| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Comments:
<0> is there a python function that returns the time a file was last modified?
<0> or should I make my own ? :(
<1> would be better to check for yahoo then ms
<2> I have a rather complicated structure of lists and dictionaries, which I need to recursively process so that I have '' instead of all None and Empty values....
<3> agilman: os.path.getmtime
<4> msg HALLO
<1> TVSET: I still don't get what you need
<2> kosh: I found examples http://www.freenetpages.co.uk/hp/alan.gauld/tutrecur.htm to be pretty close to what I need. But can I use lists and dictionaries in the same way?
<3> EvanCarroll: ircing using telnet ?
<1> I know what recursion is but I don't understand what your problem is
<4> Jerub: Not this time (messaging channels via loop)
<2> kosh: my main problem is that I don't python... and want to stay that way for the time being... but I have a task at hand that needs solving. So I am looking for the easiest solution. :)
<2> don't know - that is :)
<5> kosh: I think he has a a list or dict with nested lists and dicts, and wants to convert all None to ''
<2> cr***: exactly
<5> TVSET: well its too bad that you don't want to learn python
<2> because I have to return the result via XMLRPC, and XMLRPC does not support the None stuff
<5> its really a simple laanguage
<2> cr***: I want to learn it, but not now. ;)
<1> why not change it at creation time?
<0> Jerub, is getmtime a python2.4 function?
<6> agilman: os.path.getmtime is ancient
<2> kosh: it is created by a thirdparty piece of code... (universal feed parser, if you know) .. and that piece of software is way too complicated for me to play with
<3> or there's just os.stat(filename).st_mtime
<5> TVSET: well I think you might have to check if its a dict or list, and then loop through all the items in the dict or list recusively doing the same if you come across a list or dict, other wise check if its None
<2> cr***: how can I check if it a list or a dict?
<1> if isinstnace(yourobj, list) if isinstance(yourob, dict)
<5> TVSET: type(obj) == list or type(obj) == dict
<1> cr***: NO NO NO!
<1> using type like that is a very bad idea
<1> even isinstance is a bad idea but vastly better
<5> kosh: why so?
<0> os.path.getmtime('/home/agilman/blogs/29-01-2006')
<0> does that look right?
<5> because it takes into account derivatives?>
<1> instance works with inheritance at least, what you are doing with type is enforcing a type
<5> right
<1> however I have seen a number of cases where people make their own datatypes that follow a list or dict interface
<1> and the type check like that will fail them even though if you did not check at all the code would work without issue
<7> help(time)
<5> yeah, I know, its planning for the future :)
<8> is there an easy way to see if a given string has any letters in it?
<9> try ***igning to the dict in a try:except, and catch the exception...
<9> if 'c' in string...
<8> ? doesn't that just check for 'c'?
<10> jajaja
<1> that is the best option sysfault however since he does not want to know python I doubt trying to work with exceptions will work for him
<10> spanis
<1> alus: define what a letter is exactly
<2> hmm
<8> kosh: string.ascii_letters
<2> alternatively I could use an XMLRPC server that doesn't choke on None or Empty... any suggestions?
<1> alus: so numbers, symbols etc don't count?
<7> gmtime(0)
<8> kosh: not today
<1> alus: what version of python are you using?
<8> 2.4
<1> a = set(string.letters) b = set(yourstring) a.intersection(b)
<7> 2.4.2
<11> for win32com what's VT_ARRAY | VT_VARIANT equal to?
<7> Python 2.4.2 (#1, Jan 8 2006, 01:04:02)
<8> kosh: neat
<7> in UNIX is the epoch 1970?
<12> hi, what's the iterator protocol?
<1> Yango: for i in something:
<13> Yango: http://www.python.org/doc/lib/typeiter.html
<1> Yango: it is how that works
<12> yes
<13> the protocol itself is the two methods described in that URL
<13> alus: s.isalpha()
<8> Yhg1s: contains, not is entirely
<13> alus: the set solution or a regexp solution then.
<14> Yhg1s, pysqlite2 dont have the cur.lastworid method ?
<13> the regexp solution is quite probably faster, but if set is fast enough for you, that's probably easier.
<14> for get the last row on a integer
<13> KarlsBerg: no idea, I don't use SQLite.
<14> oh ok, yesterday i think what you use pysqlite with you answer
<13> no, I just helped you with a problem.
