powerel7
/
base
2
0
Fork 0
Code Issues Pull Requests Releases Activity
Browse Source

python-urlgrabber package update 

Signed-off-by: basebuilder_pel7ppc64bebuilder0 <basebuilder@powerel.org>
master
basebuilder_pel7ppc64bebuilder0 6 years ago
parent
9265ad8f08
commit
84a17b227e
7 changed files with 796 additions and 0 deletions
Whitespace
Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL
Split View
Diff Options
Show Stats Download Patch File Download Diff File
  1. 60
      SOURCES/BZ-1017491-respond-to-ctrl-c.patch
  2. 18
      SOURCES/BZ-1082648-curl-77-error-message.patch
  3. 19
      SOURCES/BZ-1099101-revert-curl-ctrl-c.patch
  4. 24
      SOURCES/BZ-1233329-timedhosts-parsing-error-handling.patch
  5. 279
      SOURCES/BZ-1342179-add-retry-no-cache-opt.patch
  6. 31
      SOURCES/BZ-853432-single-conn-reset.patch
  7. 365
      SPECS/python-urlgrabber.spec

60
SOURCES/BZ-1017491-respond-to-ctrl-c.patch
Unescape View File

@ -0,0 +1,60 @@ @@ -0,0 +1,60 @@
commit 42b5945c99d92d55b5d401f2d934fe8fcc19f98b
Author: Zdenek Pavlas <zpavlas@redhat.com>
Date: Mon Nov 4 15:32:43 2013 +0100

Tell curl to return immediately on ctrl-c. BZ 1017491

diff --git a/urlgrabber/grabber.py b/urlgrabber/grabber.py
index 03fa1ba..579c2ff 100644
--- a/urlgrabber/grabber.py
+++ b/urlgrabber/grabber.py
@@ -1903,6 +1903,8 @@ class PyCurlFileObject(object):
urllib.addinfourl, via. urllib.URLopener.* """
return self.url
+# tell curl to return immediately on ctrl-c
+pycurl.global_init(pycurl.GLOBAL_DEFAULT | pycurl.GLOBAL_ACK_EINTR)
_curl_cache = pycurl.Curl() # make one and reuse it over and over and over
def reset_curl_obj():
commit 857af5ac1cec0a06755b75286ae2192e398cd341
Author: Zdenek Pavlas <zpavlas@redhat.com>
Date: Wed Dec 18 13:12:00 2013 +0100

Don't fail with older pycurl.

diff --git a/urlgrabber/grabber.py b/urlgrabber/grabber.py
index 579c2ff..ef18d6a 100644
--- a/urlgrabber/grabber.py
+++ b/urlgrabber/grabber.py
@@ -1903,8 +1903,9 @@ class PyCurlFileObject(object):
urllib.addinfourl, via. urllib.URLopener.* """
return self.url
-# tell curl to return immediately on ctrl-c
-pycurl.global_init(pycurl.GLOBAL_DEFAULT | pycurl.GLOBAL_ACK_EINTR)
+if hasattr(pycurl, 'GLOBAL_ACK_EINTR'):
+ # fail immediately on ctrl-c
+ pycurl.global_init(pycurl.GLOBAL_DEFAULT | pycurl.GLOBAL_ACK_EINTR)
_curl_cache = pycurl.Curl() # make one and reuse it over and over and over
def reset_curl_obj():
commit bf0a0be71373dec515bbb54e0613a3b9b0c00b04
Author: Valentina Mukhamedzhanova <vmukhame@redhat.com>
Date: Tue Feb 4 10:11:58 2014 +0100

Treat curl errcodes 42, 55, 66 as Ctrl-C BZ#1017491

diff --git a/urlgrabber/grabber.py b/urlgrabber/grabber.py
index 04f1179..f8deeb8 100644
--- a/urlgrabber/grabber.py
+++ b/urlgrabber/grabber.py
@@ -1517,7 +1517,7 @@ class PyCurlFileObject(object):
err.url = errurl
raise err
- elif errcode == 42:
+ elif errcode in (42, 55, 56):
# this is probably wrong but ultimately this is what happens
# we have a legit http code and a pycurl 'writer failed' code
# which almost always means something aborted it from outside

18
SOURCES/BZ-1082648-curl-77-error-message.patch
Unescape View File

@ -0,0 +1,18 @@ @@ -0,0 +1,18 @@
commit e19463a8781fa5de74e62a09684259309b2f89cf
Author: Valentina Mukhamedzhanova <vmukhame@redhat.com>
Date: Mon Sep 22 14:45:10 2014 +0200

