[Mono-announce-list] ANNOUNCE: Beagle 0.2.17
joe at joeshaw.org
Tue May 8 15:40:48 EDT 2007
I'm pleased to announce the release of Beagle 0.2.17. This is a bug
THE SHORT OF IT
To download the 0.2.17 release or learn more about Beagle, visit
the Beagle web page:
WHAT IS BEAGLE?
Beagle is a Linux desktop-independent service which transparently and
unobtrusively indexes your data in real-time. For example:
* Files are immediately indexed when they are created, are re-indexed
when they are modified, and are dropped from the index upon deletion.
* E-mails are indexed upon arrival.
* IM conversations are indexed as you chat, a line at a time.
* Web pages are indexed as you view them (with a browser extension).
Beagle supports many different data sources and file formats. For a
complete listing, visit:
Beagle uses the Lucene indexing system from the Apache project and the
prodigious Doug Cutting, ported to .NET by George Aroush.
Beagle includes an optional GNOME-based graphical tool for searching
To learn more about Beagle, visit:
The Beagle daemon requires at least:
* Mono 184.108.40.206
* glib-sharp 2.4.0
* GMime 2.2.0
* libexif 0.5.7
* sqlite 2.x or 3.3.1
The GNOME user interface requires:
* gtk-sharp 2.4.0
* gnome-sharp 2.4.0
To support all of Beagle's features and file formats, you should
* Mono 1.2.2 or better
* Firefox 1.5 or Epiphany
* Evolution-sharp 0.11.1 for Evolution Data Server 1.4 or 1.6,
or 0.12.2 for EDS 1.8 or 1.9.
* libgsf 1.14.1
* gsf-sharp 0.8.1 from
* wv1 1.2.4
* pdfinfo and pdftotext from xpdf
* MPlayer or Totem 2.17.1
* Galago 0.5.x
* Linux kernel 2.6.13 with inotify
NOTABLE CHANGES BETWEEN 0.2.16 AND 0.2.16.3
* When extracting files from archives, save the temporary files with
the appropriate file extensions (rather than .tmp) so that
extension-based MIME type detection works correctly.
* Handle phrase queries that contained stop words much better.
For example, previously "no dice" would return no hits.
* Make the archive filter much more robust. [bgo #415056, bnc #244490
* Add --disable-on-battery to beagle-build-index, add support for
it in beagle-crawl-system, and enable it for the documentation
index, since it takes a while to build. [bgo #383789]
* Add --disable-text-cache to the Beagle daemon, to disable storing of
text for snippets. This saves substantial amounts of disk space at
the expense of showing snippets in the UI.
* Fix a bug in which deleted files weren't removed from the text cache
* Fix prohibited date queries (ie, -date:20070505) in most cases.
* Fix an infinite loop on certain SVG files. [bgo #407564]
* Fix an infinite loop on some truncated JPEG files. [bgo #405372]
* Limit PDF tools to 100 megs mapped memory. [bgo #335461]
* Add more well-known locations for Docbook files, since their MIME
type often cannot be determined beyond generic XML types.
* Use the TOMBOY_HOME environment variable if available. [bgo #332102]
* Fix the CHM filter. [bgo #408047]
* Fix opening of Konversation hits in beagle-search. [bgo #408101]
* Shut down the daemon sooner if requested in the middle of a large
batch of files being indexed.
* Many other minor filter and backend fixes.
NOTABLE CHANGES BETWEEN 0.2.16.3 AND 0.2.17
* Redirect low-level stdout and stderr to the log file, rather than
just .Net streams so that Mono crashes are logged and saved
* Lower CPU priority in the Beagle daemon, for instances where we get
a little CPU hungry.
* Add a new index verification process if stale lockfiles are left
behind for whatever reason (usually an index helper crash). The
process is rather CPU intensive, but overall will be less painful
than purging the index and reindexing everything.
* Watch the X connection more closely, fixing a bug in which the
daemon wouldn't exit when the user logged out if it didn't have any
work to do.
* Always start the daemon with --replace, so that no more than one
copy is running per user.
* Speed up the returning of results by about 10%.
* Remove separate crawling and indexing threads from
beagle-build-index, causing it to be slightly faster and use a lot
* Speed up beagle-index-info by returning cached document counts, and
not starting up a helper process to get them.
* Fix an infinite loop problem with empty OpenOffice documents. [bgo
* Better handle PDF files which spew tons of warnings with
pdfinfo. [bgo #425150]
* Display the first Nautilus emblem on file results, and fix an
uncommon memory leak in the beagle-search UI.
* Remove the Mono --debug argument from beagle-search, saving about 2
megs RSS and making startup faster. Also remove it from things like
* Fix a bug in which mail attachments which didn't have a file name
weren't being displayed in the UI.
* Fix bugs in counting the total number of results in the UI.
* Fix a crash caused by empty buddy icons. [Launchpad #93064]
* Fix a crash caused by empty icons from Liferea. [bgo #428336]
* Fix an exception caused by Evolution IMAP folders with commas in
their names. [bgo #432417, lp #108957]
* Redirect MS Word extractor script output to stderr so that it
doesn't interfere with the indexing process.
* Fix some file modification date handling issues in the daemon.
* Exit the beagle-crawl-system cronjob if beagle-build-index doesn't
exist. [bgo #370943]
* Always set the process name in beagle-extract-content.
* Add application/xhtml+xml as a MIME type for the HTML filter.
* Sync our copy of xdgmime with the upstream GTK+ version.
- Arabic (Djihed Afifi)
- Brazilian Portuguese (Igor Pires Soares)
- British English (David Lodge)
- Czech (Jakub Friedl)
- Dzongkha (Pema Geyleg)
- Finnish (Ilkka Tuohela)
- French (Guillaume Ayoub, Stéphane Raimbault,
Christophe Bliard, Claude Paroz)
- Galician (Ignacio Casal Quinteiro)
- Hungarian (Gabor Kelemen)
- Japanese (Takeshi Aihana)
- Macedonian (Jovan Naumovski)
- Norwegian bokmål (Kjartan Maraas)
- Simplified Chinese (Funda Wang)
- Spanisn (Roberto Majadas, Jorge Gonzalez)
- Swedish (Daniel Nylander)
Contributors to this release: Joe Shaw, Debajyoti Bera, Jose Carlos
Garcia Sogo, Alexander Macdonald.
Roughly 43 GNOME Bugzilla bugs were fixed between 0.2.16 and 0.2.17:
Full set of changes:
The Thunderbird backend is memory hungry and buggy. It is now disabled by
default, but can be reenabled by passing in --enable-thunderbird to
There are some race conditions that can occur with certain combinations of
file system operations. In very rare cases it might be necessary to stop
and restart the daemon.
At this point in development, we cannot commit to stable APIs or file
formats. You will almost certainly need to reindex your data at some
point in the future.
More information about the Mono-announce-list