2
1
What's changed in loggerhead?
3
2
=============================
5
- Add startup deamon script for Linux (Marius Kruger)
7
- Switch navigation from file_ids to paths. Fixes bugs #260363, #269365
8
and #128926. (Martin Albisetti)
10
- Fix bug #258710 ("the /files page explodes in an empty branch").
11
Also minor improvements to the /files and /changes pages.
14
- Added --port, --host and --prefix options to serve-branches
15
script. (Martin Albisetti)
17
- Fixed broken template for project browsing with start-loggerhead
20
- Added --reload options to restart the application when a python
21
file change. (Guillermo Gonzalez)
23
- Added error handling middleware. (Guillermo Gonzalez)
25
- Fix bug #243415 ("Tracebacks go to console but not log
26
file"). Also minor improvements to logging in serve-branches and
27
start-loggerhead. (Guillermo Gonzalez)
32
- Download a diffs between revisions. (Martin Albisetti)
34
- Modified templates to make loggerhead's theme easier to
37
- Default sqlite interface is now sqlite3. (Robert Collins)
39
- New ninja theme sponsored by Canonical (Martin Albisetti)
41
- Added COPYING file and clarified copyright headers (John Arbash Meinel)
43
- Remove the .py extension requiered by the Debian Policy.
46
- New startup script serve-branches will serve Loggerhead without
47
the need of configuration, and allow you to browse through directories
48
and branches. (Michael Hudson)
50
- Loggerhead is no longer a TurboGears application, but rather a
51
WSGI application built using Paste (see http://wsgi.org/ and
52
http://pythonpaste.org/ for more about WSGI and Paste).
54
- URLs now use revision numbers instead of revision ids (Martin Albisetti)
56
- The scripts no longer insist on Python 2.4 -- loggerhead works
59
- Bazaar as of version 1.5 has improved in performance enough that
60
the revision cache no longer gave any noticeable benefit, so it
61
was removed (the files-changed cache is still useful).
63
- The templates were rewritten in Zope's TAL markup, powered by
64
the simpleTAL library -- improving both the performance and
65
memory consumption of rendering by a factor of around 3 for
66
large pages over the old Kid templates.
68
- Loggerhead's poorly performing text index was disabled. bzr-search
69
is now used if the plugin is installed and a text index is present
70
on the branch being viewed. (Martin Albisetti, Robert Collins).
72
- Loggerhead no longer depends on bzrlib functions deprecated in
73
Bazaar 1.5 and removed in 1.6 (Martin Albisetti).
75
- The daemonization code was made more regular, fixing bugs
76
#139161 ("Starting loggerhead process may not close its stdin
77
and stdout properly") and #211526 ("Codebrowse log directory has
78
unnecessarily permissive permissions")
80
- Some confusion about what the 'file_id' query argument means was
81
cleared up: filter_file_id now means "filter revisions to those
82
that affect this file" in all views and file_id means "examine
83
this file/directory" in the annotate and inventory views.
85
- Dates are present more compactly.
87
- The loggerhead.conf file can specify which network interface to
88
bind to (Mattias Eriksson)
93
- The changelog view was out of order when not using the revision
98
- Michael Hudson <michael.hudson@canonical.com> has mostly taken
99
over the maintenance of loggerhead.
101
- loggerhead now has a simple test suite -- use 'nosetests' in the
102
loggerhead directory to run it.
104
- The rendering performance of pages that display large amounts of
105
text was improved by a factor of 4 or so.
107
- loggerhead no longer caches the text of the diffs between
108
revisions. It rarely helped and wasted a lot of disk space.
110
- The layout of most pages was made more "tight" to waste less
111
screen real estate, and some other UI improvements (Kent
114
- Much dead code was removed.
116
- Loggerhead now computes the files changed between revisions only
117
when it needs to know this. This is a fairly expensive
118
operation in Bazaar, and doing it less massively improves
119
performance and memory usage in some situations.
121
- Loggerhead now takes a read lock on the branch being viewed for
122
the duration of each request, which improves performance
125
- Loggerhead no longer uses the corruption-prone Berkely DB-backed
126
shelve module for its caches, and rather (ab)uses a sqlite
129
- The creation of History objects is much faster for large
130
branches (Aaron Bentley).
132
- Allow the construction of URLs using revnos and file paths as
133
well as revids and fileids (bug #98826):
135
- For changes pages, append the newest revno to display to the
136
URL, like http://.../changes/<revno>
138
- For annotate pages, append the revno to display to the URL,
139
followed by the path, like http://.../annotate/<revno>/<path>
141
- For file listing and revision pages append the revno to
142
display to the URL, like http://.../files/<revno>
144
Loggerhead still generates URLs using revision and file ids for
147
- Many bugs were fixed:
149
- Loggerhead does not escape special characters when embedding a
150
revision ID in a URL (bug #88286)
152
- Improved robustness in the face of ghosts.
154
- Don't crash on displaying a commit with an empty message (bug
157
- codebrowse fails with infinite redirections (James Henstridge,
160
- Loggerhead fails to browse revisions that change binary files
161
(James Henstridge, bug #91686)
163
- Loggerhead atom feeds expose internal hostname (James
164
Henstridge, bug #93585)
166
- loggerhead don't like page break character (0x0C) in text
169
- codebrowse source listings don't contain line number anchors
172
- only serve up unescaped user content with "Content-Disposition:
175
- viewing the file listing consumes a lot memory (bug #116869)
177
- loggerhead can't handle empty branches (bug #119228)
179
- upgrading the format of a branch behind loggerhead could make
180
it fail for that branch (bug #118673)
182
- Error parsing non-ascii content (bug #117799)
184
- Loggerhead failed on whitespace-only commit messages.
186
- Links to diffs from within a revision do not work (bug
189
- UTF-8 patches served as ISO-8859-1, fixed by served bundles as
190
application/octet-stream (bug #121336)
192
- TurboGears was turning query arguments into unicode, and bzr
193
stopped accepting unicode revids (bug #175228)
8
199
- fix broken inventory page (oops!)
10
201
- fix a few rendering problems with kid and safari
added
removed
NEWS