Add a comprehensive error message to pycurl error 77. BZ 1082648

diff --git a/urlgrabber/grabber.py b/urlgrabber/grabber.py
index fd31123..961d18a 100644
--- a/urlgrabber/grabber.py
+++ b/urlgrabber/grabber.py
@@ -1576,6 +1576,7 @@ class PyCurlFileObject(object):
67 : _("Authentication failure"),
70 : _("Out of disk space on server"),
73 : _("Remove file exists"),
+ 77 : _("Problem with the SSL CA cert (path? access rights?)"),
}
errstr = str(e.args[1]) or pyerr2str.get(errcode, '<Unknown>')
if code and not 200 <= code <= 299:

19
SOURCES/BZ-1099101-revert-curl-ctrl-c.patch
Unescape View File

@ -0,0 +1,19 @@ @@ -0,0 +1,19 @@
commit 545eb9bbc325a2f59803dfb199ef64dec6db8774
Author: Valentina Mukhamedzhanova <vmukhame@redhat.com>
Date: Wed Sep 10 10:04:10 2014 +0200

Revert bf0a0be71373dec515bbb54e0613a3b9b0c00b04 - curl error codes.

diff --git a/urlgrabber/grabber.py b/urlgrabber/grabber.py
index 69cd113..fd31123 100644
--- a/urlgrabber/grabber.py
+++ b/urlgrabber/grabber.py
@@ -1524,7 +1524,7 @@ class PyCurlFileObject(object):
err.url = errurl
raise err
- elif errcode in (42, 55, 56):
+ elif errcode == 42:
# this is probably wrong but ultimately this is what happens
# we have a legit http code and a pycurl 'writer failed' code
# which almost always means something aborted it from outside

24
SOURCES/BZ-1233329-timedhosts-parsing-error-handling.patch
Unescape View File

@ -0,0 +1,24 @@ @@ -0,0 +1,24 @@
commit 7644d090e186c44d577f8d06de57bd815e0dc8b0
Author: Valentina Mukhamedzhanova <vmukhame@redhat.com>
Date: Tue Jun 23 10:00:12 2015 +0200

Don't crash on timedhosts parsing error. BZ#1233329

diff --git a/urlgrabber/grabber.py b/urlgrabber/grabber.py
index 91cdbdd..f09d4a5 100644
--- a/urlgrabber/grabber.py
+++ b/urlgrabber/grabber.py
@@ -2377,8 +2377,11 @@ class _TH:
try:
now = int(time.time())
for line in open(filename):
- host, speed, fail, ts = line.rsplit(' ', 3)
- _TH.hosts[host] = int(speed), int(fail), min(int(ts), now)
+ try:
+ host, speed, fail, ts = line.rsplit(' ', 3)
+ _TH.hosts[host] = int(speed), int(fail), min(int(ts), now)
+ except ValueError:
+ if DEBUG: DEBUG.info('Error parsing timedhosts: line "%s"', line)
except IOError: pass
_TH.dirty = False

