| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Comments:
<ironfroggy> Eleaf: ive been walking Wolfman2000 through the implementation of a standard playing deck of cards, for a poker game he is writing.
<ironfroggy> ive taken the opportunity to build code and plot for a series of python tutorials ill be writing, and maybe some future cl***es i want to look into starting.
<Eleaf> cool
<ironfroggy> http://deadbeefbabe.org/paste/525
<ironfroggy> Eleaf: we're using pastebin for versioning of the code :-)
<Eleaf> awesome
<Eleaf> can you do magic tricks with it? ;p
<Wolfman2000> Don't push your luck
<Eleaf> I think you probably could..
<ironfroggy> what kind of tricks?
<ahmeni> 52 pickup!
<Bison> anyone know where to find docs on setup.py?
<ironfroggy> the internet
<Bison> ...
<ironfroggy> Wolfman2000: let me know when you get back, have somethings to go over.
<ironfroggy> google it Bison
<Bison> ...can't find it :(
<ironfroggy> look for distutils documentation
<Bison> okay
<Bison> woohoo!
<darkgreen> is PEAK or !TransWarp the way to go for AOP in python?
<darkgreen> looks like PEAK.
<kbrooks> what is aop?
<ironfroggy> Aspect Oriented Programming
<ironfroggy> the general idea as best as ive grasped is that things' behaviors are contextual to aspect in which you are using them. i dont really know what that means.
<babbitt> is there a way to make a temporary file that is accessable by another application?
<ironfroggy> babbitt: look at the tempfile module?
<babbitt> ironfroggy, I did, it looks like everything there is completely locked down
<babbitt> ironfroggy, oh..nevermind, its locked to user id, not process ID. I was wondering how they would have locked it to process ID
<deltab> on unix, the tmeporary file could be deleted as soon as it's been created; then only the process with it open would have access
<babbitt> yea, but I /want/ another process to have access to it :)
<babbitt> but I"Ve got it now
<babbitt> sorry for being stupid
<cantares> how kann i break a long int in several 1 byte ints for packing it into a tupel?
<sewi> hello, how to uninstall a python module after python setup.py install?
<Erwin> cantares: Why would you want this?
<deltab> babbitt: that's how it'd be done
<deltab> cantares: is this for use on an IP address?
<deltab> sewi: delete it
<cantares> I habe a big tupel with around 4 million 1 byte ints, and i want to XOR that with some other data which is in my script inside a realy big long int O.o
<sewi> deltab: but what if installation copied some other files than in /usr/lib/python2.4/site-modules/MODULE?
<babbitt> this is wierd, I'm using XMLRPC and a string keeps getting sent back to me double-base64 encoded
<deltab> cantares: why do you have that tuple?
<deltab> sewi: it doesn't provide an automatic way of deleting those
<cantares> cause i have read a binary file and i upacked it using the struct module into that tuple.
<sewi> deltab: ok, thanks
<deltab> cantares: why?
<deltab> cantares: why unpack it into a tuple?
<cantares> cause i don't know how to unpack it into one big int ? O.o
<deltab> long(s.encode('hex'))
<deltab> er
<deltab> long('0x' + s.encode('hex'))
<deltab> hmm, no, doesn't work
<deltab> long(s.encode('hex'), 16)
<kbrooks> deltab: :-)
<cantares> deltab: er, i have read the file via filedata = file.read() .. so i have a 4MB string.. where in your construct should i input my var. filedata?
<deltab> yes
<MFen> a 4MB string sounds like a good place to have a database instead
<cantares> it isn't realy a string.. it's binary data O.o
<MFen> to python it's a string
<cantares> file.read() just reads it in as a string.. i don't want that stuff in a string..
<MFen> either way, it sounds like a good place to have a database
<cantares> so i have converted the string into aprox. 4000000 1-byte ints..
<cantares> I don't want a database O.o
<MFen> what is the data?
<babbitt> a really big bitmap?
<cantares> compressed files.
<MFen> what kind of compression? python has a zip module, gzip, prolly bzip2
<MFen> tar
<cantares> that wouldn't help in my case
<MFen> well, you can either use a library that gives you an abstract layer to deal with this odd file format
<MFen> or you read it in as a string and parse it yourself
<MFen> up to you
<cantares> i don't want to extract it
<MFen> what do you want to do with it?
<cantares> i want to experiment a little bit with encryption
<cantares> and in actual state i have a tuple with 4.000.000 1-byte-ints and a realy big long int, which i want to combine with xor O.o
<CHodapp> cantares, consider using Numeric or Numarray
<CHodapp> There should be modules to read and write binary formats
<Jerub> cantares: to read binary data you can use struct.unpack.
<Jerub> oh, not sure that's the question
<ChrisLong> cantares: and you used struct.unpack? i'd have used data=array.array('B',filecontent).tolist() to get a list of unsigned chars (ints) but that is irrelevant for the moment.
