Python glob and raw string

[Xaxa Urtiz]

Le jeudi 16 janvier 2014 19:14:30 UTC+1, Neil Cerutti a écrit :
> On 2014-01-16, Xaxa Urtiz <> wrote:
> 
> > Hello everybody, i've got a little problem, i've made a script
> 
> > which look after some files in some directory, typically my
> 
> > folder are organized like this :
> 
> >
> 
> > [share]
> 
> > folder1
> 
> > ->20131201
> 
> > -->file1.xml
> 
> > -->file2.txt
> 
> > ->20131202
> 
> > -->file9696009.tmp
> 
> > -->file421378932.xml
> 
> > etc....
> 
> > so basically in the share i've got some folder
> 
> > (=folder1,folder2.....) and inside these folder i've got these
> 
> > folder whose name is the date (20131201,20131202,20131203
> 
> > etc...) and inside them i want to find all the xml files.
> 
> > So, what i've done is to iterate over all the folder1/2/3 that
> 
> > i want and look, for each one, the xml file with that:
> 
> >
> 
> > for f in glob.glob(dir +r"\20140115\*.xml"):
> 
> > ->yield f
> 
> >
> 
> > dir is the folder1/2/3 everything is ok but i want to do
> 
> > something like that :
> 
> >
> 
> > for i in range(10,16):
> 
> > ->for f in glob.glob(dir +r"\201401{0}\*.xml".format(i)):
> 
> > -->yield f
> 
> >
> 
> > but the glob does not find any file.... (and of course there is
> 
> > some xml and the old way found them...) 
> 
> > Any help would be appreciate :) 
> 
> 
> 
> I've done this two different ways. The simple way is very similar
> 
> to what you are now doing. It sucks because I have to manually
> 
> maintain the list of subdirectories to traverse every time I
> 
> create a new subdir.
> 
> 
> 
> Here's the other way, using glob and isdir from os.path, adapted
> 
> from actual production code.
> 
> 
> 
> class Miner:
> 
>     def __init__(self, archive):
> 
>         # setup goes here; prepare to acquire the data
> 
>         self.descend(os.path.join(archive, '*'))
> 
> 
> 
>     def descend(self, path):
> 
>         for fname in glob.glob(os.path.join(path, '*')):
> 
>             if os.path.isdir(fname):
> 
>                 self.descend(fname)
> 
>             else:
> 
>                 self.process(fname)
> 
> 
> 
>     def process(self, path):
> 
>         # Do what I want done with an actual file path.
> 
> 	# This is where I add to the data.
> 
> 
> 
> In your case you might not want to process unless the path also
> 
> looks like an xml file.
> 
> 
> 
> mine = Miner('myxmldir')
> 
> 
> 
> Hmmm... I might be doing too much in __init__. ;)
> 
> 
> 
> -- 
> 
> Neil Cerutti

i only have 1 level of subdirectory, it's just in the case when i don't want to process all the date (otherwise i make a glob on '/*/*', no need to do any recursion.
thanks for the answer !