to $PREFIX/share/AbiSuite
So we should change path-setting in chinese/abiword and chinese/ttfm
PR: 30600
Submitted by: Kuang-che Wu <kcwu@ck.tp.edu.tw>
Approved by: MAINTAINER
The patchset for this port simply does one thing: let 8 bit character go.
According to the submitter, the patchset is derived from twnic.net,
which allows multilingual .tw domain name registration now.
Submitted by: jihuang@gate.sinica.edu.tw
Zhcon is a fast Linux Console Chinese System which supports
framebuffer device.It can display Chinese,Japanese or Korean
double byte characters.Supported language encodings include:
GB2312,GBK,BIG5,JIS and KSCM.
It can also use input methods(table based) from Windows 98 and
UCDOS for MS-DOS.
PR: ports/30433
Submitted by: Statue <statue@freebsd.sinica.edu.tw>
a resized 16x16 bitmap font) by CMEX and DynaLab for the X Window System
(PCF). This is one of the few Big5 or Big5+ Chinese bitmap fonts on the
Internet that is DFSG-compliant.
The font was designed by DynaLab and released by CMEX as part of the
proposed Big5+ standard documentation. It was then converted to HBF and BDF
format and further modified by Wei-Lun Chao in September 1998.
Submitted by: statue@freebsd.sinica.edu.tw
Handed to me via: keichii
Enlarge buffer size for extra-long line.
This fixes core dump which is rarely happened.
PR: ports/26961
Submitted by: Yen-Ming Lee <leeym@cae.ce.ntu.edu.tw>
In this xmms release (1.2.5), only few default values should be tuned
for localization, instead of patching source code extensively.
Now I'm considering an issue: should we provide localized ports/packages
which are properly tuned, or simply tell users what variables/options
should be checked/unchecked via FreeBSD documentation project ?
It seems that tin itself now has more ability to handle MB.
More inspection on chinese/tin may be required to remove obsoleted method.
(I will forward this commit log to its MAINTAINER and
freebsd-taiwan-chinese@maillist.tw.freebsd.org)
This version fixes NTFS filenames which have both Chinese and English
characters.
Submitted by: huangant.bbs@bbs.cs.ntou.edu.tw (author of the ntfs patch)
like '\n' is definitely non-printable.
o Fix this by tin's misc.c way, exactly what ports/26402 submitted.
o Unless we have wchar and, well, Citrus Project, imported, there's
no clean way, Good Thing (tm), we can do in this case.
PR: ports/26402
Submitted by: Yen-Ming Lee <leeym@bsd.ce.ntu.edu.tw>
RUN_DEPENDS. Since we now have the correct patch in all of
x11/{XFree86,XFree86-4,XFree86-4-libraries}, its service is no
longer needed.
Reported by: cheath.bbs@bbs.cs.nthu.edu.tw
This kernel module lets users read BIG5 filenames on an NTFS. The author
is <huangant.bbs@bbs.cs.ntou.edu.tw>. The original patch is located at:
http://www.cis.nctu.edu.tw/~gis89501/bsd/ntfs/chinese_ntfs-4.3b2-ro.patch
It is slightly modified to accomodate port building.
It was tested on 4.3-BETA with Windows 2000 by the author, and 5.0-CURRENT
by me (also with 2000).
QE is a PE2-like editor program under U*nix.
p.s. Don't pursuade me putting this into editors/. If you don't have
a Chinese (big5) environment, you could only see 'funny characters.'
PR: ports/25867
Submitted by: Status <statue@freebsd.sinica.edu.tw>
after deinstallation.
This hack would be removed after pine4 author accepts patch
that will deal with l10n'd pine4.
PR: ports/25720
Submitted by: MAINTAINER
Approved by/Struggled with: MAINTAINER
This meta version of AbiWord sets up the correct environment for BIG5/GB
users(though most works are done in chinese/ttfm). Users can import a
Chinese Word97 document(just not too complex :p), edit it with an XIM
(eg. xcin25), then print it using Ghostscript.
This Big5-ized patch is modified from http://dengaku.org/yasu/dia/.
Chinese Postscript output is not available yet(it'll cause Ghostscript
crash), but users can output to .png, then convert to any type they like.
