With the upgrade to version 1.4.12 or 1.4.13, scim began failing on
DragonFly when PKG_DEVELOPER=yes option was used due to the missing
locale files outlined in the plist. All the USE_TOOLS callouts and
buildlinks were correct. The problem was traced back to the configure
tests for gettext. They were producing casting errors and number of
argument errors.
The workaround is to cache the result of the gettext/libintl test
as successful in the package makefile. Doing so results in successful
build, confirming the conftest has a problem at least on some platforms.
* libXp was used by Xaw8, but it had been obsolated, and in pkgsrc,
x11/libXaw/buildlink3.mk had been switched to pick up Xaw7 by default.
* With x11/xorg-cf-files, libXp was offered with XawClientLibs,
but updated to 1.0.4, it was removed.
* And pkgsrc had been switched to use always xorg-cf-files and imake from pkgsrc,
so all platforms should not require libXp from libXaw with Imake.
Bump PKGREVISION.
Noteworthy changes in version 1.4.1 (2012-03-29)
------------------------------------------------
* Use libskk's X keysym handling.
Noteworthy changes in version 1.4.0 (2012-01-25)
------------------------------------------------
* Initial release using libskk.
1.4.1
* Handle exceptions properly in ibus-setup-hangul
https://github.com/choehwanjin/ibus-hangul/pull/5
* Update icon files
* Support ibus icon symbol feature
1.4.0
* Require ibus 1.4
* Add desktop menu for ibus-setup-hangul
* Add an option for automatic reordering
* Add an option for a word preedit text
* Update MS IME compatible symbol table
* Use surrounding text feature on converting to hanja
2012-02-29 IRIE Shinsuke <irieshinsuke@yahoo.co.jp>
* Version0.3.2
* Added option `ibus-agent-search-paths' to fix problem that ibus-el-agent
cannot be found automatically if it has been moved to another directory
after loading ibus.el
* Fixed problem: Message of Xlib's BadWindow error sometimes is displayed
in particular environments (LP #932517)
* Changed command line options for ibus-daemon automatically launched
to start XIM server together ("-d" -> "-dx")
2012-02-14 IRIE Shinsuke <irieshinsuke@yahoo.co.jp>
* Version0.3.1
* Fixed bug: An error occurs when using old version of IBus (< 1.4.0) which
doesn't support surrounding text (LP #930015)
* Fixed bug: ibus-mode aborts if newly created buffer is immediately
killed before receiving IMContext ID from IBus daemon (LP #930820)
* Fixed inappropriate error message "Couldn't receive data from agent."
in the synchronous communication
* Fixed bug: Prefix argument doesn't work for unhandled key events
* Fixed echo area's display for multiple strokes key sequence
* Changed to execute Emacs command directly as fallback for unhandled key
events, except for partial key sequences such as C-x
* Removed README.Debian file (Debian #648596)
2011-12-24 IRIE Shinsuke <irieshinsuke@yahoo.co.jp>
* Version0.3.0
* Added support for surrounding text
* Added option `ibus-candidate-window-h-offset'
* Added option `ibus-prediction-window-h-offset'
* Added command `ibus-enable-specified-engine'
* Changed to enable setting of ibus-prediction-window-position=0 even in
minibuffer if IBus version is 1.3.8 or later
* Changed to show underlines even if their thicknesses are zero
* Fixed problem: Table created by `table-insert' command might corrupt
when preediting aborts
* Fixed problem: Preediting may abort when using some engines such as m17n
* Fixed problem: Input focus cannot be detected on Ubuntu Unity desktop
because get_input_focus() often returns root window incorrectly
* Fixed bug: auto-fill-mode doesn't work (LP #784452)
* Fixed bug: Input focus cannot be obtained correctly in StumpWM (LP #856358)
Released 1.2.7
Added --with-layout option in configure for IBusEngineDesc.
Added symbol and hotkeys property in IBusEngineDesc.
Added --with-layout option in configure for IBusEngineDesc.
New ibus 1.5 will use the IBusEngineDesc.layout but Japanese input method
inherits the system keyboard layout so I added the configuration.
Probably ibus-anthy would need IBusEngineDesc.layout = 'default'.
Added symbol and hotkeys property in IBusEngineDesc.
Fixed UTF-8 copyright on about dialog.
Fixed undefined getuid in engine.py
Released 1.0.1
Added Rupee symbol (U+20B9)
Removed gtk_set_locale() for GTK3.
Followed eek 0.90 API changes.
Added some acronyms.
Set XTEST module by default if it's installed.
Added Runes group in Custom Characters table.
Added variants and options.
libskk -- a library to deal with Japanese kana-to-kanji conversion method
Features:
* Support basic features of SKK including new word registration into
dictionary, completion, numeric conversion, abbrev mode, kuten
input, hankaku-katakana input, Lisp expression evaluation (concat
only), and re-conversion.
* Support various typing rules including romaji-to-kana, AZIK,
TUT-Code, and NICOLA.
* Support various dictionary types including file dictionary (such as
SKK-JISYO.[SML]), user dictionary, skkserv, and CDB format
dictionary.
* GObject based API with gobject-introspection support.
* Experimental support for intelligent kana-to-kanji conversion based
on Viterbi algorithm.
Remove devel/py-ctypes (only needed by and supporting python24).
Remove PYTHON_VERSIONS_ACCEPTED and PYTHON_VERSIONS_INCOMPATIBLE
lines that just mirror defaults now.
Miscellaneous cleanup while editing all these files.
Changelog:
* Simplify skk-isearch
* Add some menu items
* Display input mode indicator near cursor position
* Get annotation from DictionaryServices on Mac OS X
* Get annotation from other program
* Make $ behavior changable
* Remove some combination from skk-azik.el
* Improve Tan-Kanji transform
* Add skk-search-web.el
* Other bugfixes
Based on PR 46030 by YAMASHIRO, Jun.
