🌐 Mailman 3 April 2006 - Python-list - python.org

installing pyodbc
by timw.google May 1, 2006

May 1, 2006

I just downloaded the pyodbc source to try and install on my Linux FC3 box. I see that there is a setup.py file, but when I try to do a 'python setup.py build' (or just 'python setup.py') I get Traceback (most recent call last): File "setup.py", line 27, in ? revision = latest_revision('.') File "setup.py", line 25, in latest_revision return int(output) ValueError: invalid literal for int(): exported The README.txt doesn't really say anything about how to install, exept that the easiest way is to use the Windows installer. That's fine for Windows, but what about Linux? Thanks for any help.

2 2

Re: stdin: processing characters
by Edward Elliott May 1, 2006

May 1, 2006

Kevin Simmons wrote: > I have a python script that prompts the user for input from stdin via a > menu. I want to process that input when the user types in two characters > and not have to have the user press <CR>. As a comparison, in the bash > shell one can use (read -n 2 -p "-->" CHOICE; case $CHOICE in...). Works > great and is very I did something like this a couple years ago, curses was the easiest way I found to do it. It's pretty painless with the wrapper function, which restores your terminal on error/exit.

4 3

wxpython - new line in radiobutton's label?
by nuffnough＠gmail.com May 1, 2006

May 1, 2006

Hi. I am creating a small gui using wxGlade and wxPython. I am using a radio button, and the text next to the options needs to be over two lines. I tried using a \n character, but it seems to get ignored. Looking around in the various wikis and the example code also didn't turn up anything. TIA

2 1

How to efficiently read binary files?
by David Lees May 1, 2006

May 1, 2006

I want to process large binary files (>2GB) in Python. I have played around with prototypes in pure Python and profiled the code. Most of the time seems to be spent converting back and forth to and from strings using the struct module. Is there a way to directly read into an array of integers in Python? TIA David Lees

4 4

ANN: nose 0.9.0a1
by jpellerin+nose＠gmail.com April 30, 2006

April 30, 2006

I'm pleased to announce the 0.9.0a1 release of nose, a test runner that brings py.test convenience to unittest, without resorting to (too much) magic. nose 0.9 includes a host of new features, as well as numerous backwards-incompatible changes to interfaces and implementation. For this reason, I'm releasing it first as an alpha version. Thanks to the many folks who have contributed patches and ideas and made bug reports for the development version of 0.9, especially Mika Eloranta, Jay Parlar, Kevin Dangoor, Scot Doyle and Philip J. Eby. Here's a quick rundown of what's new in 0.9.0a1. - Plugins The most important new feature is support for plugins using setuptools entrypoints. nose plugins can select and load tests (like the builtin doctest plugin), reject tests (like the builtin attrib plugin, contributed by Mika Eloranta, that allows users to select tests by attribute), watch and report on tests (like the builtin coverage and profiler plugins), completely replace test result output (like the html result plugin in the examples directory) or any combination of the above. Writing plugins is simple: subclass nose.plugins.Plugin and implement any of the methods in nose.plugins.IPluginInterface. - Better compatibility with unittest Test loading has been consolidated into a test loader class that is drop-in compatible with unittest.TestLoader. Likewise test result output, including output capture, assert introspection, and support for skipped and deprecated tests, in nose.result.TextTestResult. If you want those features and not the rest of nose, you can use just those classes. nose.main() has also been rewritten to have the same signature as unittest.main(). - Better command line interface Command line test selection is more intuitive and powerful, enabling easy and correct running of single tests while ensuring that fixtures (setup and teardown) are correctly executed at all levels. No more -f -m or -o options; now simply specify the tests to run: nosetests this/file.py that.module Tests may be specified down to the callable: nosetests this/file.py:TestClass that.module:this_test nosetests that.module:TestClass.test_method There are also new options for dropping into pdb on errors or failures, and stopping the test run on the first error or failure (thanks to Kevin Dangoor for the idea). - More! Helpful test decorators and functions in nose.tools. Support for generators in test classes. Better import path handling -- that you can shut off! Detailed verbose logging using the logging package. And more... For more information, installation instructions, etc, see: http://somethingaboutorange.com/mrl/projects/nose/

1 0

Re: setting file permissions on a web server
by John Salerno April 30, 2006

April 30, 2006

Daniel Nogradi wrote: > This depends on your arrangements with your web server provider. > Perhaps you are allowed to ssh into that machine, perhaps not, you > need to ask your provider. In case you can use ssh, then you can log > in with putty (an ssh client for windows, grab it from here: > http://www.chiark.greenend.org.uk/~sgtatham/putty/) and you will have > a command line on the server so you can run chmod there. Turns out that my web hosting company provides ssh access on the next highest plan and up, but not mine. :) Oh well, I think I'll explore FileZilla a little more and see if there's more of an automated way to do it than just right-clicking each file (or even right-clicking multiple files at once, if possible). It shouldn't be too big of a deal anyway, but I'm always after more automated ways to do things...call me lazy! :)

