The most important change between release 0.3.8 and 0.4.0.1 is:
o) The emulation of MIPS processors has been completely rewritten;
it now uses the same portable dynamic translation system as the
ARM and PowerPC emulation modes.
On Alpha and i386 hosts (and AMD64 hosts running in 32-bit mode),
GXemul previously used translation into native code. This release
will perform worse than 0.3.8 on those host architectures.
On all other hosts (including AMD64 running in native 64-bit mode),
0.4.0 is likely to be faster than 0.3.8, when emulating MIPS.
I think that in the long term, moving towards full portability like
this is a good idea.
(0.4.0 was a bit buggy and unstable; 0.4.0.1 is a quick-fix release.)
There have also been many other changes, including, but not limited to:
o) The "test machine" functionality is more well-defined than before,
and some tutorial-like demos have been added. These could be useful
e.g. in operating system construction courses.
o) NetBSD/sgimips 3.0 works now. This is most likely due to the rewrite
of the MIPS emulation mode. Previous releases of GXemul only worked
with NetBSD/sgimips 2.1.
o) I have begun implementing rudimentary support for GDB remote serial
protocol connections. This means that you can run e.g. the Data
Display Debugger, and connect it to a GXemul instance.
No advanced GDB functionality is working yet, but starting and
stopping the emulated machine and single-stepping should work.
This is a major update, too many individual changes to list here.
Most notably: uses the gstreamer-0.10 framework now.
(That's still not that great software, there is an appearent ressource
leak which makes switching between play channels fail after a while.)
changes:
-UI improvements
* Add Impulse Tracker and MOD files to the list of supported types
* Add audio/vnd.rn-realaudio as a supported playlist format
-bugfixes
1.9.8 - 18 June 2006
o Welcome to GnuCash 1.9.8 aka "Grab that cash with both hands and
make a stash" the first release candidate of the GnuCash Open Source
Accounting Software which will eventually lead to the stable version
2.0.0. This release contains many bugfixes since the second beta
release.
o PLEASE TEST any and all features important to you. Then post any
bugs you find to bugzilla
http://bugzilla.gnome.org/enter_bug.cgi?product=GnuCash
o We would like to encourage people to test this and any further
releases as much as possible and submit bug reports in order that we
can polish GnuCash to be as stable as possible for the 2.0.0 release
in a few weeks.
o Major bugfixes include -
o More Register fixes.
o Restore the business reports menu.
o QIF import and HBCI fixes.
o Fixes for currencies.
o Fixes to business invoices.
o General UI fixes.
2006/06/19: version 2.7.7 = tag release-2-7-7
5196: IP: Store IPs with two int values instead of four to save RAM (pango)
5195: BT: Log tracker error messages in UTF-8
5194: HTML: Print warning on opening page when enable_servers is set to false
2006/06/18
5191: HTML: remove sysinfo on opening page & option motd_html
5192: Fix deactivation of verbosity "gui"
5163: Configure: CVS SCM support for Mac OS X 10.3 (gstat)
2006/06/17
5158: HTML: Use table for printing buildinfo and runinfo data (schlumpf, pango)
2006/06/16
5187: New option: BT_import_new_torrents_interval,
ignore Thumbs.db & desktop.ini in directory lists
5186: GUI: Send only non-empty real_startup_message string
2006/06/15
5156: HTML, improve Options output (schlumpf)
* new command html_mods_use_js_helptext (true|false)
to change display of helptext between javascript popup`s and html table
* fix some padding and border bugs
* add "toggle js_helptext" button to 'vo' and 'voo' to directly toggle the style
2006/06/12
5061: New option small_files_slot_limit to push release of small files (pango)
all files below the configured file size (default 10kb) are uploaded
through one reserved slot
5144: Increase default max_upload_rate to 10 (schlumpf)
5177: HTML, Options: update display after changing options (schlumpf)
5176: Swarming: Change error message text and verbosity for BAD WRITE
BAD WRITE is not really an error, MLDonkey just receives unwanted data
5175: GTK1 oldgui: Last compile fix for abstract verification bitmaps (pango)
5174: Abstract verification bitmaps:
compile fixes for GTK1 GUIs, move generic code to CDK (pango)
5173: Swarmer: Small fix for merged downloads (WIP7b') (pango)
2006/06/11
5172: EDK: Log error which causes pausing a download
5171: New command group "Driver/Users",
print help text for "Bad number of arguments" error message
5170: Fix default max_indirect_connections, minimum is 30
5169: HTML, Options: add description to option section buttons,
remove 'save' button (schlumpf)
5168: Global module to abstract verification bitmaps (pango)
5167: Swarmer: Fix saving ini data for merged downloads (pango)
5166: Swarmer: New module to abstract verification bitmaps (pango)
2006/06/07
5164: CommonSwarming: Pangos WIP5e'
* When cutting a range, the reciprocal link of the range after the right part
still references the left part
* When cutting a block in two parts can before empty, but then can also before
complete (if they have no ranges left)
* Added bad_writes_is_back patch
* When find_range can only find ranges already being downloaded, it now first
does a quick check to see if there's no other, probably more interesting,
blocks remaining. If so, it forces a block change (report no more ranges left)
* While it's a good feature in itself, it's probably a workaround for a bug in
select_block (?); It seems that blocks oversaturated with sources happen way
too often (4 or 5 sources on a single range shouldn't happen).