(enable commented out wnn4 option additionally)
Overview of changes from 1.7.2 to 1.7.3
=======================================
* Fixes
- Qt4 bridge
* Fix im-delete-text in QTextEdit
* Build fix for NetBSD ([uim-ja 322])
- uim-m17nlib
* Don't discard key release event when input mode is off (#45734)
* Fix crashes with some IMs (regressed in 1.7.0-alpha).
* Others
- Generic build fix for GNU/Hurd ([uim-ja 310])
Overview of changes from 1.7.1 to 1.7.2
=======================================
* Fixes
- libuim
* Fix crash in im-acquire-text with selection
- uim-pref-qt4
* Fix build failure when --as-needed is used (#40385, Julian Taylor)
* Enhancements
- Qt4 bridge
* Support text acquisition API with QTextEdit
* Others
- Plug very minor memory leak in uim-xim
- Plug very minor file descriptor leak in libuim-custom
Overview of changes from 1.7.0 to 1.7.1
=======================================
* Fixes
- uim-tutcode
* Fix behavior when on-key and off-key are different
- uim-dict-gtk
- uim-dict-gtk3
* Fix compilation with Canna >= 3.7
- fileio.scm
* Fix for files with no newline at the end
* Enhancements
- uim-byeoru
* Speedup in conversion by change of system dictionary format
* Introduce a new way to define a personal dictionary
- uim-tutcode
* Show stroke help temporarily by keys in kanji combination guide
* Support bushu composition which consists of three or more bushu
on auto help
* Use bushu.help file on auto help and bushu conversion
- uim-toolbar-gtk3-systray
- uim-toolbar-gtk-systray
* gnome-shell support
- uim-notify
* Support dynamic reconfiguration of the agent selection
* Others
- Make uim functional when --disable-notify is set at configure
The mastersite resolves to a parking domain, causing the html page to be
interpreted as a tarball download which doesn't pass the checksum of course.
Luckily this tarball is backed up at NetBSD, so remove the mastersite so it
can fall back to NetBSD.
Secondly, a patch from 2007 was mispelled and was never included into the
distinfo. Obviously it built anyway, but the patch was renamed and the
distinfo file regenerated.
[Changes]
- Fixed Bug 726335 (Google issue 1281)- [abrt] ibus-chewing-1.3.9.2-3.fc15: g_atomic_int_get:
Process /usr/libexec/ibus-engine-chewing was killed by signal 11 (SIGSEGV) using patch from Scott Tsai
- Fixed Bug 727018 - ibus compose file needs a symbol tag for gnome-shell
- Fixed characters duplication problem (Google issue 1245, GitHub ibus-chewing issue 2)
- Fixed KP-Enter not been able to commit preedit buffer. (Google issue 1303, GitHub ibus-chewing issue 4)
- Depends on cmake-fedora now.
- Fixed issue 1274, which is addressed by yusake's comments on d9009bf.
- Add compile flag for GNOME3.
- Add command line option: showFlags
- ibus-gnome special symbol
- Thanks Fred Chien's patch against that candidate window cannot be closed
with escape key since selected tone.
- Thanks Fourdollar's patch for Fix plain zhuyin with space as selection problem.
- Remove support for ibus-1.2 and prior.
release 1.3.2
* Check whether hic is transliteration method or not
* Ignore key event with only specific modifiers
* Use GETTEXT_PACKAGE in ibus-setup-hangul
* issue: #1210 hangul_keyboard_list should be moved
approved by minskim@ (and thanks for English translation from a Korean NEWS).
2.0.0
* imhangul-2.0 is for Gtk+-2.0
* #315984 A backspace doesn't delete a character in Firefox 4.0
* #315980 A keyboard name is not translated because of a discrepency
between imhangul and libhangul
* #306326 Add a configuration option to designate a key to switch
between Korean and English
- Set a string recognized by gtk_accelerator_parse() in imhangul.conf
* Fix a bug in the simple input method in Gtk
- Show the "preedit string" when invoked with Ctrl+Shift+u
0.9.16
* Support 2-Set old Korean, 3-Set old Korean, and Ahnmatae keyboards
* Allow Korean input with a non-qwerty US system keyboard
0.9.15
* Support a Roman keyboard (requires libhangul>=0.0.10)
* Bug fix: #305211 A strange behavior of a backspace in a gtkhtml window
approved by miskim@ (and thanks for English transration from a Korean NEWS).
nabi-0.99.9
* Bug fix: #315977 A backspace doesn't delete a character in firefox4
* Bug fix: #315744 In the Korean mode, a space is appended at the
end in Wine 1.2
* In using GtkStatusIcon
- Fix the problem of the white background issue in the tray icon
in Ubuntu 10.10
nabi-0.99.8
* Add an Ubuntu theme
* Fix bugs in the Colemak keyboard (#315056, #315058)
* Update the list of dynamic keyboard (requires libhangul>=0.0.12)
- Support the 2-Set old Korean and Ahnmatae keyboards
* Add an option to use a customized fontset in the over-the-spot mode
* Override the system keymap in default
(Handle the issue when a non-qwerty keyboard is selected)
* Add an option to sepcify a theme at build time
* Refactor code
nabi-0.99.7
* Implement a Chinese-character-only mode
* Allow Korean input even when the locales of an application and
nabi differ
* Enable "preedit string" drawing without a fontset in X Windows
* Support the Colemak keyboard
* Bug fixes
nabi-0.99.6
* Bug fix: #305368 nabi doesn't appear in the notification area
- Fix a problem that nabi does not appear automatically in the
notification area in recent distributions
nabi-0.99.5
* Support a Roman keyboard (requires libhangul>=0.0.10)
* Support pango 1.16 and below (to support CentOS 5.3)
* Bug fix: #305339 a bug resetting the input status configuration
nabi-0.99.4
* Add an "Revert to Default" button in the configuration window
* Add an option to start in the Korean mode
* Allow to turn off the tray icon
* Fix a freeze when an Enter is typed in the Korean mode in gvim
* Implement "string conversion" (not yet supported in Xlib)
* Refactoring and bug fixes
What's New in SCIM-chewing 0.3.4
---------------------------------------------------------
* Handle 0-9, "+-*/" keys, Enter and '.' in keypad.