This is a slightly modified version of japanese/gnumeric. It can
exchange simple spreadsheets with MS Excel.
TODO: Test CLE's patch, to see if a static gnomeole2 is needed to
avoid frequent crash when opening complicated .xls.
Thanks for serving Chinese people for such a long time, it's time
to retire.
Reason for aterm retiring: chinese/rxvt-* and chinese/crxvt
Reason for xcin and xa+cv retiring: Now most favored X applications,
like gtk12 based or QT based even
motif based, could just work happily
with xcin25, chinput and chinput2.
For people who wants to know what's aterm, xa+cv, and xcin, here's
brief introduction for remembrance:
Aterm and xcin use their own protocol to input Chinese (big5 actually)
words. Xa+cv is a hacky wrapper for X applications to hook "xcin protocol."
"Non-I18N-aware old-fashioned application no more supported"
They will retire before 4.3 ports freeze.
Discussed/Approved by: vanilla, foxfair, kevlo, keith, ijliao
This changes the behavior of Tk on X where X Input Methods (XIM) were
recognized and used without question.
With 8.3, they will be recognized and initialized, but not used unless
XIM is turned on (tk useinputmethods 1). This should only affect users
users with special input methods, and the new default behavior should
be more beneficial to the average user.
Please repo copy, thanks.
This changes the behavior of Tk on X where X Input Methods (XIM) were
recognized and used without question.
With 8.3, they will be recognized and initialized, but not used unless
XIM is turned on (tk useinputmethods 1). This should only affect users
users with special input methods, and the new default behavior should
be more beneficial to the average user.
BTW, please repo copy, thanks.
Chinese Internet Community.
In Taiwan, there're over 300 BBS sites using Firebird, about 50 sites in china,
and about 20 worldwide (USA, UK, Canada, Austrailla and many countries can be
find Firebird) and over thousands of users on-line at the same time in biggest
BBS site.
For more information, ref: http://www.firebird.org.tw/
PR: ports/24899
Submitted by: Edward Ping-Da Chuan <edwardc@firebird.com.tw>
mysql323-client installs libmysqlclient.so.10 (instead of
libmysqlclient.so.6).
mysql323-{client,server} are production quality, now (according to
MySQL AB at least).
An Eterm that supports BIG5 and GB encodings. Thanks to
MANTANI Nobutaka <nobutaka@nobutaka.com> for the patch.
Submitted by: Michael C. Wu <keichii@iteration.net>
Eterm is a color vt102 terminal emulator for X Window system.
It supports pixmap and pseudo-transparent background. It also
supports Japanese input with XIM (X Input Method) protocol.
This version also adds BIG5 and GB support.
Thanks to MANTANI Nobutaka <nobutaka@nobutaka.com> for the patch.
Submitted by: Michael C. Wu <keichii@iteration.net>
font-lock and pc-selection-mode.
Yet another comment is
;; Don't ask me more about emacs customization ! You should read the
;; manual or just ask google.com.
Another message for phj@cn.FreeBSD.org, I know you read cvs-all:
Could you give me the URL which introduces how to setup a
zh_CN.EUC environment ? Including all details, I think LANG=zh_CN.EUC
isn't sufficent. Otherwise I can't test and make zh-emacs20 work under
both Simplified/Trditional environment.
1. Unable to open files in the "Open File..." window
2. Change default printing fonts to DefaultMing
Submitted by: The first one is from Ming-Yen <myhsu@cyberdude.com>
VFlib is a vector font library, which can use TrueType,
Zeit, JG, and BDF fonts. This version supports both
BIG5/GB encoded fonts.
The original patch was made by Chinese Turbo Linux, and
Chih-Wei Huang <cwhuang@linux.org.tw> added the BIG5 Eten
extension.
the fonts are encoded in BIG5, the characters and symbols are from
several less commonly used layers in CNS. See
http://www.edu.tw/mandr/bbs/1-4-2/1-4-2.html for more details.
Also displays messages about WANT_XX when installing.
zh-arphicttf's pattern.
Some fonts have problems with X-TT/X-freetype, possibly due to bogus
entries, including wcl-02.ttf, wcl-03.ttf, wcl-05.ttf, wcl-06.ttf, and
wcl-07.ttf. The rest are displayed correctly.
