summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authormstenner <mstenner>2005-10-22 22:05:11 +0000
committermstenner <mstenner>2005-10-22 22:05:11 +0000
commit1e7b251a97d831627264d0f11fc76a1a45e35652 (patch)
treea9d6db4913ac23be8c1cf165f0cc9383b0164221
parent219f632bd6e8e7b93b00626dd6d4b40431baac65 (diff)
downloadurlgrabber-1e7b251a97d831627264d0f11fc76a1a45e35652.tar.gz
Updated TODO and preparing for release.
-rw-r--r--ChangeLog66
-rw-r--r--TODO10
-rw-r--r--urlgrabber/__init__.py6
3 files changed, 69 insertions, 13 deletions
diff --git a/ChangeLog b/ChangeLog
index 0859a40..c69ceca 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,69 @@
+2005-10-22 Michael D. Stenner <mstenner@linux.duke.edu>
+
+ * test/test_grabber.py, test/test_keepalive.py,
+ test/test_mirror.py, test/support/testdata/test_post.php,
+ urlgrabber/byterange.py, urlgrabber/grabber.py,
+ urlgrabber/keepalive.py, urlgrabber/mirror.py:
+
+ Added some slick logging support to urlgrabber. It can be
+ controlled by the calling application or enabled from an
+ environment variable.
+
+ Allowed for HTTP POSTs by passing in the kwarg "data", just as for
+ urllib2.
+
+2005-08-19 Michael D. Stenner <mstenner@linux.duke.edu>
+
+ * urlgrabber/progress.py:
+
+ Another minor simplification offered by the WTF folks.
+
+2005-08-18 Michael D. Stenner <mstenner@linux.duke.edu>
+
+ * urlgrabber/progress.py:
+
+ Fixed a minor formatting bug, helpfully pointed out by the nice
+ folks at the daily WTF:
+ http://thedailywtf.com/forums/41204/ShowPost.aspx
+
+2005-08-17 Michael D. Stenner <mstenner@linux.duke.edu>
+
+ * test/test_grabber.py, urlgrabber/grabber.py:
+
+ Added Menno's idea for catching KeyboardInterrupt. This allows the
+ program to do things like hit the next retry, jump to the next
+ mirror, etc. when the user hits ctrl-c.
+
+ Also modified the behavior of failure_callback slightly. It now
+ gets called for EVERY failure, even the last one. That is, if
+ there's a failure on the last try, the failure callback gets called
+ and (assuming it doesn't raise anything) the exception gets raised.
+ Previously, it would not be called on the last failure. This is
+ another MINOR compatibility break.
+
+2005-08-17 Michael D. Stenner <mstenner@linux.duke.edu>
+
+ * test/test_grabber.py, test/test_mirror.py, urlgrabber/grabber.py:
+
+ except HTTPError, which basically means all HTTP return codes that
+ are not 200 and not handled internally (such as the 401
+ Unauthorized code). Previously, these were caught as IOErrors (of
+ which HTTPError is an indirect subclass) and re-raised as
+ URLGrabErrors with errno 4. They are NOW reraised as URLGrabErrors
+ with errno 14, and with the .code and .exception attributes set to
+ the HTTP status code and HTTPError exception object respectively.
+
+ This represents a minor compatibility break, as these "errors" are
+ now not handled the same way. They will be raised with a different
+ errno and errno 14 is not in the default list of retrycodes
+ (whereas 4 is).
+
+2005-06-27 Michael D. Stenner <mstenner@linux.duke.edu>
+
+ * ChangeLog, urlgrabber/byterange.py:
+
+ Fixed two namespace-related bugs.
+
2005-05-19 Michael D. Stenner <mstenner@linux.duke.edu>
* test/test_keepalive.py, urlgrabber/keepalive.py:
diff --git a/TODO b/TODO
index 979e179..ad1dc8a 100644
--- a/TODO
+++ b/TODO
@@ -18,14 +18,6 @@ ALPHA 2:
- starting a download with some parts already read (with reget,
for example)
- * Add decent debugging code and a URLGRABBER_DEBUG environment
- variable so that users of miscellaneous applications can get the
- debugging output
-
- * Have a plan for KeyboardInterupt exception handling. This is driven by
- recent chatter on the yum list about Ctrl-C handling. Seth suggests
- Ctrl-C as a "Skip to Next Mirror" trigger when processing MirrorGroups.
-
* look at making the 'check_timestamp' reget mode work with ftp.
Currently, we NEVER get a timestamp back, so we can't compare.
We'll probably need to subclass/replace either the urllib2 FTP handler
@@ -33,8 +25,6 @@ ALPHA 2:
be worth it just for this one mode of reget. It fails safely - by
getting the entire file.
- * ipv6
-
* cache dns lookups -- for a possible approach, see
https://lists.dulug.duke.edu/pipermail/yum-devel/2004-March/000136.html
diff --git a/urlgrabber/__init__.py b/urlgrabber/__init__.py
index 5189ba2..1aca993 100644
--- a/urlgrabber/__init__.py
+++ b/urlgrabber/__init__.py
@@ -14,7 +14,7 @@
# Copyright 2002-2004 Michael D. Stenner, Ryan Tomayko
-# $Id: __init__.py,v 1.14 2005/03/08 22:54:25 mstenner Exp $
+# $Id: __init__.py,v 1.15 2005/10/22 22:05:12 mstenner Exp $
"""A high-level cross-protocol url-grabber.
@@ -44,8 +44,8 @@ following features:
automatically switching mirrors if there is a failure.
"""
-__version__ = '2.9.6'
-__date__ = '2005/03/08'
+__version__ = '2.9.7'
+__date__ = '2005/10/22'
__author__ = 'Michael D. Stenner <mstenner@linux.duke.edu>, ' \
'Ryan Tomayko <rtomayko@naeblis.cx>'
__url__ = 'http://linux.duke.edu/projects/urlgrabber/'