* Handle F1-F18 keys properly.
* Fix the unexpected behavior on candidate phrases in Google
search bar browsed by Firefox.
* Be aware of the "Keyboard Layout" setting in SCIM, Thus, users
can use a French AZERTY or German QWERTZ without having to
switch to a US layout beforehand.
What's New in libchewing 0.3.3
---------------------------------------------------------
* Fix memory access violation.
* Improved Python binding.
* Merged with libchewing-data project. (r455)
* Improved random key stroke tester.
* Fix the handling of phonetic symbols in Hsu's keyboard.
* Fix unexpected candidate disorder when doing symbol choice.
* Revised phrase choice from rearward logic.
* Fix cross compilation.
* Improved shell script to merge changes form libchewing-data.
* Successful stories:
. Google ChromeOS provides built-in Chewing IM support
. MeeGo platform provides built-in Chewing IM support
Update to 1.3.9.
Fix XML entity reference.
Update to 1.3.8.
Update virtkbd module to check engine's typing mode.
Don't connect to eekboard-server if enable_virtual_keyboard is false.
Support virtual keyboard.
Define hotkeys and icon-symbol in description.
Don't start auto conversion if there is no preceding input (Issue#36).
Don't handle ctrl+g in non-converting state (Issue#35).
Ignore errors when locale module doesn't provide bindtextdomain (Issue#34).
Version 0.1.2 (2011-06-27)
=============
- Update gettext support using autopoint.
- Fix compiling error when libthai is installed in non-standard location.
- Improve linker flags handling.
The new features of scim-input-pad are the the next tables who are included:
korea.pad includes from Korean the letters, Hangul Jamo, Hangul
Compatibility Jamo, Halfwith/fullwith
arabian.pad includes the Arabian script
armenian.pad includes the Armenian script
georgian.pad includes the Georgian script from Georgia
russian.pad includes the Russian script
hebrew.pad includes the Hebrew script from Israel
india_bengali.pad includes the Bengali script from India
india_chiki.pad includes the Ol Chiki script from India
india_devanagari.pad includes the Devanagari script from India
india_gujarati.pad includes the Gujarati script from India
india_gurmukhi.pad includes the Gurmukhi script from India
india_kannada.pad includes the Kannada script from India
india_lepcha.pad includes the Lepcha script from India
india_limbu.pad includes the Limbu script from India
india_malayalam.pad includes the Malayalam script from India
india_oriya.pad includes the Oriya script from India
india_saurashtra.pad includes the Saurashtra script from India
india_sinhala.pad includes the Sinhala script from India
india_syloti_nagri.pad includes the Syloti Nagri script from India
india_tamil.pad includes the Tamil script from India
india_telugu includes the Telugu script from India
balinese.pad includes the Balinese script from Indonesia
buginese.pad includes the Buginese script from Sulawesi in Indonesia
cham.pad includes the Cham script from Vietnam and Cambodia
kayah.pad includes the Kayah Li script from the Karen in Myanmar
(Birma) and in Thailand
khmer.pad includes the Khmer script from Cambodia
mongolian.pad includes the Mongolian script from Mongolia
lao.pad includes the Lao script from Laos
myanmar.pad includes the Myamar script from Birma
philipine_buhid.pad includes the Buhid script from the Philipines
philipine_hanunoo.pad includes the Hanunoo script from the Philipines
philipine_tagalog.pad includes the Tagalog script from the Philipines
tai_lue.pad includes the old Thai script from China, Thailand,Birma
and Vietnam
rejang.pad includes the Rejang script from Sumatra in Indonesia
sundanese.pad includes the Sundanese script from west Java in Indonesia
tai_le.pad includes the new Thai script from China, Thailand,Birma and Vietnam
thai.pad includes the Thai script from Thailand
tibet.pad includes the Tibet script from Tibet
ethiopic.pad includes the ethiopian script from ethiopia
* To compile with ja-freewnn-lib-1.11alpha22, small patches
added for mainly with adding include directive:
---
+#include <unistd.h>
+#include <stdlib.h>
+#include <stdio.h>
+#include <string.h>
----
the patches are listed below.
patch-cmd_kinput2.c
patch-lib_Canna.c
patch-lib_ConvCtrl.c
patch-lib_KIProto.c
patch-lib_XimpProto.c
patch-lib_ctext.c
patch-lib_wnnlib.c
See ChangeLog, ChangeLog.en
(too many lines to be listed here)
Please note minimum security fixes had been updated after
1.10 was committed, so not really whole thing between 1.10
and 1.1.1-a021
- And more, a021 to a022 (CVS version) equivalent patches
- Thus
PKGNAME= ja-FreeWnn-{lib,server}-1.11alpha22
- DESTDIR support (That's why a022 patches added)
- package structure adjusted
from (-lib -dict -server -server-bin) to (-lib -server)
for minimum modification from DISTFILE to pkgsrc.
- some common VARIABLES have been moved to ja-freewnn-lib/Makefile.common
ja-freewnn-dict/{DESCR,Makefile,PLIST}
ja-freewnn-server-bin/{DESCR,Makefile}
Removed
ja-freewnn-lib/Makefile.common
More variables moved in
ja-freewnn-lib/patches
70 files added for a021 -> a022 updates
ja-freewnn-server/PLIST
- some executable moved sbin to bin
by following line in Makefile
+CONFIGURE_ARGS+= --disable-traditional-layout
- 23 of dictionary files in share/wnn/ja_JP/dic
moved from ja-freewnn-dict/PLIST to ja-freewnn-server/PLIST
Thanks obache@ for correcting/proofreading for above Changes.