2 1

From the Tutorial Wiki: suggested restructure
by Gerard Flanagan April 30, 2006

April 30, 2006

For anyone interested, here's a blog entry I've added over at the Tutorial Wiki: ---------------------------------------- As mentioned by effbot in the previous post I've put together a suggested alternative factoring of the Tutorial here(http://pytut.infogami.com/gerard_refactor). At the minute it just amounts to a carving-up of the existing material into two sections: beginners and intermediate. It's a bit stilted in places, but not excessively. There's also a suggested advanced section - I'm not an advanced user myself so I don't really know what would be appropriate here. Each section in the beginners tutorial has links to relevant FAQS at pyfaq.infogami.com. I also thought the FAQS could be preceded by suitable examples, and maybe there could be a final chapter of exercises. I've been sitting for half an hour trying to articulate why I think this is a good idea, but the well has dried up...and I seem to have gone a bit snow-blind looking at the blank screen! I just remember that when I was learning Python, although the Tutorial was incredibly useful and a key source of info, it seemed that I always had to hunt excessively when I had questions like 'how does **kwargs work?', 'what does extend do again?' etc. - I always thought that things weren't ordered in the best way, and the chapter titles weren't that helpful. so this is how I think it should be done! I hope it's a positive contribution - maybe it will provoke discussion if nothing else. Unfortunately, I won't have easy access to a computer for the next few months, so I can't contribute any more at this point. ------------------------------------------------------------------ Comments welcome. Or contributions - particularly to the advanced section. All the best. Gerard

1 0

WAP page, Python and creating a file on file system - Newbie
by Digital April 30, 2006

April 30, 2006

HELP!! All I need to do is to create a file in a folder. I am running Abyss WEB server and Python. and googling many sites has taught me how to create a file in a folder. BUT. I need to do this from a WML file - This is proving difficult... I can setup an index.wml and a deck of cards. Each card points to a <something>.py file. This file creates a file in the folder of my choice but always returns an error to the WAP browser. I have tried PRINTing WML content within the Python code and thought I had formed a WML output but still the WAP browser returns an error. It seems an awful lot of trouble to go to, load Python to create a file in a folder. Is there any way I can easily do this? Before people reply with how to do stuff in OOP. I have no knowledge of OOP and class instantiation etc etc. I really do need a noddy guide in what I am doing. Regards Dave

1 0

Re: setting file permissions on a web server
by Kirk McDonald April 30, 2006

April 30, 2006

Daniel Nogradi wrote: > I have next to zero experience with windows but as far as I know > windows doesn't have file permissions at all (anyone, please correct > me if I'm wrong :)) so in windows land it doesn't make any sense to > "change file permissions". Even if it has some sort of a notion of > file permissions I wouldn't know how that gets translated into unix. This is getting a little off-topic, I admit, but the NT-derived versions of Windows do indeed have file permissions. The Windows-equivalent of chmod (and chown) is cacls ("Change Access Control Lists"): http://www.ss64.com/nt/cacls.html In essence, each file has an owner and a list of other users and groups, who may each have "Full control" (meaning they can change permissions), read-only, write-only, read-write, or no access. Windows 95 and its descendants don't have permissions any more than DOS does. (Except with respect to Windows network file-sharing.) (Heh, I checked just before posting this and someone beat me to it. Here's my post anyway.) :-) -Kirk McDonald

1 0

Re: setting file permissions on a web server
by Edward Elliott April 30, 2006

April 30, 2006

Daniel Nogradi wrote: > I have next to zero experience with windows but as far as I know > windows doesn't have file permissions at all (anyone, please correct > me if I'm wrong :)) so in windows land it doesn't make any sense to > "change file permissions". Actually Windows has a quite good permission scheme, with full owner/group ACLs and granular permissions for read, write, append, delete, and several others I can't recall. The problem is no one uses them much. The manipulation tools are a pain, and developers routinely ignore them (including Microsoft's own developers, although I hear they've gotten much better lately). > I guess so. But that's not a terribly big problem even if you can't > use ssh. What I would do is write a script (in python of course :)) > that does the file permission changing and run that script over the > web. Since this can have serious security implications my strategy > would be to place this script somewhere which is not reachable through > the web and only relocate it to a web accessible directory when you > want to run it over the web, and when you are done, you place it back > to its secure location so nobody can reach it. You're right about the security concerns, but your solution sounds like more trouble than it's worth. With a bit more effort, you could write a local python script that acts as an ftp client and applies permissions remotely. But why bother when a good ftp client already has that built-in (and is much less likely to contain destructive bugs). > You can find out how to > write this script from http://docs.python.org/lib/os-file-dir.html You're worried about security and you turn a novice with a scripting language loose in the briar patch of unix permissions, on a remote server no less? And I thought I was sadistic! :)

1 0