279
SOURCES/BZ-1342179-add-retry-no-cache-opt.patch
Unescape View File

@ -0,0 +1,279 @@ @@ -0,0 +1,279 @@
diff -up urlgrabber-3.10/test/test_mirror.py.orig urlgrabber-3.10/test/test_mirror.py
--- urlgrabber-3.10/test/test_mirror.py.orig 2013-08-26 09:09:07.000000000 +0200
+++ urlgrabber-3.10/test/test_mirror.py 2016-06-29 18:26:06.790393129 +0200
@@ -268,33 +268,55 @@ class ActionTests(TestCase):
self.assertEquals(self.g.calls, expected_calls)
self.assertEquals(urlgrabber.mirror.DEBUG.logs, expected_logs)
+import thread, socket
+LOCALPORT = 'localhost', 2000
class HttpReplyCode(TestCase):
def setUp(self):
+ # start the server
+ self.exit = False
def server():
- import socket
s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
s.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
- s.bind(('localhost', 2000)); s.listen(1)
+ s.bind(LOCALPORT); s.listen(1)
while 1:
c, a = s.accept()
+ if self.exit: c.close(); break
while not c.recv(4096).endswith('\r\n\r\n'): pass
c.sendall('HTTP/1.1 %d %s\r\n' % self.reply)
+ if self.content is not None:
+ c.sendall('Content-Length: %d\r\n\r\n' % len(self.content))
+ c.sendall(self.content)
c.close()
- import thread
- self.reply = 503, "Busy"
+ s.close()
+ self.exit = False
thread.start_new_thread(server, ())
+ # create grabber and mirror group objects
def failure(obj):
self.code = getattr(obj.exception, 'code', None)
return {}
self.g = URLGrabber()
- self.mg = MirrorGroup(self.g, ['http://localhost:2000/'], failure_callback = failure)
+ self.mg = MirrorGroup(self.g, ['http://%s:%d' % LOCALPORT],
+ failure_callback = failure)
+
+ def tearDown(self):
+ # shut down the server
+ self.exit = True
+ s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
+ s.connect(LOCALPORT); s.close() # wake it up
+ while self.exit: pass # poor man's join
def test_grab(self):
+ 'tests the propagation of HTTP reply code'
+ self.reply = 503, "Busy"
+ self.content = None
+
+ # single
self.assertRaises(URLGrabError, self.mg.urlgrab, 'foo')
self.assertEquals(self.code, 503); del self.code
+ # multi
err = []
self.mg.urlgrab('foo', async = True, failfunc = err.append)
urlgrabber.grabber.parallel_wait()
diff -up urlgrabber-3.10/test/test_mirror.py.orig urlgrabber-3.10/test/test_mirror.py
--- urlgrabber-3.10/test/test_mirror.py.orig 2016-06-29 18:26:06.790393129 +0200
+++ urlgrabber-3.10/test/test_mirror.py 2016-06-29 18:26:58.886148544 +0200
@@ -268,13 +268,14 @@ class ActionTests(TestCase):
self.assertEquals(self.g.calls, expected_calls)
self.assertEquals(urlgrabber.mirror.DEBUG.logs, expected_logs)
-import thread, socket
+import threading, socket
LOCALPORT = 'localhost', 2000
class HttpReplyCode(TestCase):
def setUp(self):
# start the server
self.exit = False
+ self.process = lambda data: None
def server():
s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
s.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
@@ -282,7 +283,10 @@ class HttpReplyCode(TestCase):
while 1:
c, a = s.accept()
if self.exit: c.close(); break
- while not c.recv(4096).endswith('\r\n\r\n'): pass
+ data = ''
+ while not data.endswith('\r\n\r\n'):
+ data = c.recv(4096)
+ self.process(data)
c.sendall('HTTP/1.1 %d %s\r\n' % self.reply)
if self.content is not None:
c.sendall('Content-Length: %d\r\n\r\n' % len(self.content))
@@ -290,7 +294,8 @@ class HttpReplyCode(TestCase):
c.close()
s.close()
self.exit = False
- thread.start_new_thread(server, ())
+ self.thread = threading.Thread(target=server)
+ self.thread.start()
# create grabber and mirror group objects
def failure(obj):
@@ -305,7 +310,7 @@ class HttpReplyCode(TestCase):
self.exit = True
s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
s.connect(LOCALPORT); s.close() # wake it up
- while self.exit: pass # poor man's join
+ self.thread.join()
def test_grab(self):
'tests the propagation of HTTP reply code'
@@ -323,6 +328,45 @@ class HttpReplyCode(TestCase):
self.assertEquals([e.exception.errno for e in err], [256])
self.assertEquals(self.code, 503); del self.code
+ def test_retry_no_cache(self):
+ 'test bypassing proxy cache on failure'
+ def process(data):
+ if 'Pragma:no-cache' in data:
+ self.content = 'version2'
+ else:
+ self.content = 'version1'
+
+ def checkfunc_read(obj):
+ if obj.data == 'version1':
+ raise URLGrabError(-1, 'Outdated version of foo')
+
+ def checkfunc_grab(obj):
+ with open('foo') as f:
+ if f.read() == 'version1':
+ raise URLGrabError(-1, 'Outdated version of foo')
+
+ self.process = process
+ self.reply = 200, "OK"
+
+ opts = self.g.opts
+ opts.retry = 3
+ opts.retry_no_cache = True
+
+ # single
+ opts.checkfunc = checkfunc_read