The new features of scim-tables are the the next tables who are included:
Kirshenbaum input method for emacs
polytonic Greek added
Hebrew added
Hindi-remington added
Hindi-inscript improved
Marathi-remington added
Punjabi-remington added
Russian Yawerty improved
Tamil-tamil99 added
Uyghu-standard added
Uyghu-romanized added
Chinese Cangjie 6 added
Chinese ZhuYin improved
Chinese ZhuYin-Big improved
Chinese Cantonese improved
Based on PR#44989 by YAMASHIRO, Jun. add a missing PLIST entry.
Overview of changes from 1.7.0-alpha to 1.7.0
=============================================
* Fixes
- Generic IM
* Fix sort order of Wubi86 and Zhengma tables in 1.7.0-alpha
- GTK+3 bridge
* Avoid repeated changes of subwindow size in some annotation text
Overview of changes from 1.6.x to 1.7.0-alpha
=============================================
* New features
- GTK+3 bridge
- uim-candwin-gtk3
- uim-candwin-tbl-gtk3
- uim-dict-gtk3
- uim-im-switcher-gtk3
- uim-input-pad-ja-gtk3
- uim-pref-gtk3
- uim-toolbar-gtk3
- uim-toolbar-gtk3-systray
* Support GTK+3
- uim-toolbar-applet-gnome3
* Support GNOME3
* Fixes
- uim Scheme library
* Fix im-acquire-text brokeness
- uim-toolbar-qt4
* Fix bug: Can't set starting position of Qt4 toolbar (#34034)
- uim-notify
* Fix compilation with libnotify >= 0.7.0
- uim-tutcode
* Fix behavior with interactive bushu conversion
- uim-skk
* Fix crash upon an illegal line in SKK dictionary
- Emacs bridge
* Don't disturb other leim's modelines (#35318)
- Qt4 bridge
* Fix crash of applications upon exit when widget style of
QGtkStyle is used
- uim-dict-gtk
* Fix a minor bug
* Enhancements
- uim-tutcode
* Postfix mazegaki conversion
* Postfix bushu conversion
* Add inflection support for postfix and prefix mazegaki conversion
* Kanji code (JIS/Kuten/Unicode) input mode
* History input mode
- libuim
* Support locale with EUC-JIS-2004 and EUC-JISX0213 encodings
- generic-predict
* Add support for completion with Google Suggest
* Others
- GTK+2 bridge
* Discard support for version below 2.14.
patch provided by YAMASHIRO, Jun via PR#44861.
pkgsrc changelog:
* remove "canna" option from default.
Overview of changes from 1.6.0 to 1.6.1
=======================================
* Fixes
- IM management system
* Fix crash while switching to the IM which has been once used but
then disabled and re-enabled again by "Enabled input methods"
widget of uim-pref.
- uim-skk
* Don't crash with malformed words having brackets which are not
intended for skk-henkan-strict-okuri-precedence.
- libuim-skk.so
* Fix possible contamination of dictionary files of SKK-JISYO and
tutcode's mazegaki dictionary when user switches between uim-skk
and uim-tutcode dynamically.
* Enhancements
- uim-tutcode
* Interactive bushu conversion
* Postfix bushu conversion in yomi of mazegaki.
* Completion
* Input prediction in mazegaki conversion
* Input prediction in bushu conversion
* Kanji combination guide on completion and input prediction
* Another symbol (kigou) input mode like "KI"
* Add function to redisplay last auto help
* Show kanji on auto help
* Show hint kanji for intermediate sequence on virtual keyboard
- uim-pref
* Add configuration 'uim-pref-suppress-dot-uim-warning-dialog?'
(Set this variable #t in ~/.uim if you'd like to suppress the
dialog)
- uim.el
* Support Emacs24 (by Kan-Ru Chen, #31539)
* Others
- Enable compilation even when the home directory of chrooted build
environment is non-writable (e.g. nobody user)
- tcode, trycode, and hangul modules are no longer enabled by
default. Use more sophisticated tutcode and byeoru modules
instead.
* Release 1.3.2
* Add --with-gtk=2.0|3.0|no configure option.
* Fix build with ibus-1.4.
* Use minput_list() instead of mdatabase_list().
* Make class name parsing robuster.
* Add "~/.m17n.d/" to observed-paths.
* Use wildcard pattern in default.xml to set the default value.
* Lower the rank of si-samanala engine.
* Read default config from a file.
* Use gnome-autogen.sh to generate configure and Makefiles.
* Don't compile m17nutil.c twice for engine and setup.
* Use ibus_m17n_scan_engine_name() to simplify the code.
* Assign separate GType to each m17n engine to make config per-class.
* Clear errno to 0 before calling strtol/strtoul.
* Simplify AltGr handling.
* Give higher rank for all Indic engines.
* Use MInputContext::arg and remove the table for ic and engine.
* Don't init IBusBus in print_engines_xml().
* Fix bug in ibus_m17n_engine_reset.
* Add code coderview.settings.
ChangeLog:
Do not normalize keyevent on romaja keyboard
Dynamic hangul keyboard listing
Normalize keyval for position before use
Ignore modifier keys only when they are in hanja key list
Clear preedit text on ibus_hangul_engine_reset()
Update lookup table managing code
Add a new feature
commit the candidate when it clicked by the user.
issue #921
0.0.12
* add dubeolsik yetgeul keyboard
* make sebeolsik yetgeul keyboard layout conform to HWP
* add API to get builtin keyboard list
* use Unicode jamo for incomplete syllable
* gettextize
* bug fix for Unicode 5.2
* fix wrong characters in ahnmatae(#31548)
* add command line tool hangul
0.0.11
* fixed preedit string bug: #315256
* added Ahnmatae method
* updated hanja frequency data and reordered hanja table according to it:
#315200
* fixed bugs in hanja data
* updated old hangul jamo combination table according to Unicode standard 5.2:
#315127
Random summary of changes:
14.2
- Support multiple dictionaries.