<cantares> Jerub: I do that, unpack gave me the 4.000.000 1byte ints
<Jerub> oh, good.
<ChrisLong> cantares: instead of converting the int to a tuple, did you consider converting the tuple to an int?
<Jerub> cantares: so you can do something like reduce(operator.xor, mylist) now, right?
<cantares> i used it in,,, filedata = struct.unpack("@" + str(len(filedata)) + "B",filedata) that way
<cantares> ChrisLong: the only way i know to do that would use exzessiv processor time..
<ChrisLong> the result should be the same... except for the difference between a tuple and a list
<cantares> maybe a list is more usefull?
<ChrisLong> cantares: i'm not sure if it makes a big diff in cpu time to go from tupkle to int or from int to tuple
<Jerub> oh, a list and a tuple are interchangable in the reduce() thing
<babbitt> does base64.b64encode() have a maximum size, or is there a string maximum size? I'm trying to read in a file and am only getting a small chunk of it
<ChrisLong> jerub: i think cantares wants something like [a ^ b for a,b in zip(mytuple, myint_as_tuple)]
<ChrisLong> cantares: have look at http://deadbeefbabe.org/paste/535
<fserb> Any AST expert around?
<fserb> I want to know two things: Should I use the parser module instead of the compiler for everything? Second, on the ast.Function cl***, why "def x(a=1,b)" and "def x(a,b=1)" returns the same parameters (how do differentiate both?)
<cantares> i think it would be easier to write the stuff to a file again
<ChrisLong> cantares: maybe there are better ways, but those came to my mind
<cantares> ... , when both ware in tuple form
<ChrisLong> how does that help with the xor'ing?
<ChrisLong> or do you mean instead of putting the xor result in a list, write it to file?
<cantares> i have already written a experimental code part for tupel xoring that works
<cantares> no, i want to read a file, xor it with generated date and save the xor-data back into a file
<cantares> date -> data
<ChrisLong> i'll go now. it's nearly 2 am. i have to get some sleep
<Yhg1s> fserb: 'def x(a=1,b)' is invalid Python.
<Kevin_p> Anyone know of something to turn python programs into stand-alone executables?
<kosh> py2exe
<Kevin_p> kosh: Thank you :P
<_lester_> kosh: what about in *nix?
<kosh> you don't do it in unixes
<Jerub> _lester_: you build a .deb or a .rpm and have Depends: Python
<kosh> you just use the system python
<Jerub> oh osx, you use py2app
<Kevin_p> most linux distributions already come with Python anyways
<Jerub> (most linux distributions use python in the installer anyways)
<Jerub> ;)
<_lester_> yeah I am looking for running a python program on some embedded systems...
<_lester_> without proper filesystem support. :)
<Jerub> _lester_: huh?
<Jerub> _lester_: how can you have an embedded system without filesystem support?
<kosh> the point though is that for regular unixes you don't want it all bundled together since that makes life very difficult later for upgrades
<_lester_> Jerub: when u build your own...
<kosh> when I do a apt-get dist-upgrade I expect that everything gets updated so I don't have to manually track changes in stuff
<_lester_> kosh: yeah true.
<Jerub> _lester_: okay, how can you a) be competant enough to make a workable embedded system and b) be stupid enough not to implement a real filesystem.
<Kevin_p> I'm stupid enough not to do either, so HA.
<Jerub> (and (c) why the hell not use a bsd or linux base anyway)
<squiggly> has AES encryption been ported to python yet?
<_lester_> Jerub: because I have a day job and I'm not really competant. :)
<kosh> python2.4-pyopenssl
<Jerub> squiggly: google says yes.
<kosh> that should do AES
<bsmntbombdood> squiggly: I don't know of AES written in python, that would be really slow. But there are crypto librarys that include AES
<Jerub> _lester_: :)
<kosh> no idea how though but it should
<bsmntbombdood> squiggle: py-crypto is a good one
<ahmeni> the enigma cipher is in the python STL if you're up for some oldschool encryption
<squiggly> hum
<Yhg1s> no, rotor was removed in 2.4
<squiggly> can anyone recommend any good vcards that run on agp 4/8?
<squiggly> ;p
<cantares> how is the performance ratio between well c implementet encryption and well python implented?
<mykhal> I wish to p*** an instance's attribute value as the default value of an argument of the cl*** method.. something like this wrong way ... http://deadbeefbabe.org/paste/536 is there a simple way to do it ?
<bsmntbombdood> cantares: probably a lot
<Jerub> cantares: nearly nothing.
<Jerub> cantares: because a decent python implementation of a cypto algorithm is a python module written in C!
<cantares> i mean completly in python written encryption
Return to
#python
or
Go to some related
logs:
konverter rpm FC5
#php
unknown symbol pci_pretty_name
#mysql
nvclock temperature wrong
#perl
#lgp
#linux
#perl
sis 760gx ubuntu
|
|