<12> thanks Yhg1s
<14> Yhg1s, you perceive that I do not speak english perfect? jiji
<14> i must train :P
<13> yes, it's pefectly obvious that English is not your native tongue ;)
<15> Is it possible to use the timit module to accumulatively measure something that occurs in an inner loop? Would I do something like time_functionx += t.timeit.Timer("functionx(...)").timeit(1)?
<14> jijijijijijij
<1> Yhg1s: yeah the regex one is probaly faster to run but I figured that the set one was VASTLY easier to write and read again later so it was the best default solution
<13> Kuwanger: that would be quite inaccurate. why would you want that?
<15> Yhg1s: What do you mean exactly?
<13> Kuwanger: no, what do *you* mean exactly? :)
<13> Kuwanger: timing a single functioncall makes overhead have a *lot* more impact
<13> Kuwanger: so it will be quite inaccurate.
<13> Kuwanger: what are you trying to measure?
<15> Yhg1s: I'm trying to figure out the approximate ratio of time spent in each function in an inner-loop.
<12> Kuwanger you need to test them in a loop, the bigger, the most accurate
<13> Kuwanger: use the profiler instead.
<12> just one call will do you harm, no good
<15> Yhg1s: I didn't know of the profiler.
<12> Kuwanger: http://docs.python.org/lib/profile.html
<15> Yango: Yes, I know now.
<12> I'm trying to compare a dialect found by the Sniffer cl*** in the CSV module to existing Dialects... is there an easier way than to compare attribute by attribute?
<8> is there a list of python keywords in some module?
<13> you may be able to compare __dict__'s
<13> alus: do you mean reserved words? There's a list in the language reference, and one in the tokenize module. Why do you need it in a module, though?
<13> hm, no, not one in the tokenize module.
<8> well I could make the list myself, just wondering if it already exists
<8> yeah reserved words
<8> also standard type names and such would be nice
<13> well, its' not in a module
<13> but docs.python.org/lib has the builtin names, and docs.python.org/ref the list of reserved words
<12> alus: you could built it with a couple of python lines :-)
<8> I think I've got a pretty good set
<13> if you use those two URL's, you have the complete set ;P
<0> I wrote a small module, where would I place it so that everyone could access it?
<0> Im on linux
<13> /usr/lib/python<version>/site-packages/
<16> Woah.
<0> gratis
<2> I need a little help with this one - http://deadbeefbabe.org/paste/2756
<2> I managed to change all Falses to 0s, but Nones don't want to become ''s for some reason...
<8> print repr(L)
<2> alus: prints nothing... hmmm
<13> TVSET: what you think is None is apparently not really None.
<2> Yhg1s: what is it then? ;)
<13> something other.
<2> hmm
<13> like the string "None", perhaps.
<2> this is confusing
<2> can't be
<13> so print repr(L) in the 'else' clause.
<2> Yhg1s: prints nothing :(
<13> it doesn't print even a newline, or does it print ""?
<13> have you run your code in the interactive interpreter?
<13> you'll see that it treats None just like you want:
<13> >>> clean(None)
<13> ''
<13> >>> clean([None])
<13> ['']
<13> (etc)
<13> so, it isn't the code that's wrong, just the data.
<2> Yhg1s: it didn't print anything... not even a new line
<13> then the whole 'else' clause isn't executed, so one of the earlier conditions is satisfied instead.
<2> Yhg1s: it should get executed eventually...
<13> then something else is wrong. But have you tried what I suggested? Run it on some other data, instead of the feedparser output?
<2> if section on lists and dicts are executed, than it should somehow get to the point when list/dict values are processed in the else clause
<13> have you printed the repr() of data, to see if it actually contains None's?
<2> Yhg1s: yes, I printed the repr of data and it has None's... but all False's were converted to 0 as I wanted
<2> hence the puzzle ;)
<15> Yhg1s: Thanks.
<2> somehow repr(False) doesn't print anything...
<13> TVSET: and have you run your clean function on a single None, or a dict or list containing a None, like I have?
<13> >>> print repr(False)
<13> False
Return to
#python
or
Go to some related
logs:
error while loading shared libraries: libc.so.6 LD_ASSUME_KERNEL getconf GNU_LIB
#lisp
#javascript
#fedora
grub install hpt374
#lisp
gentoo alsasound' not found; aborting
la commande timport
ubuntu sound via 58 echo choppy
small ascii debian logo
|
|