+ try:
+ self.mg.urlread('foo')
+ except URLGrabError as e:
+ self.fail(str(e))
+
+ # multi
+ opts.checkfunc = checkfunc_grab
+ self.mg.urlgrab('foo', async=True)
+ try:
+ urlgrabber.grabber.parallel_wait()
+ except URLGrabError as e:
+ self.fail(str(e))
+
def suite():
tl = TestLoader()
return tl.loadTestsFromModule(sys.modules[__name__])
diff -up urlgrabber-3.10/urlgrabber/grabber.py.orig urlgrabber-3.10/urlgrabber/grabber.py
--- urlgrabber-3.10/urlgrabber/grabber.py.orig 2016-06-29 18:25:53.964453346 +0200
+++ urlgrabber-3.10/urlgrabber/grabber.py 2016-06-29 18:26:58.886148544 +0200
@@ -171,6 +171,12 @@ GENERAL ARGUMENTS (kwargs)
The libproxy code is only used if the proxies dictionary
does not provide any proxies.
+ no_cache = False
+
+ When True, server-side cache will be disabled for http and https
+ requests. This is equivalent to setting
+ http_headers = (('Pragma', 'no-cache'),)
+
prefix = None
a url prefix that will be prepended to all requested urls. For
@@ -383,10 +389,11 @@ RETRY RELATED ARGUMENTS
identical to checkfunc, except for the attributes defined in the
CallbackObject instance. The attributes for failure_callback are:
- exception = the raised exception
- url = the url we're trying to fetch
- tries = the number of tries so far (including this one)
- retry = the value of the retry option
+ exception = the raised exception
+ url = the url we're trying to fetch
+ tries = the number of tries so far (including this one)
+ retry = the value of the retry option
+ retry_no_cache = the value of the retry_no_cache option
The callback is present primarily to inform the calling program of
the failure, but if it raises an exception (including the one it's
@@ -431,6 +438,19 @@ RETRY RELATED ARGUMENTS
passed the same arguments, so you could use the same function for
both.
+ retry_no_cache = False
+
+ When True, automatically enable no_cache for future retries if
+ checkfunc performs an unsuccessful check.
+
+ This option is useful if your application expects a set of files
+ from the same server to form an atomic unit and you write your
+ checkfunc to ensure each file being downloaded belongs to such a
+ unit. If transparent proxy caching is in effect, the files can
+ become out-of-sync, disrupting the atomicity. Enabling this option
+ will prevent that, while ensuring that you still enjoy the benefits
+ of caching when possible.
+
BANDWIDTH THROTTLING
urlgrabber supports throttling via two values: throttle and
@@ -1001,6 +1021,8 @@ class URLGrabberOptions:
self.half_life = 30*24*60*60 # 30 days
self.default_speed = 1e6 # 1 MBit
self.ftp_disable_epsv = False
+ self.no_cache = False
+ self.retry_no_cache = False
def __repr__(self):
return self.format()
@@ -1077,7 +1099,8 @@ class URLGrabber(object):
if callback:
if DEBUG: DEBUG.info('calling callback: %s', callback)
obj = CallbackObject(exception=exception, url=args[0],
- tries=tries, retry=opts.retry)
+ tries=tries, retry=opts.retry,
+ retry_no_cache=opts.retry_no_cache)
_run_callback(callback, obj)
if (opts.retry is None) or (tries == opts.retry):
@@ -1089,6 +1112,8 @@ class URLGrabber(object):
if DEBUG: DEBUG.info('retrycode (%i) not in list %s, re-raising',
retrycode, opts.retrycodes)
raise
+ if retrycode is not None and retrycode < 0 and opts.retry_no_cache:
+ opts.no_cache = True
def urlopen(self, url, opts=None, **kwargs):
"""open the url and return a file object
@@ -1429,11 +1454,15 @@ class PyCurlFileObject(object):
self.curl_obj.setopt(pycurl.SSLKEYPASSWD, opts.ssl_key_pass)
#headers:
- if opts.http_headers and self.scheme in ('http', 'https'):
+ if self.scheme in ('http', 'https'):
headers = []
- for (tag, content) in opts.http_headers:
- headers.append('%s:%s' % (tag, content))
- self.curl_obj.setopt(pycurl.HTTPHEADER, headers)
+ if opts.http_headers is not None:
+ for (tag, content) in opts.http_headers:
+ headers.append('%s:%s' % (tag, content))
+ if opts.no_cache:
+ headers.append('Pragma:no-cache')
+ if headers:
+ self.curl_obj.setopt(pycurl.HTTPHEADER, headers)
# ranges:
if opts.range or opts.reget:
@@ -2055,7 +2084,8 @@ class _ExternalDownloader:
'ssl_key_pass',
'ssl_verify_peer', 'ssl_verify_host',
'size', 'max_header_size', 'ip_resolve',
- 'ftp_disable_epsv'
+ 'ftp_disable_epsv',
+ 'no_cache',
)
def start(self, opts):
@@ -2236,6 +2266,8 @@ def parallel_wait(meter=None):
except URLGrabError, ug_err:
retry = 0 # no retries
if opts.tries < retry and ug_err.errno in opts.retrycodes:
+ if ug_err.errno < 0 and opts.retry_no_cache:
+ opts.no_cache = True
start(opts, opts.tries + 1) # simple retry
continue