* Compute the size of unselected ranges in each blocks, so that sources don't
"rush" on blocks just left by a source.
* Some BAD WRITEs seems to happen because we forget ranges we requested
(when switching blocks ?)
* Replaced call to clear_uploader_ranges by a call to clear_uploader_intervals
in BitTorrent support Choke message handler.
5063: EDK: remove duplicate entries in shared_files_new.ini (pango)
5160: Remove stale avifile.ml, clean commonMultimedia logging
5159: Compute magic values only when needed
5157: add some line breaks to option help text for shorter lines (schlumpf)
2006/06/01
5153: New command: force_web_infos to manually download web_infos URLs
5152: Do not connect friends from disabled networks
5151: Options: New stringvalue functions needed for multiuser support
5150: Log: Fix log when server list is empty in check_blocked_servers
2006/05/31
5149: Log: Some additions for commonSwarming
5148: Mail: Fix for multiline SMTP responses (zet)
5143: HTML/Telnet: Country codes for servers
5147: G1/G2: Display proper filename (and therefore support rename) (zet)
5145: EDK: Fix md4 hashing when filesize is an exact multiple
of edk chunk size, MLDonkey is now eMule-compatible (zet)
2006/05/30
5141: Swarmer: New option swarming_block_selection_algorithm (pango)
* Allow to choose swarmer block selection algorithm at runtime.
- Algorithm 1 is the current default algorithm
- Algorithm 2, from WIP3o patch, should try harder to complete partially
downloaded chunks, specially with Bittorrent. It was previously
reverted because of unknown performance problem (slower download ?)
5140: Update .cvsignore files (pango)
5138: Catch ip banned exception (zet)
5137: HTML: Fix display of file magic values with '
5136: FileTP: fixes and enhancements (zet)
* Add "FTP-chunk_size" option (0 = no chunks = download full file at once)
-- A chunk_size would be required if downloading from > 1 source
* Add FTP downloads to bandwidth controller (for stats & speed limiter)
* Support FTP sites that do not buffer input (MS FTP)
* Add support for user/pass (http://user:pass@site or ftp://user:pass@site).
This also adds user/pass support to http_client.
* Error codes from HTTP (ie: 404) and some from FTP are supported (530 sometimes
signals a retry, sometimes a login failure so a regex was used (lftp)).
Downloads are paused when a hard error is received (and noted in the log).
2006/05/25
5131: New option gui_log_size: number of lines for GUI console messages
5129: HTML: Display client_bind_addr in options menu
5124: Log: Login messages can be disabled with verbosity "no-login"
5130: CommonSwarming: Pangos WIP4a'
thanks for testing goes to SchAmane, CruX, jave (forgot someone?)
That's a huge patch over CommonSwarming module (whose job is to select what data
to ask from connected sources, and the merging of received data),
I hope I'm not forgetting anything important :
* first, it's a mantainance patch; Lots of code have been reworked for
readability and compactness (iterators...), functions grouped together,
sometimes renamed to more descriptive names. Lots of comments have been added,
including the description of invariants I've found, and a nice ASCII art
schema of the swarming datastructures ;)
* naming has been made more consistant, to differenciate chunks (from each
network view) from blocks (internal to swarming), and intervals (list of
int64 * int64 beggining/end offset pairs)
* Support for bitmaps encoded as strings (AvailableCharBitmap) removed, since
it's no longer used anywhere; use the more compact AvailableBitv instead
* Removed range_current_begin from ranges, it's enough for a range to have a
beginning and a end ;)
* Rewrote compute_block_num to have precise preconditions, and algorithm proof
* Added some assertions to better enforce invariants
(on verification bitmaps, for example)
* Allocation of ranges is now delayed until the very last moment, instead of
when a chunk starts being downloaded. Main benefit (beside small memory gain)
is that requests for different ranges sizes can be fulfilled at any time.
* Modified received function so that no BAD WRITE message can happen anymore:
condition when data that does not start from the beginning of a range is
correctly detected and data discarded
(since ranges cannot handle that case anyway)
* Rewrote a much simpler version of present_intervals
* Rewrote block selection algorithm for readability and flexibility; select_block
simply "folds" over all available blocks, using compare_choices function to
keep the best available one(s). May need some more CPU (no fast exit)
* Fixed donkey uploaders calls when disconnected
* Reenable the copy of identical chunks between files. Opportunities for copies
are checked every 5 minutes.