Further testings showed that all these fonts can't be used in GS6. It
seems that the patched GS6 is confused identifying Big5 and GB TTFs,
so for now only the newer Unicode-encoded TTFs can be used.
Before isprint(3) get fixed or any one knows how to use isprint(3) to
identify multibytes characters, this temporary fix would remain here.
PR: 23196
Reviewed by: People on tw.bbs.comp.386bsd newsgroup
Many of them are unnecessary, and most of the task are handled by
ttfm now.
It's also aware of chinese/{CJK, ghostscript6}. It supports XFree86
3.x or 4.x by default. Users only need to define GS6=yes or CJK=yes
when installing, then it'll do most of the work. CJK and ghostscript6
ports should specify their respective variables via DEPENDS_ARGS.
chinese/{moettf,moefonts-cid,wangttf} will have similar changes.
undefined behaviors, some of which are coherent in 0.9.2,
others like CID & TTF naming conventions are not standardized
yet.
Future versions should move to FreeType 2, since it has more
font support, especially for CID fonts, so that some really
hacks can be removed from TTFM.
This port adds Chinese support to enscript, which means it can convert
a BIG5/CNS plain text to a PS document, with all the fancy features
provided by enscript. Unlike bg5ps, it can't manage GB yet, but it's
easy to add. Another useful feature is with the ps2pdf and zh-ghostscript6,
it is now possible to generate font-embedded Chinese pdf document from
plain text on the fly, which also means platform-independent.
Due to some problems in the CMap files provided by Adobe, Eten-B5-H and
B5-H CMaps are unable to map correct ascii glyphs in TTFs. BIG5 users
are suggested to use B5pc-H for the time being.
Maybe it's time to re-organize ports/chinese/* font relationships.
This port adds Chinese support to enscript, which means it can convert
a BIG5/CNS plain text to a PS document, with all the fancy features
provided by enscript. Unlike bg5ps, it can't manage GB yet, but it's
easy to add. Another useful feature is with the ps2pdf and zh-ghostscript6,
it is now possible to generate font-embedded Chinese pdf document from
plain text on the fly, which also means platform-independent.
Due to some problems in the CMap files provided by Adobe, Eten-B5-H and
B5-H CMaps are unable to map correct ascii glyphs in TTFs. BIG5 users
are suggested to use B5pc-H for the time being.
Maybe it's time to re-organize ports/chinese/* font relationships.
This port adds Big5/GB TTF support to Ghostscript 6. Since GS6
treated TTF as translated CID fonts, users can then use them in
various applications that generate PS documents, like Netscape
or enscript.
The major drawback, compared to Ghostscript+VFlib, is that GS6
doesn't support automated font variations like bold or italic.
The font can only be used as the way it is.
This port adds Big5/GB TTF support to Ghostscript 6. Since GS6
treated TTF as translated CID fonts, users can then use them in
various applications that generate PS documents, like Netscape
or enscript.
The major drawback, compared to Ghostscript+VFlib, is that GS6
doesn't support automated font variations like bold or italic.
The font can only be used as the way it is.
This version fixes some minor typos and bugs in ttfm.sh and X modules,
and the ghostscript6.ttfm module was also added. The new module is to
be used along with the forthcoming zh-ghostscript6 and zh-enscript. It
can also be used as a CID font manager. Relative info can be found at:
http://www.aihara.co.jp/~taiji/tops/
The new GS6 module is written with Japan/Korean in mind, meaning there
are some barebones for identifying/using Japan/Korean TTF/CIDs.
TODO modules:
* Abiword
* ChiTeX
* CID for XFree86-4
* CJK
* cwTeX
* GS5.5/vflib
This port installs ten Chinese Big5-encoded ttfs, made by
Dr. Hann-Tzong Wang. They are distributed under GPL.
There should be better integrations of arphicttf/moettf/wangttf,
ttfm, and those apps using ttf.
Especially thanks to the maintainer, Tai-hwa Liang (avatar), for
ending those long lasting evil +400k files/patch-*. :-)
According to kkenn, pine4's security will exist until it's totally
rewritten, so I'll leave pkg-install untouched, which shows the
security warning.
