How to get file extension correctly?

Question

I know that this question is asked many times on this website. But I found that they missed an important point: only file extension with one period was taken into consider like *.png *.mp3, but how do I deal with these filename with two period like .tar.gz.

The basic code is:

filename = '/home/lancaster/Downloads/a.ppt'
extention = filename.split('/')[-1]

But obviously, this code do not work with the file like a.tar.gz. How to deal with it? Thanks.

By the way in your code, the variable `extension` is actually storing the complete filename, it does the same thing regardless of the type of extension. — mastazi, Jun 18 '16 at 11:31
your code is a wrong example: it's giving you the " basename " of the path, not the extension, which is equivalent to `import os;os.path.basename('/home/lancaster/Downloads/a.ppt')` — zmo, Jun 18 '16 at 11:32
a.tar.gz is a "gzip-compressed" tar file. So the extension of this file is `gz` and not `tar.gz`. So this question comes down to finding substrings ".tar.gz" etc, in the file names. If you see Rahul's edit, you will find that this is true. — gaganso, Jun 18 '16 at 11:38
Hey guys, I found an more interesting thing that if I compress `a.ppt` the default filename will be `a.ppt.tar.gz`, so there will be more disturbances. Please take this into consider. — Page David, Jun 18 '16 at 11:38
@SilentMonk But if I rename `a.tar.gz` to `a.tar(2).gz`, I cannot open it properly, so the extension is `tar.gz`. — Page David, Jun 18 '16 at 11:42
You can open `a.tar(2).gz` just fine: `gunzip` will work fine on that file. — , Jun 18 '16 at 11:59
@Evert But I got `unknown suffix -- ignored` from `gunzip a.ppt.tar\(2\)`. What goes wrong? — Page David, Jun 18 '16 at 12:16
Obviusly, you should *not* run `gunzip` on `a.ppt.tar\(2\)`, but on `a.ppt.tar\(2\).gz`... — , Jun 18 '16 at 17:42
Possible duplicate of [Extracting extension from filename in Python](http://stackoverflow.com/questions/541390/extracting-extension-from-filename-in-python) — Trevor Boyd Smith, Mar 15 '17 at 13:13
@Rahul 's answer is the better way. i.e using `os.path.splitext('/path/to/your/file')` If you need a one line code you can use something like `(os.path.splitext('path/to/file.ext')[1]).split('.')[1]` — Tharanga, Jun 22 '17 at 06:35

score 100 · Answer 1 · answered Jul 12 '17 at 14:28

100

Python 3.4

You can now use Path from pathlib. It has many features, one of them is suffix:

>>> from pathlib import Path
>>> Path('my/library/setup.py').suffix
'.py'
>>> Path('my/library.tar.gz').suffix
'.gz'
>>> Path('my/library').suffix
''

If you want to get more than one suffix, use suffixes:

>>> from pathlib import Path
>>> Path('my/library.tar.gar').suffixes
['.tar', '.gar']
>>> Path('my/library.tar.gz').suffixes
['.tar', '.gz']
>>> Path('my/library').suffixes
[]

answered Jul 12 '17 at 14:28

Or Duan

13,142
6
60
65

Something that frustrated me: I googled for `Path` from pathlib documentation, and it told me to use .ext . This kept failing and eventually I learned of .suffix. Not sure if the first documentation was for an older version of the Path() library? – Starman Jan 22 '18 at 20:59
@Starman `pathlib` introduced in Python 3.4, make sure you're not looking at Python 2 docs. – Or Duan Jan 22 '18 at 21:26

Rahul K P · Answer 2 · 2016-06-18T11:59:42.993

26

Here is a in build module in os. More about os.path.splitext.

In [1]: from os.path import splitext
In [2]: file_name,extension = splitext('/home/lancaster/Downloads/a.ppt')
In [3]: extension
Out[1]: '.ppt'

If you have to fine the extension of .tar.gz,.tar.bz2 you have to write a function like this

from os.path import splitext
def splitext_(path):
    for ext in ['.tar.gz', '.tar.bz2']:
        if path.endswith(ext):
            return path[:-len(ext)], path[-len(ext):]
    return splitext(path)

Result

In [4]: file_name,ext = splitext_('/home/lancaster/Downloads/a.tar.gz')
In [5]: ext
Out[2]: '.tar.gz'

Edit

Generally you can use this function

from os.path import splitext
def splitext_(path):
    if len(path.split('.')) > 2:
        return path.split('.')[0],'.'.join(path.split('.')[-2:])
    return splitext(path)

It will work for all extensions.

Working on all files.