31
SOURCES/BZ-853432-single-conn-reset.patch
Unescape View File

@ -0,0 +1,31 @@ @@ -0,0 +1,31 @@
From 93ad446fc6533304e33b231cd5786572773440b9 Mon Sep 17 00:00:00 2001
From: Zdenek Pavlas <zpavlas@redhat.com>
Date: Mon, 21 Oct 2013 16:52:53 +0200
Subject: [PATCH] single-conn mode: Kill the cached idle connection. BZ 853432

The multi-downloader manages connections as necessary, caching
at most one downloader per host. But there's one more connection,
used by the legacy downloader. In case keepalive=1 (default),
it's still connected, possibly to the same connection-limited
host. Kill it.
---
urlgrabber/grabber.py | 3 +++
1 file changed, 3 insertions(+)

diff --git a/urlgrabber/grabber.py b/urlgrabber/grabber.py
index b004f4d..d06cdae 100644
--- a/urlgrabber/grabber.py
+++ b/urlgrabber/grabber.py
@@ -2221,6 +2221,9 @@ def parallel_wait(meter=None):
# Turn on the max_connections=1 override. BZ 853432
if DEBUG: DEBUG.info('max_connections(%s) %s => 1', key, limit)
single.add(key)
+ # When using multi-downloader the parent's _curl_cache
+ # object is idle. Kill it, as it might use keepalive=1.
+ reset_curl_obj()
retry = opts.retry or 0
if opts.failure_callback:
--
1.7.11.7

365
SPECS/python-urlgrabber.spec
Unescape View File

@ -0,0 +1,365 @@ @@ -0,0 +1,365 @@
%{!?python_sitelib: %define python_sitelib %(python -c "from distutils.sysconfig import get_python_lib; print get_python_lib()")}

Summary: A high-level cross-protocol url-grabber
Name: python-urlgrabber
Version: 3.10
Release: 8%{?dist}
Source0: http://urlgrabber.baseurl.org/download/urlgrabber-%{version}.tar.gz
Patch1: BZ-853432-single-conn-reset.patch
Patch2: BZ-1017491-respond-to-ctrl-c.patch

# rhel-7.1.0
Patch10: BZ-1099101-revert-curl-ctrl-c.patch
Patch11: BZ-1082648-curl-77-error-message.patch

# rhel-7.2
Patch20: BZ-1233329-timedhosts-parsing-error-handling.patch

# rhel-7.3
Patch25: BZ-1342179-add-retry-no-cache-opt.patch

License: LGPLv2+
Group: Development/Libraries
BuildRoot: %{_tmppath}/%{name}-%{version}-%{release}-root
BuildArch: noarch
BuildRequires: python-devel, python-pycurl
Url: http://urlgrabber.baseurl.org/
Provides: urlgrabber = %{version}-%{release}
Requires: python-pycurl

%description
A high-level cross-protocol url-grabber for python supporting HTTP, FTP
and file locations. Features include keepalive, byte ranges, throttling,
authentication, proxies and more.

%prep
%setup -q -n urlgrabber-%{version}
%patch1 -p1
%patch2 -p1

# rhel-7.1.0
%patch10 -p1
%patch11 -p1

# rhel-7.2
%patch20 -p1

# rhel-7.3
%patch25 -p1

%build
python setup.py build

%install
rm -rf $RPM_BUILD_ROOT
python setup.py install -O1 --root=$RPM_BUILD_ROOT
rm -rf $RPM_BUILD_ROOT/%{_docdir}/urlgrabber-%{version}

%clean
rm -rf $RPM_BUILD_ROOT

%files
%defattr(-,root,root,-)
%doc ChangeLog LICENSE README TODO
%{python_sitelib}/urlgrabber*
%{_bindir}/urlgrabber
%attr(0755,root,root) %{_libexecdir}/urlgrabber-ext-down

%changelog
* Thu Jun 30 2016 Valentina Mukhamedzhanova <vmukhame@redhat.com> - 3.10-8
- Add no_cache and retry_no_cache options.
- Resolves: bug#1342179

* Tue Jun 30 2015 Valentina Mukhamedzhanova <vmukhame@redhat.com> - 3.10-7
- Don't crash on timedhosts parsing error.
- Resolves: bug#1233329

* Wed Sep 24 2014 Valentina Mukhamedzhanova <vmukhame@redhat.com> - 3.10-6
- Add a comprehensive error message to curl error 77.
- Resolves: bug#1082648

* Wed Sep 10 2014 Valentina Mukhamedzhanova <vmukhame@redhat.com> - 3.10-5
- Revert treating curl errcodes 55 and 56 as Ctrl-C.
- Resolves: bug#1099101

* Tue Feb 4 2014 Valentina Mukhamedzhanova <vmukhame@redhat.com> - 3.10-4
- Treat curl errcodes 55 and 56 as Ctrl-C.
- Resolves: bug#1017491