PR: ports/22436
A True Type Font Manager that handles true type font installation,
uninstallation, list, and default font setting. It was originally
in the CLE Linux distribution, modified by Cheng Yuan-Chung from
Joerg Pommnitz's ttmkfdir, to have Big5/GB TTF better detection.
There are four modules as of version 0.9.0 for their respective apps:
chitex: For ChiTeX, a Chinese TeX/LaTeX
ttf2pk: Generate pk/tfm for CJK
xfreetype: For Xfsft in XFree86 3.3.x and "freetype" module in 4.x
xttfm: For X-TT in XFree86 3.3.x and "xtt" module in 4.x, available
in both tcl and sh versions.
Since ChiTeX hasn't been ported yet, CJK needs further tuning, only
xfreetype & xttfm are available right now. Future applications which
uses TTF(like ghostscript) should use TTFM to manage TTF. The importing
of this port also implies existing ports like arphicttf/moettf should
be re-written.
/me is always able to find trouble for himself. :-)
dot.*rc in ${PREFIX}/share/skel/zh_TW.Big5, instead of `make install-user`.
I also have to admit the RUN_DEPENDS section is too ugly, since it
strives identifying whether an installed port/package is Big5-ized.
For those which is unique or have "cookies", there should be no problem.
But the following still have problems, especially slave ports:
mutt
nvi-big5(if another version of nvi-m17n is installed)
pine4(not sure yet)
rxvt
aterm
tin
tintin++
xmms
at the BSDCon2000 I18N meeting. User only needs to install
the port to gain a fully localized Traditional Chinese system.
PR: 22673
Submitted by: Michael C. Wu <keichii@peorth.iteration.net>
USE_MOTIF and generates a LIB_DEPENDS to x11-toolkits/open-motif. As
before, it implies USE_XPM (and therefore USE_XLIB). Motif-dummy is
removed from PKG_IGNORE_DEPENDS, whose sole resident is now XFree86-3.
Most of the simple ".if defined(HAVE_MOTIF)"s are removed to always
have USE_MOTIF. ftp/moxftp will define USE_MOTIF unless
WANT_ATHENA_VERSION (new variable) is defined. I merged the X cases
in cad/mars, so USE_MOTIF is used iff WITHOUT_X11 is not defined.
I will remove x11-toolkits/Motif-dummy (which has been repo copied to
open-motif) in a few days.
mkisofs is a pre-mastering program that generates binary ISO9660
filesystem image. Users can then write the image to devices like
CD-R or CD-RW. This port can also generate Chinese Big5 filenames
on Joliet filesystems. Use the -J option to activate it.
Although it works flawlessly, further improvements are possible:
1. Accept GB(Simplified Chinese) filenames also.
2. Upgrade to mkisofs 1.13. 1.13 uses a better structure to manage
I18N, which is (almost all) borrowed from Linux kernel's fs code.
Better word wrapping for CJK strings
Specify default font's encoding explicitly
Better XLFD handling
Selection resync fixes
Numerous gtkrc.$locale samples
Other minor fixes
One flaw of this port is that CLE only supports i386 platform, so alpha
users still have to use x11-toolkits/linux-gtk.
Submitted by: Jing-Tang Keith Jang <keith@freebsd.sinica.edu.tw>
Introduce zh-XFree86-aoutlibs port.
From the description of the originator:
x11/XFree86-aoutlibs doesn't support Big5 encoding, so Big5 users
can't input Chinese using XIM in aout applications like Netscape.
This port installs 3.3.6 aout library binaries distributed by XFree86,
which solves the problem.
Submitted by: Jing-Tang Keith Jang <keith@freebsd.sinica.edu.tw>
Reviewed by: Michael C. Wu <keichii@peorth.iteration.net>
Active zh-netscape47-navigator.
Chinese Netscape navigator, slave port of chinese/netscape47-communicator.
Submitted by: Jing-Tang Keith Jang <keith@freebsd.sinica.edu.tw>
Reviewed by: Michael C. Wu <keichii@peorth.iteration.net>
Active zh-netscape47-navigator.
Submitted by: Jing-Tang Keith Jang <keith@sinica.edu.tw>
Reviewed by: Michael C. Wu <keichii@peorth.iteration.net>
Active zh-netscape47-communicator.