- Support "tooltip".
- Support UNICODE display.
- Support face colors.
- Support character input by code number.
- Improve JIX X 0213 support.
14.1
- Support CDB directly.
- Improve info.
- Improve dynamic completion.
- Display character code.
- Various conversion improvements.
2010-11-03 IRIE Shinsuke <irieshinsuke@yahoo.co.jp>
* Version0.2.1
* Added support for vim-mode
* Added libexec directories to search paths for `ibus-agent-file-name'
* Changed to execute fallback command in latin mode to simulate original
behavior for the key event (IBus issue #1058)
* Modified custom types of `ibus-kana-onbiki-x-keysym' and
`ibus-kana-onbiki-key-symbol' so that they can be disabled
* Modified to reset cursor color in consideration of color themes
* Modified not to unnecessarily set cursor color for non-X frame
* Modified not to unnecessarily check current buffer while command is running
* Fixed bug: Keymap might not be updated when switching input focus from
non-X frame to X frame
* Fixed bug: ibus-mode is abnormally exited when automatically turning off
monitor for power saving
* Fixed bug: If starting emacs-w3m in one of multiple frames, ibus-mode's
keymaps are incorrectly disabled in one of the other buffers
* Fixed problem (maybe): ibus-el-agent crashes if failing to connect to
X display when ibus-mode starts or display is changed (RH bug #627358)
- New tables which was not include in original:
+ array30, array30-big, wubi-haifeng
- Table removed from original version because of license issues:
+ zhengma, ziranma
- Add package review ID
- Add build tag for el6
* Fri Nov 19 2010 Ding-Yi Chen <dchen at redhat.com> - 1.3.9.2
- Fixed Bug 652909
- Added back a Changlog item that mention Nils Philippsen's change.
- Apply Jim Huang's patch for zh_TW.po
* Fri Nov 12 2010 Ding-Yi Chen <dchen at redhat.com> - 1.3.8
- Quick fix for f15 and ibus-1.4 build
- Version scheme change.
Fixed deleting voiced consonants. (#1148)
Fixed romaji_typing_rule and kana_typing_rule receives UTF-8 keys. (#1146)
Fixed the custom tables with new keys.
Fixed SEGV when key tables are customized in new gconf.
Added a warning for ibus.
Supported ibus.engine.update_preedit_text_with_mode
Changes:
* Fix compatible issues with gnome-do and chromium-browser.
* Fix XIM compatible issue in ppc64 platform.
* Fix some build problems with gtk3
* Sort input method in setup with strcoll
* Fix some problem in Makefile for NteBSD
ChangeLog
Comment out test-keymap test case.
Limit typing rate to 800 hits/minutes, and fix make distcheck error.
Catch exceptions from os.kill.
Fix CPU 100% usage with signal.SIGCHLD
Fix keyval and state in __keycode_button_clicked_cb
Destroy existing connection before creating a new connection, and only create the fake context once.
Move the candidate window just above the cursor when the window and a preedit string overlap.
Remove compiler warnings.
Enable key snooper by default again, except Chrome/Chromium browsers, to fix application compatibility issues like 1068.
Always consume the hotkey, even if the hotkey associated engine already activated.
Fix race condition in bus_ibus_impl_create_engine()
Fix a build error of IBus-1.0.gir with gobject-introspection 0.9.6
Fix a typo.
Remove unused dbus-1 package in gir includes
Implement stress tool for ibus.
Fix a build error with GTK3 gdkkeysyms.h
Update gu.po from Sweta Kothari <swkothar@redhat.com>
Fix g_assertion failure in bus_ibus_impl_{load,save}_global_engine_name functions.
Add some comments about registry
Add Emacs modeline for .c and .h files.
Move the log handler code from bus/main.cc to src/ibusshare.cc.
Add codereview.settings.
Add time info in log message from satorux@chromium.org
Fix memory leak in ibus_input_context_process_key_event by Fujiwarat
Fix segv in bus_engine_proxy_process_key_event_reply_cb
Set CFLAGS to "-Wall -Werror" in autogen.sh
Fix property issue.
Update fr.po
Fix a build warning
Packagesd by Ryo ONODERA in pkgsrc-wip, import requested via PR#43819.
multiskkserv is a skk server which can use multiple dictionaries.
This server uses cdb format as the dictionary format. cdb is the
efficient and constant (i.e. cannot add, delete, modify) database
format by Dan J. Bernstein, which I think is the appropriate format
for the system-wide constant dictionary storage.
Based on PR#43872 by YAMASHIRO, Jun.
some fixes and improvements by me.
uim Changelog:
- New IM uim-social-ime
- New IM uim-ajax-ime
- New IM uim-yahoo-jp
- Support Wubi86 and ZhengMa tables with Generic IM
- Improved uim-tutcode with many nice features with table style
candidate window
- Stable Qt4 bridge
- Annotation system
- Internal plugin system reorganization
- uim-elatin enhancement
- Japanese multiple-segment IM enhancement
This project provide a table that defines Telex/VNI Input Method phrases for
Vietnamese. Most of Vietnamese in North Vietnam use Telex Typing, but VNI
typing in South Vietnam.
[Changes]
- Input style of ibus-chewing is decommissioned under ibus >=1.3.
Now the input style is determined solely on the setting of
"Embed preedit in application window" in IBus.
- Resolves: #608991, #632043
- Fixed Issue 1022: chewing commit some text in reset method
(patched by Peng Huang).
- Fixed Issue 1032: [ibus-chewing] Chewing not commit some single Chinese
char into application when press enter.
- Rewrite CMake modules to make them cleaner, and documents in cmake module
help format.
