > the number of all others combined (last stats I saw were from last > August, where Windows was "winning" by an easy factor of 2) Well, speaking of lame anecdotes, :), now that I'm writing software in python, this is the way my conversations go with potential users: Me - "I've got some more of the code done. Do you guys want to try it out?" Linux User - "Sure. Do I need python installed?" Me - "Type 'python'. Does anything happen?" Linux User - "Yeah, a >>> prompt comes up." Me - "OK, just download the software." (5 minutes later) Linux User - "Got the file; ... what do I do with it?" Me - "On the command line, tar xvzf mysoftware.tar.gz, then cd mysoftware; ./run" Linux User - "OK, it's running. Thanks!" Windows User - "What do I need?" Me - "Well, first you'll need WinZip ... download it from www.winzip.com" (20 minutes later) Windows User - "OK installed. What else do I need?" Me - "You need Python: go to www.python.org, download it, and install it." (20 minutes later) Windows User - "Grr. OK installed. It asked me about tcl/tk but I didn't install that." Me - "No, you need that. Uninstall and then reinstall and install that too." (5 minutes later) Windows User - "Augh! Done. OK now what?" Me - "Download mysoftware.zip, and double-click on it. That should open up WinZip...." (the windows user will almost always become frustrated at this point and forget about it, being "just a user" and confused by all this high-falutin terminology of "archives" and "virtual machines", assuming that WinZip works at all. I need to build a friggin' installer soon so this stops happening.) The point being, of course, more Linux users have tar than Windows users have WinZip, and more Linux users have Python than Windows users do; but I am sure that GNU tar is downloaded less than WinZip. I have yet to encounter a Linux user (regardless of distribution) who doesn't have python installed, although most don't know that it's there. Given how many really useful utilities are written in it, and how many environments depend on it, it's becoming a basic component of most modern linux distros. > not-admitting-that-a-little-known-part-of-python-startup-sends- > email-to-python.org-whenever-python's-invoked-ly y'rs - tim putting-python-in-a-chroot-jail-with-no-network-access- from-now-on-(and-i'd-like-to-see-you-do-that-on -windows-<wink>)-ly y'rs, -- __________________________________________ | ______ __ __ _____ _ _ | | | ____ | \_/ |_____] |_____| | | |_____| |_____ | | | | | | @ t w i s t e d m a t r i x . c o m | | http://www.twistedmatrix.com/~glyph/ | `__________________________________________'

24 47