* Thu Jan 16 2014 Valentina Mukhamedzhanova <vmukhame@redhat.com> - 3.10-3
- Tell curl to return immediately on ctrl-c.
- Resolves: bug#1017491

* Fri Dec 27 2013 Daniel Mach <dmach@redhat.com> - 3.10-2
- Mass rebuild 2013-12-27

* Mon Oct 21 2013 Zdenek Pavlas <zpavlas@redhat.com> - 3.10-1
- single-conn mode: Kill the cached idle connection. BZ 853432

* Thu Oct 17 2013 Zdenek Pavlas <zpavlas@redhat.com> - 3.10-0
- clamp timestamps from the future. BZ 894630, 1013733
- Fix the condition to enter single-connection mode. BZ 853432
- Fix unit tests

* Tue Oct 8 2013 Zdenek Pavlas <zpavlas@redhat.com> - 3.9.1-33
- Fix the condition to enter single-connection mode (again)

* Thu Oct 3 2013 Zdenek Pavlas <zpavlas@redhat.com> - 3.9.1-32
- Update to latest HEAD.
- Fix traceback when entering single-connection mode with DEBUG off
- Fix traceback when limit==None and DEBUG is on
- Fix the condition to enter single-connection mode

* Thu Oct 3 2013 Zdenek Pavlas <zpavlas@redhat.com> - 3.9.1-31
- Update to latest HEAD.
- Switch to max_connections=1 after refused connect. BZ 853432
- Never display negative downloading speed. BZ 1001767
- clamp timestamps from the future. BZ 894630, 1013733

* Thu Aug 29 2013 Zdenek Pavlas <zpavlas@redhat.com> - 3.9.1-30
- Update to latest HEAD.
- add ftp_disable_epsv option. BZ 849177
- Spelling fixes.
- docs: throttling is per-connection, suggest max_connections=1. BZ 998263
- More robust "Content-Length" parsing. BZ 1000841

* Tue Jun 18 2013 Zdenek Pavlas <zpavlas@redhat.com> - 3.9.1-29
- Update to latest HEAD.
- Fix parsing of FTP 213 responses
- Switch to max_connections=1 after timing out. BZ 853432
- max_connections=0 should imply the default limit.

* Fri May 17 2013 Zdenek Pavlas <zpavlas@redhat.com> - 3.9.1-28
- Update to latest HEAD.
- Add the "minrate" option. BZ 964298
- Workaround progress "!!!" end for file:// repos.
- add URLGrabError.code to the external downloader API
- Disable GSSNEGOTIATE to work around a curl bug. BZ 960163

* Wed Mar 27 2013 Zdenek Pavlas <zpavlas@redhat.com> - 3.9.1-26
- Update to latest HEAD.
- Handle HTTP 200 response to range requests correctly. BZ 919076
- Reset curl_obj to clear CURLOPT_RANGE from previous requests. BZ 923951

* Thu Mar 7 2013 Zdeněk Pavlas <zpavlas@redhat.com> - 3.9.1-25
- Update to latest HEAD.
- fix some test cases that were failing. BZ 918658
- exit(1) or /bin/urlgrabber failures. BZ 918613
- clamp timestamps from the future. BZ 894630
- enable GSSNEGOTIATE if implemented correctly.
- make error messages more verbose.

* Thu Feb 14 2013 Fedora Release Engineering <rel-eng@lists.fedoraproject.org> - 3.9.1-24
- Rebuilt for https://fedoraproject.org/wiki/Fedora_19_Mass_Rebuild

* Mon Jan 7 2013 Zdeněk Pavlas <zpavlas@redhat.com> - 3.9.1-23
- Update to latest HEAD.
- Handle checkfunc unicode exceptions. BZ 672117

* Thu Dec 6 2012 Zdeněk Pavlas <zpavlas@redhat.com> - 3.9.1-22
- Update to latest HEAD.
- Improve URLGRABBER_DEBUG, add max_connections. BZ 853432

* Thu Nov 1 2012 Zdeněk Pavlas <zpavlas@redhat.com> - 3.9.1-21
- Update to latest HEAD.
- Get rid of "HTTP 200 OK" errors. BZ 871835.

* Tue Sep 4 2012 Zdeněk Pavlas <zpavlas@redhat.com> - 3.9.1-20
- Update to latest HEAD.
- Fixed BZ 851178, 854075.

