Python gzip: is there a way to decompress from a string?

Question

I've read this SO post around the problem to no avail.

I am trying to decompress a .gz file coming from an URL.

url_file_handle=StringIO( gz_data )
gzip_file_handle=gzip.open(url_file_handle,"r")
decompressed_data = gzip_file_handle.read()
gzip_file_handle.close()

... but I get TypeError: coercing to Unicode: need string or buffer, cStringIO.StringI found

What's going on?

Traceback (most recent call last):  
  File "/opt/google/google_appengine-1.2.5/google/appengine/tools/dev_appserver.py", line 2974, in _HandleRequest
    base_env_dict=env_dict)
  File "/opt/google/google_appengine-1.2.5/google/appengine/tools/dev_appserver.py", line 411, in Dispatch
    base_env_dict=base_env_dict)
  File "/opt/google/google_appengine-1.2.5/google/appengine/tools/dev_appserver.py", line 2243, in Dispatch
    self._module_dict)
  File "/opt/google/google_appengine-1.2.5/google/appengine/tools/dev_appserver.py", line 2161, in ExecuteCGI
    reset_modules = exec_script(handler_path, cgi_path, hook)
  File "/opt/google/google_appengine-1.2.5/google/appengine/tools/dev_appserver.py", line 2057, in ExecuteOrImportScript
    exec module_code in script_module.__dict__
  File "/home/jldupont/workspace/jldupont/trunk/site/app/server/tasks/debian/repo_fetcher.py", line 36, in <module>
    main()
  File "/home/jldupont/workspace/jldupont/trunk/site/app/server/tasks/debian/repo_fetcher.py", line 30, in main
    gziph=gzip.open(fh,'r')
  File "/usr/lib/python2.5/gzip.py", line 49, in open
    return GzipFile(filename, mode, compresslevel)
  File "/usr/lib/python2.5/gzip.py", line 95, in __init__
    fileobj = self.myfileobj = __builtin__.open(filename, mode or 'rb')
TypeError: coercing to Unicode: need string or buffer, cStringIO.StringI found

please post more from the traceback (which line is failing). — tonfa, Oct 09 '09 at 13:12
Would using stringIO instead of cStringIO solve the problem? — recursive, Oct 09 '09 at 13:15
@recursive: I've already added the necessary checks for the availability of cStringIO and revert to StringIO if not. — jldupont, Oct 09 '09 at 13:18

score 56 · Answer 1 · edited Nov 02 '15 at 11:15

56

If your data is already in a string, try zlib, which claims to be fully gzip compatible:

import zlib
decompressed_data = zlib.decompress(gz_data, 16+zlib.MAX_WBITS)

Read more: http://docs.python.org/library/zlib.html‎

edited Nov 02 '15 at 11:15

Abyx

12,345
5
44
76

answered Aug 19 '13 at 17:21

Anon Mouse

668
5
4

8

This was also what came to me in the beginning. But it quickly turned out that it does not work:"zlib.error: Error -3 while decompressing data: incorrect header check" – HongboZhu Sep 19 '13 at 19:37
8

Add argument so the call looks like `decompressed_data=zlib.decompress(gz_data, 16+zlib.MAX_WBITS)`, works like a charm for me. Thanks to here http://stackoverflow.com/a/2695575/3635816 – RickardSjogren Jan 22 '15 at 19:50

score 38 · Accepted Answer · edited Jun 07 '13 at 11:47

gzip.open is a shorthand for opening a file, what you want is gzip.GzipFile which you can pass a fileobj

open(filename, mode='rb', compresslevel=9)
    #Shorthand for GzipFile(filename, mode, compresslevel).

vs

class GzipFile
   __init__(self, filename=None, mode=None, compresslevel=9, fileobj=None)
   #    At least one of fileobj and filename must be given a non-trivial value.

so this should work for you

gzip_file_handle = gzip.GzipFile(fileobj=url_file_handle)

score 7 · Answer 3 · answered Nov 03 '20 at 08:37

7

You can use gzip.decompress from the gzip builtin Python library(available for Python 3.2+).

Example on how to decompress bytes:

import gzip
gzip.decompress(gzip_data)

Documentation

https://docs.python.org/3.5/library/gzip.html#gzip.decompress

answered Nov 03 '20 at 08:37

Montoya

2,819
3
37
65

user1444978 · Answer 4 · 2017-01-20T14:45:57.660

Consider using gzip.GzipFile if you don't like passing obscure arguments to zlib.decompress.

When you deal with urllib2.urlopen response that can be either gzip-compressed or uncompressed:

import gzip
from StringIO import StringIO

# response = urllib2.urlopen(...

content_raw = response.read()
if 'gzip' in response.info().getheader('Content-Encoding'):
    content = gzip.GzipFile(fileobj=StringIO(content_raw)).read()

When you deal with a file that can store either gzip-compressed or uncompressed data:

import gzip

# some_file = open(...

try:
    content = gzip.GzipFile(fileobj=some_file).read()
except IOError:
    some_file.seek(0)
    content = some_file.read()

The examples above are in Python 2.7

using `try/except` with `read` on opened IO buffers is not the best way to infer format, because each attempt might consume some bytes. — Eli Korvigo, Dec 27 '17 at 13:01

Python gzip: is there a way to decompress from a string?

4 Answers4

Linked