- [For developer and distro maintainer]
Various targets changed. Use 'make help' to obtain a list of available
targets.
Packaged by Ryo ONODERA in pkgsrc-wip.
A part of import request by PR#43818.
dbskkd-cdb is a dictionary server for SKK Japanese Input Method
system. For building the executable file, TinyCDB is required.
Packaged by Ryo ONODERA in pkgsrc-wip.
A part of import request by PR#43818.
A basic set of Kana-Kanji conversion dictionary developed by SKK project.
This package is cdb format dictionary for dbskkd-cdb and so on.
* Added some acronyms.
* Fixed the default selection in GtkTreeView.
* Don't load kbdui/foo.so in addition to kbdui/foo.la.
* Added debian build.
* Fixed#1 input-pad.pc doesn't include gtk+-2.0
Fix make distcheck errors.
Tiny modification to m4/boost.m4.
Pass $CXXFLAGS to $ac_cpp to make cross-compilation of ibus-pinyin easy.
Remove sqlite3 in create_db.py
Remove duplicated EXTRA_DIST value in lua/Makefile.am
Add PY prefix of all file names to avoid name conflict on some file
system than is case insensitive.
change the default fuzzy settings
* fix 'hh' not handled correctly.
* Don't use "==" in sh script (Issue#2).
* Don't set $ENV in Makefile (Issue#3).
* Add "Initial Input Mode" configuration.
* Add test for zl.
* Fix zl conversion.
* Add links to external resources.
to trigger/signal a rebuild for the transition 5.10.1 -> 5.12.1.
The list of packages is computed by finding all packages which end
up having either of PERL5_USE_PACKLIST, BUILDLINK_API_DEPENDS.perl,
or PERL5_PACKLIST defined in their make setup (tested via
"make show-vars VARNAMES=..."), minus the packages updated after
the perl package update.
sno@ was right after all, obache@ kindly asked and he@ led the
way. Thanks!
* Fix build error with binutils-gold
* Add debian package support from Li Daoing's repository
* Fix build error with cmake-2.6
* Use if(NOT ..) to replace if(NOT DEFINED ..) in CMakeLists.txt
* Use memset to init memory block.
* Merge remote branch 'origin/debian'
* Fix background and foreground in preedit QTextCharFormat.
* Added hotkeys in IBus::EngineDesc .
ibus.el is a IBus client for GNU Emacs. This program allows users on-the-spot
style input with IBus. The input statuses are individually kept for each
buffer, and prefix-keys such as C-x and C-c can be used even if IBus is active.
So you can input various languages fast and comfortably by using it.
This package provide Chinese extra phrases for ibus-table based IME,
such as ibus-table-zhengma, ibus-table-wubi, ibus-table-cangjie5,
ibus-table-erbi and etc.
IBus-chewing is an IBus front-end of Chewing, an intelligent Chinese input
method for Zhuyin (BoPoMoFo) users.
It supports various Zhuyin keyboard layout, such as standard (DaChen),
IBM, Gin-Yeah, Eten, Eten 26, Hsu, Dvorak, Dvorak-Hsu, and DaChen26.
Chewing also support toned Hanyu pinyin input.
ibus-array is desined to implement a full-functional Array 30 input method
engine for iBus. Array 30 is a traditional Chinese input method used in Taiwan.
IBus means an Intelligent Input Bus. It is a new input framework for Linux OS.
It provides full featured and user friendly input method user interface.
It also may help developers to develop input method easily.
Based on PR#43649 by Ryo ONODERA.
Back from snapshot release, so version scheme also back from timestamp.
This release is still alpha release, but, yes, better than snapshot,
and need to support of emacs23.
While here,
* set LICENSE=gnu-gpl-v2
* simplify PLIST.
* fixes paths for tutorial file in setup file
* change to use MAKE_FLAGS instead of MAKEFLAGS, it affect subsequent
make, e.g. failed to build of emacs for resolving dependency.
Based on packaged by uebayasi@ in pkgsrc-wip-jp,
Modified by Ryo ONODERA in pkgsrc-wip.
A basic set of Kana-Kanji conversion dictionary developed by SKK project.
nbitools is not enough to maintain for recent platforms and no good support for
user-destdir (now, canna-lib and canna-canuum is user-destdir ready)
no objection in tech-pkg-ja@ about three months.
Zinnia is a simple, customizable and portable online hand recognition system
based on Support Vector Machines.
This package contains Zonna recognition models trained with Tomoe data.
Zinnia is a simple, customizable and portable online hand recognition system
based on Support Vector Machines.
This package is python module for Zinnia.
Zinnia is a simple, customizable and portable online hand recognition system
based on Support Vector Machines. Zinnia simply receives user pen strokes as a
sequence of coordinate data and outputs n-best characters sorted by SVM
confidence. To keep portability, Zinnia doesn't have any rendering
functionality. In addition to recognition, Zinnia provides training module that
allows us to create any hand-written recognition systems with low-cost.
im-ja is a Japanese input module for GTK2 and also has a XIM server.
Currently supports hiragana, katakana, half-width katakana, zenkaku, canna, wnn,
anthy, skk and kanji character recognition inputs.
Conversion hotkeys, status window, preedit text colors, etc. can be customized
through a GUI. An optional applet is also included for the gnome-panel which
can be used to display and change the input method.
Kimera is an input method (XIM) for the X Window System utilizing
the Qt4 toolkit that provides Japanese language input support. It
provides Kanji conversion via Canna or Anthy, as well as supporting
the ability to assign input styles. We are also considering a
simple key assign selection method like the one used in Windows.
2009-07-29 Takuro Ashie <ashie@clear-code.com>
* preparation for release (ver 2.5)
2009-07-21 Takuro Ashie <ashie@clear-code.com>
* fix:
- Fix paddings in the KasumiAddWindow.
2009-07-16 Takuro Ashie <ashie@clear-code.com>
* fix:
- Use stock objects for buttons on KasumiAddWindow.