This port provides a Big5 localized Netscape, including menus,
toolbars, error messages and many others. It also contains a
libX11 fix, which allows Netscape to show Big5 buttons in html
documents.
The patches are mainly contributed by thhsieh@linux.org.tw &
hcchu@r350.ee.ntu.edu.tw. The localized Netscape.Big5 is
translated by Antony Shen <u8618806@cc.nctu.edu.tw>, with some
modifications.
Submitted by: Jing-Tang Keith Jang <keith@sinica.edu.tw>
Reviewed by: Michael C. Wu <keichii@peorth.iteration.net>
Obtained from: outta-port@freebsd.sinica.edu.tw
Add new port : cle_base.
CLE(Chinese Linux Extension) is a project providing Chinese L10N for
Linux, mainly the RedHat distribution. This port picks packages from
a minimal CLE installation. In addition to Linux emulation, you can
run L10N Linux applications like Netscape or (hopefully) Star Office.
WWW: http://cle.linux.org.tw
Submitted by: Michael C. Wu <keichii@peorth.iteration.net>
Add new port rxvt-gb(2.6.3) into ports/chinese, this is for GB(simp-Chinese)
and it is a slave port from x11/rxvt.
Submitted by: Michael C. Wu <keichii@peorth.iteration.net>
Active rxvt-big5
Add new port rxvt-big5(2.6.3), it enables big5 locale in rxvt andis a slave
port from x11/rxvt.
Mike Silbersack <silby@silby.com>. Add a build- and install-time
warning about the probable existence of further remote security
problems, and remove FORBIDDEN.
Submitted by: Mike Silbersack <silby@silby.com>
Reorder variables to satisfy portlint.
Approved by: Jing-Tang Keith Jang <keith@freebsd.sinica.edu.tw>
(MAINTAINER of chinese/xemacs)
KIRIYAMA Kazuhiko <kiri@pis.toba-cmt.ac.jp>
(MAINTAINER of xemacs-*-packages)
From pkg-descr:
pyDict is a Chinese(Big5) <-> English dictionary, written in Python.
It can be used as a console app or X11 window(with -x option). The
data is prepared by TurboLinux in simplified Chinese, then converted
to traditional Chinese by CLE(Chinese Linux Extension) team.
PR: ports/14643
Submitted by: Jing-Tang Keith Jang <keith@freebsd.sinica.edu.tw>
(a) Its name was changed from Wnn to FreeWnn because Wnn6 which is a
commercial software exits
(b) Its license was changed to GPL.
(c) The method to configure was changed from imake to GNU configure.
(d) Relatively to the original Wnn, the Wnn in the ports tree were
modified by me a lot. Most of the modifications were adopted
into FreeWnn.
(c) Header and library files are installed into
${LOCALBASE}/{lib,include} instead of ${X11BASE}/{lib,include}.
(2) FreeWnn is divided into two ports FreeWnn-lib and FreeWnn-server
in chinese, korean and japanese categories. The former is for libwnn
and header files to compile client commands, and the files used in
client commands. The latter is for a server to convert KANA to KANJI
(Chinese character), and dictionaries and files used by the server.
Notice: I forgot to commit FreeWnn-{lib,server} (^_^;;
(a) Its name was changed from Wnn to FreeWnn because Wnn6 which is a
commercial software exits
(b) Its license was changed to GPL.
(c) The method to configure was changed from imake to GNU configure.
(d) Relatively to the original Wnn, the Wnn in the ports tree were
modified by me a lot. Most of the modifications were adopted
into FreeWnn.
(c) Header and library files are installed into
${LOCALBASE}/{lib,include} instead of ${X11BASE}/{lib,include}.
(2) FreeWnn is divided into two ports FreeWnn-lib and FreeWnn-server
in chinese, korean and japanese categories. The former is for libwnn
and header files to compile client commands, and the files used in
client commands. The latter is for a server to convert KANA to KANJI
(Chinese character), and dictionaries and files used by the server.
cross-language master-slave ports.
I just wonder how I could overlook this. I thought I tested with Wnn,
netscape and so on. *sigh*
Yes, I know how I can fix it, but I'd certainly back out first.
Reported by: steve
Pointy hat to: knu