* Made file writing a bit smarter in Unix32 module so that writing zeroed chunks
(as can happen a lot with chunks copying) on an otherwise sparse temp file
does not allocate disk space. A bit hackish.
2006/05/23
5121: EDK: Remove obsolete option emule_compression
5122: improve block_list output and runinfo ip blocking (schlumpf)
5123: Fix Invalid_argument("UTF8.length") bug in HTML, down-/uploaders
RMagick 1.13.0
o Added transform, transform!, transpose, transpose! methods to Image class
(available with ImageMagick 6.2.8)
o Feature #4844, add auto_orient, auto_orient! methods to Image class
(suggestion from John Oordopjes, available with ImageMagick 6.2.8)
o Added adaptive_sharpen, adaptive_sharpen_channel methods to Image class
(available with ImageMagick 6.2.7)
o Added composite_image_channel, composite_image_channel! methods to Image
class (added in ImageMagick 6.2.6)
o Added radial_blur_channel method to Image class (available in
ImageMagick 6.2.4)
o Fix bug #4806, add hash, eql? methods to Pixel class (bug report from
Tim Pease)
o Change extension filename to match RubyGems 0.9.0 expectations.
o Fix bug #4821, correct doc for Image#rotate (bug report from Tim Pease)
o Update the Draw#annotate documentation
The PKGDBDIR and PKGDBDIR.refcount directories are excluded from the
list of files to check, because for unprivileged pkgsrc, they are
subdirectories of PKGSRCDIR.
Removed the dependency on libnbcompat for NetBSD.
SciGraphica is an Open Source data analysis and visualisation tool. It
pretends to be a clone of Microcal Origin but with the aim to be fully
featured, cross-platform, user-friendly scientific package that through
its Open Source nature can be extended and/or bugfixed by users, as well
as being user-extensible with Python modules.
This package contains the support library, which is separated from the
graphical user interface.
This change includes all the things I thought of during the freeze of
2006Q2.
Some variables have disappeared from pkgsrc without notice. Marked them
deprecated to help the package authors at least a bit.
Renamed some ACL abbreviations in makevars.map and adjusted the variable
definitions.
Added the command line options -Dtrace, -Dunchecked and -Dunused.
Some variables may contain unexpected white-space and therefore should
not be used in .for loops.
The -Dmisc option replaces the --verbose option (and the log_info
subroutine), which does not exist any longer. This eliminates all
instances of "OK:" from the output, which I didn't like from the
beginning. The -Dmisc option also takes over many warnings that have
been issued by --debug before.
Improved the check for absolute pathnames. Especially, everything that
follows ${DESTDIR} or $(DESTDIR) in a Makefile is considered an absolute
pathname. This reveals many wrong uses of DESTDIR (as defined by the GNU
coding standards), for example $(DESTDIR)/$(prefix), which should be
$(DESTDIR)$(prefix) instead.
Almost every use of a make(1) variable is checked for spelling mistakes.
Before, only the definitions of variables have been checked.
Restricted the use of variables to specific files, which are defined in
makevars.map. This catches especially buildlink3.mk files with
unexpected side effects.
In shell commands, neither "echo" nor "echo -n" or "${ECHO} -n" should
be used. Since the INSTALL_*_DIR commands can only handle one directory
at a time, suggest to use INSTALLATION_DIRS instead.
Greatly improved the checks for dependency specifications, especially to
find foo-*, which should rather be foo-[0-9]*.
Fixed the incorrect handling of LICENSE_FILE (now using $line->text
instead of $line).
Improved the wording of the "plural names" warnings, so that for known
variables it is "The += operator should only be used with lists." now.
In buildlink3.mk files the uppercase and lowercase names of the package
should correspond with the package name. This catches many
copy-and-paste mistakes. Fixed many null pointer exceptions as well. In
addition, every buildlink3.mk file needs a definition of
BUILDLINK_API_DEPENDS.
In patch files, absolute pathnames and unportable macro names are
checked not only in added lines, but also in the context lines.
In the pkgsrc root directory and the category directories, the Makefiles
are checked like every other Makefile, too.
Fixed the place where the global variables for the package context are
defined and later undefined.
Variables that are mentioned in EVAL_PREFIX may be used in the current
file and are not flagged as "used but not defined".
When parsing shell words in Makefiles, recognize $<, $@, $^ and $/.
(The latter is used extensively by openoffice to represent a directory
separator.)
Fixed some minor bugs:
* $makevar had been defined unintentionally by using the -> operator.
dollar character before the opening parenthesis. This should really have
been detected by bmake, but somehow the parser seems to be tolerant with
respect to syntax errors.