- Remove neeldless key accelerators for KasumiAddWindow.
2009-07-16 Takuro Ashie <ashie@clear-code.com>
* fix:
- Fix paddings in the main window.
- Fix minimum size of columns in the word list view.
- Remove needless key accelerators in main window.
Use mnemonic instead.
- Remove some needless codes.
2009-07-13 Takuro Ashie <ashie@clear-code.com>
* fix:
- Fix a compile time warning.
- Rename some strang variable names.
2009-07-13 Takuro Ashie <ashie@clear-code.com>
* enhancement:
- Added some word type subcategories. Thanks to Jun Oizumi.
2009-06-25 Takuro Ashie <ashie@clear-code.com>
* fix
- Kill duplicated Ikuya Awashiro in man page! Thanks to Jun Oizumi.
2009-06-25 Takuro Ashie <ashie@clear-code.com>
* fix
- Remove "SIKU" conjugation. Thanks to Jun Oizumi.
2009-01-29 Takuro Ashie <ashie@clear-code.com>
* preparation for release (ver 2.4)
2009-01-20 Ikuya Awashiro <ikuya@fruitsbasket.info>
* KasumiAddWindow.cxx:
- addded an include file. Thanks to Mike Fabian.
* KasumiWord.cxx:
- DITTO.
* main.cxx:
- DITTO.
2009-01-20 Takuro Ashie <ashie@clear-code.com>
* enhancement:
- Support ANTHY_UTF8_ENCODING and set it as default.
You can switch to ANTHY_EUC_JP_ENCODING by -E or --eucjp option.
2007-10-30 Ikuya Awashiro <ikuya@fruitsbasket.info>
* fix:
- FTBFS with GCC 4.3: missing #includes
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=417266
Thaks to Martin Michlmayr <tbm@cyrius.com>.
- kasumi dumps core on x86_64 systems when trying
to add new words to the dictionary
https://bugzilla.novell.com/show_bug.cgi?id=253788
Thanks to Mike Fabian.
- fail in the judgment of the version.
http://kasumi.sourceforge.jp/index.php?%A5%D0%A5%B0%CA%F3%B9%F0%2F95
Thanks to dtana.
* preparation for release (ver 2.3)
2006-11-29 Takashi Nakamoto <bluedwarf@bpost.plala.or.jp>
* preparation for release (ver 2.2)
2006-11-28 Takashi Nakamoto <bluedwarf@bpost.plala.or.jp>
* enhancement:
- change the text color on the search entry box if Kasumi failed
to find specified text
- ask Gtk+ library to keep above the exclusive mode window
* fix:
- scroll to the undermost of the view after adding a new word
2006-11-27 Takashi Nakamoto <bluedwarf@bpost.plala.or.jp>
* fix:
- the bug raised while sorting
2006-10-15 Takashi Nakamoto <bluedwarf@bpost.plala.or.jp>
* enhancement:
- add exclusive mode
2006-10-14 Ikuya Awashiro <ikuya@fruitsbasket.info>
* configure.in: 2.1
* This release is a result at the CodeFest Kyoto 2006
hosted by FSIJ.
2006-10-14 Takashi Nakamoto <bluedwarf@bpost.plala.or.jp>
* enhancement:
- add new word type category combo box
* fix:
- export embedded string to po files
- changed adding window title
- remove unnecessary words
- improve portability
- support dakuon mark
2006-06-28 Takashi Nakamoto <bluedwarf@bpost.plala.or.jp>
* fix: AUTHORS
- reconstruction of AUTHORS file
- added Italian l10n team
2006-06-26 Takashi Nakamoto <bluedwarf@bpost.plala.or.jp>
* fix: KasumiException.cxx
- security fix
* fix: KasumiDicEventListener.hxx KasumiWordEventListener.hxx
- add virtual destructor to avoid compilation warnings
* fix: configure.in main.cxx Makefile.am acconfig.h
- removed acconfig.h deprecated in autoconf 2.5
- correctly handling "--disable-nls" option on no-glibc platform
- thanks to Diego's patch
* fix: AUTHORS
- typo of my email address :(
2006-06-25 Takashi Nakamoto <bluedwarf@bpost.plala.or.jp>
* fix: KasumiAddWindow.cxx po/*.po po/*.pot
- fixed typo: "Invlid" -> "Invalid"
* New feature:
- Italian localization
2006-06-23 Ikuya Awashiro <ikuya@fruitsbasket.info>
* KasumiMainWindow.hxx:
- fixed compile error on gcc 4.1. thanks to sgtom.
2006-06-21 Takashi Nakamoto <bluedwarf@bpost.plala.or.jp>
* fix: configure.in
- require Gtk+ version is 2.6 or higher for GtkCellRendererCombo
2006-06-20 Takashi Nakamoto <bluedwarf@bpost.plala.or.jp>
* fix: KasumiDic.cxx
- fix the defect that Kasumi failes when the private dictionary
doesn't have any word.
2006-06-03 Ikuya Awashiro <ikuya@fruitsbasket.info>
* kasumi.desktop.in:
- added.
* Makefile.am, configure.in:
- resistered in GNOME/K menu.
2006-06-02 Ikuya Awashiro <ikuya@fruitsbasket.info>
* Makefile.am:
- kasumi.png was placed in /usr/share/pixmaps instead /usr/share/kasumi.
- add man page.
* kasumi.1:
- added.