* Mon Aug 27 2012 Zdeněk Pavlas <zpavlas@redhat.com> - 3.9.1-19
- timedhosts: defer 1st update until a 1MB+ download. BZ 851178

* Wed Aug 22 2012 Zdeněk Pavlas <zpavlas@redhat.com> - 3.9.1-18
- Update to latest HEAD, lots of enhancements.

* Wed Aug 10 2012 Zdeněk Pavlas <zpavlas@redhat.com> - 3.9.1-17
- Fix a bug in progress display code. BZ 847105.

* Wed Aug 8 2012 Zdeněk Pavlas <zpavlas@redhat.com> - 3.9.1-16
- Update to latest head.
- Improved multi-file progress, small bugfixes.

* Fri Jul 20 2012 Zdeněk Pavlas <zpavlas@redhat.com> - 3.9.1-15
- Update to latest head, misc bugfixes: BZ 832028, 831904, 831291.
- Disable Kerberos auth. BZ 769254
- copy_local bugfix. BZ 837018
- send 'tries' counter to mirror failure callback

* Mon May 21 2012 Zdeněk Pavlas <zpavlas@redhat.com> - 3.9.1-14
- timedhosts: sanity check on dl_time

* Fri May 18 2012 Zdeněk Pavlas <zpavlas@redhat.com> - 3.9.1-13
- fix file:// profiling. BZ 822632.

* Mon May 14 2012 Zdeněk Pavlas <zpavlas@redhat.com> - 3.9.1-12
- Update to latest HEAD
- Merge multi-downloader patches

* Sat Jan 14 2012 Fedora Release Engineering <rel-eng@lists.fedoraproject.org> - 3.9.1-11
- Rebuilt for https://fedoraproject.org/wiki/Fedora_17_Mass_Rebuild

* Wed Feb 09 2011 Fedora Release Engineering <rel-eng@lists.fedoraproject.org> - 3.9.1-10
- Rebuilt for https://fedoraproject.org/wiki/Fedora_15_Mass_Rebuild

* Fri Sep 3 2010 Seth Vidal <skvidal at fedoraproject.org> - 3.9.1-9
- new update to latest head with a number of patches collected from
older bug reports.

* Mon Aug 30 2010 Seth Vidal <skvidal at fedoraproject.org> - 3.9.1-8
- update to latest head patches

* Thu Jul 22 2010 David Malcolm <dmalcolm@redhat.com> - 3.9.1-7
- Rebuilt for https://fedoraproject.org/wiki/Features/Python_2.7/MassRebuild

* Tue Apr 13 2010 James Antill <james@fedoraproject.org> 3.9.1-6
- Update to upstream HEAD.
- LOWSPEEDLIMIT and hdrs

* Fri Feb 19 2010 Seth Vidal <skvidal at fedoraproject.org> - 3.9.1-5
- add patch to allow reset_curl_obj() to close and reload the cached curl obj

* Thu Nov 12 2009 Seth Vidal <skvidal at fedoraproject.org> - 3.9.1-4
- reset header values when we redirect and make sure debug output will work

* Wed Nov 11 2009 Seth Vidal <skvidal at fedoraproject.org> - 3.9.1-3
- fixing a bunch of redirect and max size bugs

* Fri Sep 25 2009 Seth Vidal <skvidal at fedoraproject.org> - 3.9.1-2
- stupid patch

* Fri Sep 25 2009 Seth Vidal <skvidal at fedoraproject.org> - 3.9.1-1
- 3.9.1

* Tue Aug 18 2009 Seth Vidal <skvidal at fedoraproject.org> - 3.9.0-8
- ssl options, http POST string type fixes

* Mon Aug 10 2009 Seth Vidal <skvidal at fedoraproject.org> - 3.9.0-6
- reget fixes, tmpfiles no longer made for urlopen() calls.

* Wed Aug 5 2009 Seth Vidal <skvidal at fedoraproject.org> - 3.9.0-5
- apply complete patch to head fixes: timeouts, regets, improves exception raising

* Tue Aug 4 2009 Seth Vidal <skvidal at fedoraproject.org> - 3.9.0-4
- timeout patch for https://bugzilla.redhat.com/show_bug.cgi?id=515497


* Thu Jul 30 2009 Seth Vidal <skvidal at fedoraproject.org> - 3.9.0-1
- new version - curl-based

* Wed Apr 8 2009 James Antill <james@fedoraproject.org> 3.0.0-15
- Fix progress bars for serial consoles.
- Make C-c behaviour a little nicer.

