Re: 内核升级后X启动问 题 ?

2008-07-17 文章 Kov Chai
On Thu, Jul 17, 2008 at 11:42:05AM +0800, [EMAIL PROTECTED] wrote:
 那个脚本是个自解压脚本,你先把它解压,再做修改吧。
没错。不过如果自己没有把握的话,修改不是很安全。

 
 以Wiki模式创建Linux中文文档,欢迎加入
 http://www.linux-wiki.cn/
 
 2008/7/17 zwang [EMAIL PROTECTED]:
 
  感谢!
 
  我用的是 NVIDIA-Linux-x86-173.14.09-pkg1.run, 好像没看这个版本的patch啊。
我不大清楚了。不过你可以多试试 google,这是我找到的。
http://www.nvnews.net/vbulletin/showthread.php?t=113737page=2
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=481485

  还有就是想直接编辑一下脚本中xen相关的部分,但用Vi打开乱码!

 
  2008/7/16 Kov Chai [EMAIL PROTECTED]:
 
  On Wed, Jul 16, 2008 at 04:30:23PM +0800, Jason wrote:
再请教一个简单点的问题:
/etc/network/interfaces
里面默认的配置是dhcp模式的,我将其改成了静态IP模式,但每次系统启动后都没有IP,需要手动执行sudo ifup eth0才能获得。
   
为什么改成静态之后该脚本就不再调用了昵(原来dhcp模式时每次启动都有IP的,说明该文件有被调用啊) ?
   
  
/etc/network/interfaces加上:
   auto eth0
  
   
2008/7/16 zwang [EMAIL PROTECTED]:
   
各位好:
   
系统是debian lenny, 现象系统有两个版本的内核(新的2.6.25-2是前两天刚更新的)。
   
   
   
  在使用旧内核时,为了使用compiz而装了nvidia驱动;现在使用新内核进行启动时,X启动不起来了,看Log中说好像是因为nvidia模块加载失败的问题!(搜索了一下nvidia.ko,nvidia好像是被装到老的内核module中去了)
   
然后用新内核启动后准备重新安装一下nvidia驱动,可提示说该内核是xen版本,nvidia不支持,那么我该如果让 X 在新内核下工作昵?
BTW:用旧的内核启动都还是正常的。
  我是 sid, NV Geforce2, bug#477643 [1],已经有更新的驱动包了。其实只要把 NV 驱动
  (脚本) 里面关于xen 检测的代码删掉就可以了。
 
  zwang, 你不妨可以自己打 patch 或者用新的 etch 的 driver package。其实 google
  一下就清楚了 [2]。=)
 
   
  
   那么你装的是xen版本的内核喽?新力德里搜一下 2.6.25,装和老内核一样后缀的版本.
   另外,如果nvidia的驱动是nvidia网站上下的话,装了新内核之后,要重编译nvidia的内核模块.切换到终端,
   sudo /etc/init.d/gdm stop
   sudo YOUR_DRIVER_FILE_PATH_AND_FILE -K
 
  [1] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=477643
  [2]
 
  http://www.google.com/search?q=Xen+kernel+debian+nvidia+ie=utf-8oe=utf-8aq=trls=org.debian:en-US:unofficialclient=iceweasel-a
 
  --
  Regards,
  Kov Chai
  2008.07.16 Wed
 
  --
  PI=
  3.141592653589793238462643383279502884
  19716939937510582097494459230781640628
  6208998628034825342117067982148086513...
 