2006-05-29 Takashi Nakamoto <bluedwarf@bpost.plala.or.jp>
* preparation for release (ver 2.0)
* fix: KasumiWord.cxx, configure.in
- accept ASCII character in sound entries
* fix: AUTHORS
- add Ikuya to AUTHORS
* fix: POTFILES.in ja.po
- translated newly created message
* enhancement: KasumiMainWindow.cxx KasumiMainWindow.hxx
- obsolete "Add Mode" button
2006-05-28 Takashi Nakamoto <bluedwarf@bpost.plala.or.jp>
* fix: main.cxx
- changed copyright notice
* fix: KasumiWordType.hxx, KasumiWord.cxx
- don't issue modification event when the same value are set
2006-05-27 Takashi Nakamoto <bluedwarf@bpost.plala.or.jp>
* fix: KasumiDic.cxx, KasumiWordType.cxx
- set an alternative word type if specified one does not exist
- measures against an anthy's defect that it returns error code
even when it is right and no word is registered.
* fix:
- inserted copyright and license terms into the head of each
source file or header file.
* fix: KasumiAddWindow.cxx
- changed the page size of the spin button
2006-05-17 Takashi Nakamoto <bluedwarf@bpost.plala.or.jp>
* fix: KasumiDic.cxx
- change the condition of anthy version
2006-05-14 Takashi Nakamoto <bluedwarf@bpost.plala.or.jp>
* fix
- KasumiAddWindow
- search function for KasumiMainWindow
* enhancement
- implement of KasumiWordEventListener
2006-05-7 Takashi Nakamoto <bluedwarf@bpost.plala.or.jp>
* drastic changes for new version 2.0
2006-03-26 Ikuya Awashiro <ikuya@fruitsbasket.info>
* KasumiMainWindow.hxx, KasumiAddWindow.hxx:
Fixed with G++ 4.1: friends not declared
Thanks to Martin Michlmayr <tbm@cyrius.com>
Major changes from version 3.0 to 3.1:
* Kinput2 now supports ATOK. The source code is kindly contributed
by Justsystem Corporation. Thanks!
To compile for Atok, define symbol `UseAtok' in Kinput2.conf.
* Minimum support of Wnn7 is implemented.
* Two IC attributes, XNPreeditState and XNResetState both of which
are introduced by X11R6, are supported.
* Kinput2's `egg'-emulation is improved.
* A workaround for bugs in some Xlib implementations has been added.
* Previously, when kinput2 cannot connect to jserver, kinput2 seems
to be frozen (actually it keeps retrying). Now kinput2 beeps and
ends conversion in such a case. The next time you start conversion,
it tries to connect again.
Re-arranged packages:
* sj3-lib:removed
* sj3-dict: removed
* sj3-server-bin: removed
* sj3-client: added (lib + client tools)
* sj3-server: no more meta-pkg (server + dict)
* sj3: added (sj3-server + sj3-client)
Notable changes since 2.0.1.21:
* changed to gnu-configure.
* sj3lib is shared library.
* using lua for server settings.
* sj3 command was removed.
* sj3serv no longer listen network socket. Set "false" to "strict_auth" and
invoke sj3proxy additionaly if require it.
* dictionaty directry was changed, please refer MigrationGuide from HOMEPAGE.
On pkgsrc, new location is ${VARBASE}/sj3/dict.
pkgsrc changelog:
* Enable Anthy's UTF-8 dictionary support.
* Improved configure option for 'eb' option.
* Add qt4/kde4 option
(Note: UIM widget for KDE4("kde4" option) still buggy)
Overview of changes from 1.5.6 to 1.5.7
=======================================
* New features
- uim-chardict-qt4
- uim-toolbar-qt4
- uim-pref-qt4
- uim-candwin-qt4
- uim-applet-kde4
* Experimental support of Qt4 tools by Muneyuki Noguchi (#24123,
#24620)
* Enhancement
- uim-toolbar-gtk-systray
* Support transparent background
* Fixes
- GTK+ bridge
* Fix crash in combination with Firefox and flashplayer plugin
(#25139)
- uim-fep
* Fix compilation on NetBSD current ([uim-ja 148])
- uim.el
* Fix bug: uim.el doesn't work when linum-mode is activated and
the inline candidates displaying mode is enabled ([uim-en 245])
* Fix bug: Some special keys (such as arrow keys) cannot be used
when uim.el is enabled on emacs23 with -nw option.
- Qt4 bridge
* Don't enable knotify3 when qt4-immodule is enabled in configure
time (#19877)
* Fix compilation when srcdir != builddir
- uim-notify
* Disable uim-notify on build phase to prevent dependency library
mismatch
- uim-chardict-qt
* Install translation file properly ([uim-ja 150])
Overview of changes from 1.5.5 to 1.5.6
=======================================
* Enhancement
- uim-dict-gtk
* Support UTF-8 characters with Anthy personal dictionary
* Fixes
- XIM bridge (uim-xim)
* Fix segfault with latest X.Org
* Fix segfault with PRIME
* Others
- Apply build fix from Fedora
Based on patch provided by Ryo ONODERA in PR 42416 (include user-destdir support).
While here set LICENSE=gnu-gpl-v2.
* 2.06.3 -> 2.06.4
- close IM by Shift-Space even if NumLock or CapsLock.
- fix skkinput.ja.man install path
A HMM-based(Hidden Markov Model) sentence pinyin input method for Chinese.
This Project starts from my graduate life in BUPT. Evolved from a research
prototype to a real project. This input method is based on scim-pinyin, a
popular Linux input method.
Hopefully this project will help you speed up inputing Chinese characters.
This is SCIM IMEngine module for Thai, based on libthai library.
Currently, it supports 3 keyboard layouts:
- Ketmanee
- TIS-820.2538
- Pattachote
In addition to keyboard layout suuports, it also validate input sequences
in 3 levels of strictness:
- Passthrough (no check)
- Basic Check
- Srict Check
For applications that support surrounding text retrieval/deleting,
it also corrects the invalid input sequences.
This package includes many data files for SCIM Generic Table Input Method
module (table). They support Chinese (Simplified and Traditional), Japanese,
Korean and some other languages, such as Russian etc. The data files will be
converted to binary format and installed to ${prefix}/share/scim/tables/.
SCIM table im module will load them automatically.