* Fri Mar 13 2009 Seth Vidal <skvidal at fedoraproject.org>
- kill deprecation warning from importing md5 if anyone uses keepalive

* Mon Mar 9 2009 Seth Vidal <skvidal at fedoraproject.org>
- apply patch for urlgrabber to properly check file:// urls with the checkfunc

* Thu Feb 26 2009 Fedora Release Engineering <rel-eng@lists.fedoraproject.org> - 3.0.0-12
- Rebuilt for https://fedoraproject.org/wiki/Fedora_11_Mass_Rebuild

* Fri Nov 28 2008 Ignacio Vazquez-Abrams <ivazqueznet+rpm@gmail.com> 3.0.0-11
- Rebuild for Python 2.6

* Wed Oct 14 2008 James Antill <james@fedoraproject.org> 3.0.0-10
- Have the progress bar have a small bar, for a virtual size doubling.

* Thu Jul 10 2008 James Antill <james@fedoraproject.org> 3.0.0-9
- Make urlgrabber usable if openssl is broken
- Relates: bug#454179

* Sun Jun 15 2008 James Antill <james@fedoraproject.org> 3.0.0-9
- Don't count partial downloads toward the total

* Sat May 18 2008 James Antill <james@fedoraproject.org> 3.0.0-8
- Tweak progress output so it's hopefully less confusing
- Add dynamic resizing ability to progress bar
- Resolves: bug#437197

* Fri May 2 2008 James Antill <james@fedoraproject.org> 3.0.0-7
- Fix reget's against servers that don't allow Range requests, also tweaks
- reget == check_timestamp, if anyone/thing uses that.
- Resolves: bug#435156
- Fix minor typo in progress for single instance.

* Mon Apr 7 2008 James Antill <james@fedoraproject.org> 3.0.0-6
- Fix the ftp byterange port problem:
- Resolves: bug#419241
- Fixup the progress UI:
- add function for total progress
- add total progress percentagee current download line
- add rate to current download line
- use dead space when finished downloading
- don't confuse download rate on regets.

* Sat Mar 15 2008 Robert Scheck <robert@fedoraproject.org> 3.0.0-5
- Make sure, that *.egg-info is catched up during build

* Mon Dec 3 2007 Jeremy Katz <katzj@redhat.com> - 3.0.0-4
- Ensure fds are closed on exceptions (markmc, #404211)

* Wed Oct 10 2007 Jeremy Katz <katzj@redhat.com> - 3.0.0-3
- fix type checking of strings to also include unicode strings; fixes
regets from yum (#235618)

* Mon Aug 27 2007 Jeremy Katz <katzj@redhat.com> - 3.0.0-2
- fixes for package review (#226347)

* Thu May 31 2007 Jeremy Katz <katzj@redhat.com> - 3.0.0-1
- update to 3.0.0

* Wed Dec 6 2006 Jeremy Katz <katzj@redhat.com> - 2.9.9-5
- rebuild for python 2.5

* Wed Dec 6 2006 Jeremy Katz <katzj@redhat.com> - 2.9.9-4
- fix keepalive (#218268)

* Sat Nov 11 2006 Florian La Roche <laroche@redhat.com>
- add version/release to "Provides: urlgrabber"

* 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

* Tue May 16 2006 Jeremy Katz <katzj@redhat.com> - 2.9.9-1
- update to 2.9.9

* Tue Mar 14 2006 Jeremy Katz <katzj@redhat.com> - 2.9.8-2
- catch read errors so they trigger the failure callback. helps catch bad cds

* Wed Feb 22 2006 Jeremy Katz <katzj@redhat.com> - 2.9.8-1
- update to new version fixing progress bars in yum on regets

* Fri Dec 09 2005 Jesse Keating <jkeating@redhat.com>
- rebuilt

* Wed Sep 21 2005 Jeremy Katz <katzj@redhat.com> - 2.9.6-4
- don't use --record and list files by hand so that we don't miss
directories (#158480)

* Wed Sep 14 2005 Jeremy Katz <katzj@redhat.com> - 2.9.6-3
- add directory to file list (#168261)

* Fri Jun 03 2005 Phil Knirsch <pknirsch@redhat.com> 2.9.6-2
- Fixed the reget method to actually work correctly (skip completely transfered
files, etc)

* Tue Mar 8 2005 Jeremy Katz <katzj@redhat.com> - 2.9.6-1
- update to 2.9.6

* Mon Mar 7 2005 Jeremy Katz <katzj@redhat.com> - 2.9.5-1
- import into dist
- make the description less of a book

* Mon Mar 7 2005 Seth Vidal <skvidal@phy.duke.edu> 2.9.5-0
- 2.9.5

* Thu Feb 24 2005 Seth Vidal <skvidal@phy.duke.edu> 2.9.3-0
- first package for fc3
- named python-urlgrabber for naming guideline compliance

Write Preview
Loading…
Cancel
Save