1
2
What's changed in loggerhead?
2
3
=============================
7
- New ninja theme sponsored by Canonical (Martin Albisetti)
9
- Remove the .py extension requiered by the Debian Policy.
12
- Loggerhead is no longer a TurboGears application, but rather a
13
WSGI application built using Paste (see http://wsgi.org/ and
14
http://pythonpaste.org/ for more about WSGI and Paste). XXX more.
16
- zpt.cleaner_urls bug 138021 ("loggerhead should generate links
17
based on revision numbers and file paths"). XXX more.
19
- The scripts no longer insist on Python 2.4 -- loggerhead works
22
- Bazaar as of version 1.5 has improved in performance enough that
23
the revision cache no longer gave any noticeable benefit, so it
24
was removed (the files-changed cache is still useful).
26
- The templates were rewritten in Zope's TAL markup, powered by
27
the simpleTAL library -- improving both the performance and
28
memory consumption of rendering by a factor of around 3 for
29
large pages over the old Kid templates.
31
- Loggerhead's poorly performing text index was disabled. bzr-search
32
is now used if the plugin is installed and a text index is present
33
on the branch being viewed. (Martin Albisetti, Robert Collins).
35
- Loggerhead no longer depends on bzrlib functions deprecated in
36
Bazaar 1.5 and removed in 1.6 (Martin Albisetti).
38
- The daemonization code was made more regular, fixing bugs
39
#139161 ("Starting loggerhead process may not close its stdin
40
and stdout properly") and #211526 ("Codebrowse log directory has
41
unnecessarily permissive permissions")
43
- Some confusion about what the 'file_id' query argument means was
44
cleared up: filter_file_id now means "filter revisions to those
45
that affect this file" in all views and file_id means "examine
46
this file/directory" in the annotate and inventory views.
48
- Dates are present more compactly.
50
- The loggerhead.conf file can specify which network interface to
51
bind to (Mattias Eriksson)
56
- The changelog view was out of order when not using the revision
62
- Michael Hudson <michael.hudson@canonical.com> has mostly taken
63
over the maintenance of loggerhead.
65
- loggerhead now has a simple test suite -- use 'nosetests' in the
66
loggerhead directory to run it.
68
- The rendering performance of pages that display large amounts of
69
text was improved by a factor of 4 or so.
71
- loggerhead no longer caches the text of the diffs between
72
revisions. It rarely helped and wasted a lot of disk space.
74
- The layout of most pages was made more "tight" to waste less
75
screen real estate, and some other UI improvements (Kent
78
- Much dead code was removed.
80
- Loggerhead now computes the files changed between revisions only
81
when it needs to know this. This is a fairly expensive
82
operation in Bazaar, and doing it less massively improves
83
performance and memory usage in some situations.
85
- Loggerhead now takes a read lock on the branch being viewed for
86
the duration of each request, which improves performance
89
- Loggerhead no longer uses the corruption-prone Berkely DB-backed
90
shelve module for its caches, and rather (ab)uses a sqlite
93
- The creation of History objects is much faster for large
94
branches (Aaron Bentley).
96
- Allow the construction of URLs using revnos and file paths as
97
well as revids and fileids (bug #98826):
99
- For changes pages, append the newest revno to display to the
100
URL, like http://.../changes/<revno>
102
- For annotate pages, append the revno to display to the URL,
103
followed by the path, like http://.../annotate/<revno>/<path>
105
- For file listing and revision pages append the revno to
106
display to the URL, like http://.../files/<revno>
108
Loggerhead still generates URLs using revision and file ids for
111
- Many bugs were fixed:
113
- Loggerhead does not escape special characters when embedding a
114
revision ID in a URL (bug #88286)
116
- Improved robustness in the face of ghosts.
118
- Don't crash on displaying a commit with an empty message (bug
121
- codebrowse fails with infinite redirections (James Henstridge,
124
- Loggerhead fails to browse revisions that change binary files
125
(James Henstridge, bug #91686)
127
- Loggerhead atom feeds expose internal hostname (James
128
Henstridge, bug #93585)
130
- loggerhead don't like page break character (0x0C) in text
133
- codebrowse source listings don't contain line number anchors
136
- only serve up unescaped user content with "Content-Disposition:
139
- viewing the file listing consumes a lot memory (bug #116869)
141
- loggerhead can't handle empty branches (bug #119228)
143
- upgrading the format of a branch behind loggerhead could make
144
it fail for that branch (bug #118673)
146
- Error parsing non-ascii content (bug #117799)
148
- Loggerhead failed on whitespace-only commit messages.
150
- Links to diffs from within a revision do not work (bug
153
- UTF-8 patches served as ISO-8859-1, fixed by served bundles as
154
application/octet-stream (bug #121336)
156
- TurboGears was turning query arguments into unicode, and bzr
157
stopped accepting unicode revids (bug #175228)
163
8
- fix broken inventory page (oops!)
165
10
- fix a few rendering problems with kid and safari
added
removed
NEWS