  -BEGIN PGP SIGNATURE-
  Version: GnuPG v1.4.9 (GNU/Linux)
 
  iEYEARECAAYFAkh98IgACgkQkt4oayDNJDakuQCgn5VbUXj//RCEGEQpyjNVEWPy
  piEAnRGuC6v+NlZc3eAIH06LXU9QBKUM
  =j7iC
  -END PGP SIGNATURE-
 
 
 

-- 
Regards,
Kov Chai
2008.07.17 Thu

--
THEOREM: 1=2
PROOF: 
  1. Let a=b. 
  2. Then a^2 = ab, 
  3. a^2 + a^2 = a^2 + ab, 
  4. 2 a^2 = a^2 + ab, 
  5. 2 a^2 - 2 ab = a^2 + ab - 2 ab, 
  6. and 2 a^2 - 2 ab = a^2 - ab. 
  7. Writing this as 2 (a^2 - a b) = 1 (a^2 - a b), 
  8. and cancelling the (a^2 - ab) from both sides gives 1=2. 


signature.asc
Description: Digital signature


Re: lenny下配置sendmai l服务器的问题

2008-07-17 文章 Kov Chai
On Thu, Jul 17, 2008 at 10:52:25AM +0800, Menghan Zheng wrote:
 我想配置实现一定功能的邮件服务器。
 已经有的文档是关于sendmail的。在debian中,默认的MTA是exim4。据说exim4可
 以伪装成sendmail,
 我在/usr/sbin下找到了sendmail-exim4的链接。但是sendmail相关的配置文件却
 没有在系统中找到。
只不过很多 exim4 的命令行与 sendmail 兼容。

 
 我推测,如果想让exim4伪装成sendmail服务,一定要有sendmail的配置文件,如
什么叫伪装啊?让 exim4 也能读 sendmail 的配置么?

 /etc/mail/sendmail.cf。
 文档中的一些配置也与这些文件有关,但是我的系统里没有。
我的 exim4 也没有这个文件。

 
 请问我如何做,才能实现文档中关于邮件服务器的配置
能告诉大家是什么文档么?

 
 -- 
 Keep In Touch!
 Electronics Engineering Department, Tsinghua University
 
 Dorm Tel: 8610 5153 4319
 Mobile: 86 13401088180
 MSN: [EMAIL PROTECTED]
 Email: [EMAIL PROTECTED], [EMAIL PROTECTED] 
 
 
 
 -- 
 To UNSUBSCRIBE, email to [EMAIL PROTECTED]
 with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
 

-- 
Regards,
Kov Chai
2008.07.17 Thu

--
THEOREM: The limit as n goes to infinity of sin x/n is 6. 
PROOF: cancel the n in the numerator and denominator. 


signature.asc
Description: Digital signature


Re: 内核升级后X启动问 题 ?

2008-07-16 文章 Kov Chai
On Wed, Jul 16, 2008 at 04:30:23PM +0800, Jason wrote:
  再请教一个简单点的问题:
  /etc/network/interfaces
  里面默认的配置是dhcp模式的,我将其改成了静态IP模式,但每次系统启动后都没有IP,需要手动执行sudo ifup eth0才能获得。
 
  为什么改成静态之后该脚本就不再调用了昵(原来dhcp模式时每次启动都有IP的,说明该文件有被调用啊) ?
 
 
  /etc/network/interfaces加上:
 auto eth0
 
 
  2008/7/16 zwang [EMAIL PROTECTED]:
 
  各位好:
 
  系统是debian lenny, 现象系统有两个版本的内核(新的2.6.25-2是前两天刚更新的)。
 
 
  在使用旧内核时,为了使用compiz而装了nvidia驱动;现在使用新内核进行启动时,X启动不起来了,看Log中说好像是因为nvidia模块加载失败的问题!(搜索了一下nvidia.ko,nvidia好像是被装到老的内核module中去了)
 
  然后用新内核启动后准备重新安装一下nvidia驱动,可提示说该内核是xen版本,nvidia不支持,那么我该如果让 X 在新内核下工作昵?
  BTW:用旧的内核启动都还是正常的。
我是 sid, NV Geforce2, bug#477643 [1],已经有更新的驱动包了。其实只要把 NV 驱动
(脚本) 里面关于xen 检测的代码删掉就可以了。

zwang, 你不妨可以自己打 patch 或者用新的 etch 的 driver package。其实 google 
一下就清楚了 [2]。=)

 
 
 那么你装的是xen版本的内核喽?新力德里搜一下 2.6.25,装和老内核一样后缀的版本.
 另外,如果nvidia的驱动是nvidia网站上下的话,装了新内核之后,要重编译nvidia的内核模块.切换到终端,
 sudo /etc/init.d/gdm stop
 sudo YOUR_DRIVER_FILE_PATH_AND_FILE -K

