Extract values from within strings in a list - python

Question

I have a list in my python code with the following structure:

file_info = ['{file:C:\\samples\\123.exe, directory:C:\\}','{file:C:\\samples\\345.exe, directory:C:\\}',...]

I want to extract just the file and directory values for every value of the list and print it. With the following code, I am able to extract the directory values:

for item in file_info:

    print item.split('directory:')[1].strip('}')

But I am not able to figure out a way to extract the 'file' values. The following doesn't work:

print item.split('file:')[1].strip(', directory:C:\}')

Suggestions? If there is any better method to extract the file and directory values other than this, that would be great too. Thanks in advance.

What?! Does [{file:file1, directory... etc. have quotes around it? You write that you have a list of dicts and treat it like a string! — Two-Bit Alchemist
– Two-Bit Alchemist, Commented Apr 2, 2014 at 19:16
Is there any reason you're not using a list of dicts and using what seems to be a string instead? — matt
– matt, Commented Apr 2, 2014 at 19:19
Sorry, forgot the quotes.. it's not a list of dicts, it's just a list of strings. So it's file_info = ['{file:file1, directory:dir1}','{file:file2, directory:directory2}',...] — user2251144
– user2251144, Commented Apr 2, 2014 at 19:22
@user2251144 ok, list of strings. A simple example would help a lot. Thanks. — alecxe
– alecxe, Commented Apr 2, 2014 at 19:25

alecxe · Accepted Answer · 2014-04-02 19:44:46Z

3

If the format is exactly the same you've provided, you'd better go with using re:

import re

file_info = ['{file:file1, directory:dir1}', '{file:file2, directory:directory2}']

pattern = re.compile(r'\w+:(\w+)')
for item in file_info:
    print re.findall(pattern, item)

or, using string replace(), strip() and split() (a bit hackish and fragile):

file_info = ['{file:file1, directory:dir1}', '{file:file2, directory:directory2}']

for item in file_info:
    item = item.strip('}{').replace('file:', '').replace('directory:', '')
    print item.split(', ')

both code snippets print:

['file1', 'dir1']
['file2', 'directory2']

If the file_info items are just dumped json items (watch the double quotes), you can use json to load them into dictionaries:

import json

file_info = ['{"file":"file1", "directory":"dir1"}', '{"file":"file2", "directory":"directory2"}']

for item in file_info:
    item = json.loads(item)
    print item['file'], item['directory']

or, literal_eval():

from ast import literal_eval

file_info = ['{"file":"file1", "directory":"dir1"}', '{"file":"file2", "directory":"directory2"}']

for item in file_info:
    item = literal_eval(item)
    print item['file'], item['directory']

both code snippets print:

file1 dir1
file2 directory2

Hope that helps.

edited Apr 2, 2014 at 19:44

answered Apr 2, 2014 at 19:39

alecxe

476k127 gold badges1.1k silver badges1.2k bronze badges

Sign up to request clarification or add additional context in comments.

9 Comments

user2251144 Over a year ago

Thank you for the suggestions! I think the json.loads might be more relevant to my code, so I'll try to restructure file_info. Thanks again :)

eyquem Over a year ago

The solutions with json.loads and ast.literal_eval are incorrect since the treatment shown by OP in his/her question reveals that the strings are not as you modified them to make json.load and ast.literal_eval to work.

alecxe Over a year ago

@eyquem probably. But, as you can see, I've asked for the example twice :) That's why I've provided several options to choose from depends on the input format. Thanks.

eyquem Over a year ago

@user2251144 If you can restructure file_info , you should do it avoiding to repeat 'file' and 'directory' that doesn't bring any interesting information because this info is contained in the fact that the file is always in first position and the directory in second position in the string. In my opinion, you're not in a good process of solving a problem: you try to solve it at the level of the consequences while the problem IMO is at the definition of file_info level. I may be wrong. But that's the characteristics of XY problem to not give full information, origine and consequences.

eyquem Over a year ago

Yes , you asked two times. And (s)he didn't edit its question. I don't think it's a good practice to give quantities of soluces to balance the lack of a correct question. What if the file_info is composed of '{"file":file1, "directory":"dir1"}' ? You should provide a soluce for this case too.

|

eyquem · Accepted Answer · 2014-04-02 21:32:08Z

0

I would do:

import re

regx = re.compile('{\s*file\s*:\s*([^,\s]+)\s*'
                  ','
                  '\s*directory\s*:\s*([^}\s]+)\s*}')

file_info = ['{file:C:\\samples\\123.exe, directory  :  C:\\}',
             '{  file:  C:\\samples\\345.exe,directory:C:\\}'
             ]

for item in file_info:
    print '%r\n%s\n' % (item,
                        regx.search(item).groups())

result

'{file:C:\\samples\\123.exe, directory  :  C:\\}'
('C:\\samples\\123.exe', 'C:\\')

'{  file:  C:\\samples\\345.exe,directory:C:\\}'
('C:\\samples\\345.exe', 'C:\\')

answered Apr 2, 2014 at 21:32

eyquem

27.7k7 gold badges43 silver badges46 bronze badges

Collectives™ on Stack Overflow

Extract values from within strings in a list - python

2 Answers 2

9 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

9 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related