2.9.9-2 - add ssl byterange and keepalive patch

epel9
jbowes 19 years ago
parent 49adcbb3b6
commit c354ed18ed

@ -3,9 +3,10 @@
Summary: A high-level cross-protocol url-grabber
Name: python-urlgrabber
Version: 2.9.9
Release: 1.1
Release: 2
Source0: urlgrabber-%{version}.tar.gz
Patch0: urlgrabber-read-error.patch
Patch1: urlgrabber-ssl-byterange-keepalive.patch
License: LGPL
Group: Development/Libraries
BuildRoot: %{_tmppath}/%{name}-%{version}-%{release}-buildroot
@ -22,6 +23,7 @@ authentication, proxies and more.
%prep
%setup -n urlgrabber-%{version}
%patch0 -p0
%patch1 -p0
%build
python setup.py build
@ -41,6 +43,9 @@ rm -rf $RPM_BUILD_ROOT
%{_bindir}/urlgrabber
%changelog
* Mon Jul 17 2006 James Bowes <jbowes@redhat.com> - 2.9.9-2
- Add support for byte ranges and keepalive over HTTPS
* Wed Jul 12 2006 Jesse Keating <jkeating@redhat.com> - 2.9.9-1.1
- rebuild

@ -0,0 +1,157 @@
Index: urlgrabber/byterange.py
===================================================================
RCS file: /cvsroot/urlgrabber/cvs-root/urlgrabber/urlgrabber/byterange.py,v
retrieving revision 1.11
diff -u -r1.11 byterange.py
--- urlgrabber/byterange.py 22 Oct 2005 21:57:28 -0000 1.11
+++ urlgrabber/byterange.py 13 Jul 2006 19:48:26 -0000
@@ -71,6 +71,15 @@
# HTTP's Range Not Satisfiable error
raise RangeError('Requested Range Not Satisfiable')
+class HTTPSRangeHandler(HTTPRangeHandler):
+ """ Range Header support for HTTPS. """
+
+ def https_error_206(self, req, fp, code, msg, hdrs):
+ return self.http_error_206(req, fp, code, msg, hdrs)
+
+ def https_error_416(self, req, fp, code, msg, hdrs):
+ self.https_error_416(req, fp, code, msg, hdrs)
+
class RangeableFileObject:
"""File object wrapper to enable raw range handling.
This was implemented primarilary for handling range
Index: urlgrabber/grabber.py
===================================================================
RCS file: /cvsroot/urlgrabber/cvs-root/urlgrabber/urlgrabber/grabber.py,v
retrieving revision 1.46
diff -u -r1.46 grabber.py
--- urlgrabber/grabber.py 22 Mar 2006 20:09:33 -0000 1.46
+++ urlgrabber/grabber.py 13 Jul 2006 19:48:26 -0000
@@ -402,24 +402,25 @@
# This is a convenient way to make keepalive optional.
# Just rename the module so it can't be imported.
import keepalive
- from keepalive import HTTPHandler
+ from keepalive import HTTPHandler, HTTPSHandler
except ImportError, msg:
- keepalive_handler = None
+ keepalive_handlers = ()
else:
- keepalive_handler = HTTPHandler()
+ keepalive_handlers = (HTTPHandler(), HTTPSHandler())
try:
# add in range support conditionally too
import byterange
- from byterange import HTTPRangeHandler, FileRangeHandler, \
- FTPRangeHandler, range_tuple_normalize, range_tuple_to_header, \
- RangeError
+ from byterange import HTTPRangeHandler, HTTPSRangeHandler, \
+ FileRangeHandler, FTPRangeHandler, range_tuple_normalize, \
+ range_tuple_to_header, RangeError
except ImportError, msg:
range_handlers = ()
RangeError = None
have_range = 0
else:
- range_handlers = (HTTPRangeHandler(), FileRangeHandler(), FTPRangeHandler())
+ range_handlers = (HTTPRangeHandler(), HTTPSRangeHandler(),
+ FileRangeHandler(), FTPRangeHandler())
have_range = 1
@@ -454,7 +455,7 @@
global DEBUG
DEBUG = DBOBJ
- if keepalive_handler and keepalive.DEBUG is None:
+ if keepalive_handlers and keepalive.DEBUG is None:
keepalive.DEBUG = DBOBJ
if have_range and byterange.DEBUG is None:
byterange.DEBUG = DBOBJ
@@ -582,7 +583,8 @@
def close_all():
"""close any open keepalive connections"""
- if keepalive_handler: keepalive_handler.close_all()
+ for handler in keepalive_handlers:
+ handler.close_all()
def urlgrab(url, filename=None, **kwargs):
"""grab the file at <url> and make a local copy at <filename>
@@ -1012,7 +1014,7 @@
return self.opts.opener
elif self._opener is None:
handlers = []
- need_keepalive_handler = (keepalive_handler and self.opts.keepalive)
+ need_keepalive_handler = (keepalive_handlers and self.opts.keepalive)
need_range_handler = (range_handlers and \
(self.opts.range or self.opts.reget))
# if you specify a ProxyHandler when creating the opener
@@ -1043,7 +1045,7 @@
# -------------------------------------------------------
if need_keepalive_handler:
- handlers.append( keepalive_handler )
+ handlers.extend( keepalive_handlers )
if need_range_handler:
handlers.extend( range_handlers )
handlers.append( auth_handler )
Index: urlgrabber/keepalive.py
===================================================================
RCS file: /cvsroot/urlgrabber/cvs-root/urlgrabber/urlgrabber/keepalive.py,v
retrieving revision 1.14
diff -u -r1.14 keepalive.py
--- urlgrabber/keepalive.py 4 Apr 2006 21:00:32 -0000 1.14
+++ urlgrabber/keepalive.py 13 Jul 2006 19:48:26 -0000
@@ -172,7 +172,7 @@
else:
return dict(self._hostmap)
-class HTTPHandler(urllib2.HTTPHandler):
+class KeepAliveHandler:
def __init__(self):
self._cm = ConnectionManager()
@@ -207,9 +207,6 @@
self._cm.remove(connection)
#### Transaction Execution
- def http_open(self, req):
- return self.do_open(HTTPConnection, req)
-
def do_open(self, http_class, req):
host = req.get_host()
if not host:
@@ -324,6 +321,20 @@
if req.has_data():
h.send(data)
+class HTTPHandler(KeepAliveHandler, urllib2.HTTPHandler):
+ def __init__(self):
+ KeepAliveHandler.__init__(self)
+
+ def http_open(self, req):
+ return self.do_open(HTTPConnection, req)
+
+class HTTPSHandler(KeepAliveHandler, urllib2.HTTPSHandler):
+ def __init__(self):
+ KeepAliveHandler.__init__(self)
+
+ def https_open(self, req):
+ return self.do_open(HTTPSConnection, req)
+
class HTTPResponse(httplib.HTTPResponse):
# we need to subclass HTTPResponse in order to
# 1) add readline() and readlines() methods
@@ -425,6 +436,9 @@
class HTTPConnection(httplib.HTTPConnection):
# use the modified response class
response_class = HTTPResponse
+
+class HTTPSConnection(httplib.HTTPSConnection):
+ response_class = HTTPResponse
#########################################################################
##### TEST FUNCTIONS
Loading…
Cancel
Save