[1] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=477643
[2]
http://www.google.com/search?q=Xen+kernel+debian+nvidia+ie=utf-8oe=utf-8aq=trls=org.debian:en-US:unofficialclient=iceweasel-a

-- 
Regards,
Kov Chai
2008.07.16 Wed

--
PI=
3.141592653589793238462643383279502884
19716939937510582097494459230781640628
6208998628034825342117067982148086513...


signature.asc
Description: Digital signature


Re: Debian Installer sublevel 1: Please update the PO translation for the package Debian Installer sublevel 1

2008-06-22 文章 Kov Chai
Hi,

对不起,似乎前几封信没有发到列表上。附件是我的翻译。

-- 
Regards,
Kov Chai
2008.06.22 Sun

--
Pascal is for building pyramids - imposing, breathtaking, static structures
built by armies pushing heavy blocks into place.
Lisp is for building organisms - imposing, breathtaking, dynamic structures
built by squads fitting fluctuating myriads of simpler organisms into place.
-- Alan J. Perlis, in the Foreword to SICP


zh_CN.po.gz
Description: Binary data


signature.asc
Description: Digital signature


Re: Debian Installer sublevel 1: Please update the PO translation for the package Debian Installer sublevel 1

2008-06-21 文章 Kov Chai
Hi Ming,

2008/6/21 Ming Hua [EMAIL PROTECTED]:


 我最近没有时间,很感谢你帮忙。

谢谢你才对。=)


  我想,用 alioth 的帐号应该也有 check in 的权限吧?如果有做得不对的地方,
  能告诉我么?

 目前似乎是只有 Carlos 和我能 commit,你可以试一试,总不会有坏处的。:-P

啊,的确是这样。你能帮我 commit 么?我在 debian-boot 申请了一下 d-i SVN 的权限,但是还没有人回应我。

谢谢啦。

-- 
Regards
Kov Chai


Re: 请校对 w.d.o/releases/etch 网页的翻译

2008-06-21 文章 Kov Chai
Hi ,


2008/6/21 Ji ZhengYu [EMAIL PROTECTED]:

 On Wed, Jun 11, 2008 at 8:43 PM, Vern Sun [EMAIL PROTECTED] wrote:
  proof listing:
 
 http://people.openrays.org/~s5unty/www.d.o/releases/etch/index.zh-cn.htmlhttp://people.openrays.org/%7Es5unty/www.d.o/releases/etch/index.zh-cn.html
 
 http://people.openrays.org/~s5unty/www.d.o/releases/etch/index.zh-tw.htmlhttp://people.openrays.org/%7Es5unty/www.d.o/releases/etch/index.zh-tw.html
 

 ---Debian GNU/Linux 4.0r2 已经于2007年12月26日发布。此发行版包含很多重要的更新, 更多描述信息请见我们的
 新闻稿 和 发行备注。
 +++(新的原文)Debian GNU/Linux 4.0r3 was released February 17th, 2008. The
 release included many major changes, described in our press release
 and the Release Notes.
 +++Debian GNU/Linux 4.0r3 已经于2008年2月17日发布。

 ---如何获取和安装 Debian GNU/Linux,请看 安装程序 页面和 安装手册。怎样从一个旧的 Debian
 发行版基础上升级,请看 发行备注 中的介绍。
 +++(新的原文)To obtain and install Debian GNU/Linux, see the installation
 information page and the Installation Guide. To upgrade from an older
 Debian release, see the instructions in the Release Notes.
 +++如何获取和安装 Debian GNU/Linux,请看 安装信息 页面和 安装手册。怎样从一个旧的 Debian
 发行版基础上升级,请看 发行备注 中的介绍。

关于如何获取和安装 Debian GNU/Linux,请参考安装信息页面和 安装手册。要在一个旧的 Debian
发行版基础上升级,请看发行备注中的介绍。

其中,从xxx基础上升级,不是非常通顺。改为在xxx的基础上升级 似乎更合适一些。



 ---虽然该发行版被称为稳定版 (stable),但是它仍有可能会意外的出现一些问题。我们已经完成了一个 已知重要问题
 的列表,另外您可以随时 报告其它问题 给我们。
 最后要强调的是,我们准备了一个肩负此发行版面世使命的 人员列表。
 +++(新的原文)
 Contrary to our wishes, there may be some problems that exist in the
 release, even though it is declared stable. We've made a list of the
 major known problems, and you can always report other issues to us.

