Projects_python gzipstream

== python-gzipstream - Streaming zlib (gzip) support for python ==

A streaming gzip handler.

[http://spacewalk.redhat.com/documentation/python-gzipstream/gzipstream.GzipStream-class.html gzipstream.GzipStream] extends the functionality of the [http://docs.python.org/release/3.1.3/library/gzip.html gzip.GzipFile] class to allow the processing of streaming data.

== Example ==

{{{ #!/usr/bin/python -u """ test on huge file which tends to expose "interesting" behavior """ import os import sys import time

print "NOTE: intended to be run from the gzipstream directory in CVS" print

from gzipstream import GzipStream

INPUT = os.path.join(os.path.dirname(sys.argv[0]), 'huge-all-zeros.txt.gz') BUFSIZE = 1024*16

t0 = time.time()

print "WARNING: This test takes a while to complete as it works on a very" print " file. It will:" print " - GunzipStream %s and count chars" % repr(INPUT) print " - compare with a zcat|wc" print

try: # python 2.* only class File(file): seek = None fin = File(INPUT, 'rb') print "Opened file as file object with no seek for a 'socket-like' test." except: print "Opened file as a normal file object." fin = open(INPUT, 'rb')

gzin = GzipStream(fin, 'r')

sys.stderr.write('reading...\n') x = gzin.read(BUFSIZE) count = len(x)

while x: x = gzin.read(BUFSIZE) count = count + len(x)

print "GzipStream's char count:\n%s" % count

print "'zcat | wc -m' char count:" os.system('zcat %s | wc -m' % INPUT) print "They should match."

print 'Elapsed:', (time.time() - t0)/60.0, 'mins' }}}

== Documentation ==

See http://spacewalk.redhat.com/documentation/python-gzipstream

== Get the code ==

Follow GitGuide and code is located in directory projects/python-gzipstream/ within checkout. If you run: {{{ tito build --tgz }}} in that directory you will get latest tar.gz file.

Or download [https://fedorahosted.org/releases/s/p/spacewalk/python-gzipstream-1.6.2.tar.gz python-gzipstream-1.6.2.tar.gz]