In [6]: inputs = ['a.tar.gz', 'b.tar.lzma', 'a.tar.lz', 'a.tar.lzo', 'a.tar.xz','a.png']
In [7]: for file_ in inputs:                                                                    
    file_name,extension = splitext_(file_)
    print extension
   ....:     
tar.gz
tar.lzma
tar.lz
tar.lzo
tar.xz
.png

edited Jun 18 '16 at 11:59

answered Jun 18 '16 at 11:29

Rahul K P

15,740
4
35
52

1

I tried it ,this not work for `/home/lancaster/Downloads/a.tar.gz` – Page David Jun 18 '16 at 11:34
So I need to write `['tar.gz', 'tar.lzma', 'tar.lz', 'tar.lzo', 'tar.xz', etc...]` into my code? There must be many extension like this. – Page David Jun 18 '16 at 11:45
@DavidPage Updated my answer again. – Rahul K P Jun 18 '16 at 11:58
I think @no11 's solution is better because you used more `split` then him, and his code works well. – Page David Jun 18 '16 at 12:22
@DavidPage I tried to get the file name as well as the extension. It's a part of the coordination of in build function of `splitext`. That's why there is more occurrence of `split` – Rahul K P Jun 18 '16 at 12:26
I think change the fourth line to `return '.'.join(path.split('.')[:-2]),'.'.join(path.split('.')[-2:])` will be better for `a.ppt.tar.gz`. – Page David Jun 18 '16 at 12:40
I think the extension of `a.tar.gz` file is .gz :) not `.tar.gz` isn't it? so `os.path.splitext()` gives right result. :) – Tharanga Jun 22 '17 at 06:38
1

@Tharanga Solution according the OP's requirement. :) – Rahul K P Jun 22 '17 at 09:06

score 9 · Accepted Answer · answered Jun 18 '16 at 11:47

9

The role of a file extension is to tell the viewer (and sometimes the computer) which application to use to handle the file.

Taking your worst-case example in your comments (a.ppt.tar.gz), this is a PowerPoint file that has been tar-balled and then gzipped. So you need to use a gzip-handling program to open it. Using PowerPoint or a tarball-handling program wouldn't work. OK, a clever program that knew how to handle both .tar and .gz files could understand both operations and work with a .tar.gz file - but note that it would do that even if the extension was simply .gz.

The fact that both tar and gzip add their extensions to the original filename, rather than replace them (as zip does) is a convenience. But the base name of the gzip file is still a.ppt.tar.

answered Jun 18 '16 at 11:47

John Burger

3,662
1
13
23

@David, this is what I tried to convey in the comment. – gaganso Jun 18 '16 at 11:48
Thanks for your explain, based on this idea, I think @no11 's code is simple and good enough for me, but I still think you have got the best answer. – Page David Jun 18 '16 at 12:13
@Rahul 's answer is the better way. i.e using `os.path.splitext('/path/to/your/file')` If you need a one line code you can use something like `(os.path.splitext('path/to/file.ext')[1]).split('.')[1]` – Tharanga Jun 22 '17 at 06:34

score 3 · Answer 4 · answered Jun 18 '16 at 11:37

3

Simplest One:

import os.path
print os.path.splitext("/home/lancaster/Downloads/a.ppt")[1]
# '.ppt'

answered Jun 18 '16 at 11:37

Saket Mittal

3,726
3
29
49

score 2 · Answer 5 · edited Jun 18 '16 at 11:59

2

One possible way is:

Slice at "." => tmp_ext = filename.split('.')[1:]

Result is a list = ['tar', 'gz']

Join them together => extention = ".".join(tmp_ext)

Result is your extension as string = 'tar.gz'

Update: Example:

>>> test = "/test/test/test.tar.gz"
>>> t2 = test.split(".")[1:]
>>> t2
['tar', 'gz']
>>> ".".join(t2)
'tar.gz'

edited Jun 18 '16 at 11:59

Rahul K P

15,740
4
35
52

answered Jun 18 '16 at 11:31

no11

62
7

1

this will crash for filenames without a dot but except for that it will give what the OP asked for – Maximilian Peters Jun 18 '16 at 11:42
Ok. But for that case he already has a solution. So the core-question would be solved with my solution. But okay... – no11 Jun 18 '16 at 11:44
1

Although I found this solution pretty smart, it does not work for some cases. For example, some users put dots in the file name: 'my_document_on11.11.2019.docx'. – vmontazeri Apr 19 '19 at 04:06

score 0 · Answer 6 · answered Nov 17 '16 at 11:06

>>> import os
>>> import re

>>> filename = os.path.basename('/home/lancaster/Downloads/a.ppt')  
>>> extensions = re.findall(r'\.([^.]+)', basename)
['ppt']


>>> filename = os.path.basename('/home/lancaster/Downloads/a.ppt.tar.gz')  
>>> extensions = re.findall(r'\.([^.]+)', basename)
['ppt','tar','gz']

LetzerWille · Answer 7 · 2018-02-23T21:12:11.390

0

with re.findall and python 3.6

filename = '/home/Downloads/abc.ppt.tar.gz'

ext = r'\.\w{1,6}'

re.findall(f'{ext}\\b | {ext}$', filename,  re.X)

['.ppt', '.tar', '.gz']

edited Feb 23 '18 at 21:12

answered Feb 23 '18 at 20:47

LetzerWille

5,355
4
23
26

2

While this code may answer the question, providing additional context regarding why and/or how this code answers the question improves its long-term value. – rollstuhlfahrer Feb 24 '18 at 00:21

score -1 · Answer 8 · answered Jun 18 '16 at 11:33

-1

filename = '/home/lancaster/Downloads/a.tar.gz'
extention = filename.split('/')[-1]

if '.' in extention:
  extention = extention.split('.')[-1]
  if len(extention) > 0:
    extention = '.'+extention
    print extention

answered Jun 18 '16 at 11:33

akshay.s.jagtap

35
5

what's problem in this? – akshay.s.jagtap Jun 18 '16 at 11:36

How to get file extension correctly?

8 Answers8

Python 3.4