 Last but not least, we have a list of people who take credit for
 making this release happen.

 +++与我们所期望的相反,虽然该发行版被称为稳定版(stable),但它还是有些问题存在的。我们已经完成了一个已知重要问题的列表,另外您也可以随时报告其它问题给我们。

愿望是美好的,现实是残酷的。虽然该发行版被称为稳定版(stable),但是它还是可能会有些问题。我们备有一个列表,记录了那些已知的比较严重的问题,也欢迎你向我们报告其它问题。



 最后(但不限于这些人),我们准备了一份人员列表,他们为此版的正常发行做出了贡献。

最后,也是很重要的一点,我们有一个列表,它上面列有为此次发行做出了贡献的人们的名字。


 Last but not least 不知道译对了没有……

 
应该是,最后的,但不是最不重要的,这句话在英语里面经常用在最后,因为被放到最后的内容通常被认为不是很重要,所以对于虽然被安排在文章(或者是演说)末尾那些还是很重要的内容,可以用``last
but not least''。





-- 
Regards
Kov Chai


Re: Debian Installer sublevel 1: Please update the PO translation for the package Debian Installer sublevel 1

2008-06-18 文章 Kov Chai
Hi 大家好, 

由于 d-i level 1 很重要,我自做主张翻译了 fuzzy 和未翻译的部分,并做了一些细微的调整。能帮着校对一下么?如果没有问题,Ming Hua 
也没有异议的话,我就 check in 了。

Hi Ming,
你是最后一个译者,但是看到几天都没有回应。而且 SVN 中的版本也是 Christian 发的版本,所以我把 d-i 的 level 1 中 fuzzy 
和未翻译的部分翻译了,还根据 SVN 里 partman-basicfilesystems.templates 的提示改了一条翻译 (搜索 news)。

我想,用 alioth 的帐号应该也有 check in 的权限吧?如果有做得不对的地方,能告诉我么?

谢谢大家。




On Sun, Jun 15, 2008 at 09:29:37AM +0200, Christian Perrier wrote:
 Hi,
 
 Debian Installer sublevel 1 is still incomplete for your language. If it is
 not complete as of the release of Debian lenny, your language will be
 disabled in the installation process.
 
 We therefore recommend you to update this file ASAP and later keep it updated.
 
 To be added to the automatic notification list to be aware of changes,
 please answer to that mail.
 
 Thanks in advance,
 



-- 
Regards,
Kov Chai
2008.06.18 Wed

--
Group theory - An exceedingly beautiful branch of pure mathematics
used for showing in how many ways blocks of wood can be painted. 
 -- R. Ainsley in Bluff your way in Maths, 1988


signature.asc
Description: Digital signature


Re: iceweasel 总出 现 segmentation fault

2008-06-02 文章 Kov Chai
On Sun, Jun 01, 2008 at 04:12:50PM -0400, Michael Yang wrote:
 2008/5/31 Kov Chai [EMAIL PROTECTED]:
 
  On Fri, May 30, 2008 at 10:51:10PM -0400, Michael Yang wrote:
   经常用到途中突然crash,出现的错误信息只有一条 segmentation fault (从终端启动返回的)
  
   怎么找原因阿?
  你为了你的 iceweasel 装了 add-on 或者 plugin 么?会不会是它们造成的?
  用 iceweasel -safe-mode 启动 iceweasel,可以让 iceweasel 禁用 add-on。你要不 试试看,会不会还出现
  seg fault。
 
 
 去掉了一些不重要的addon,还是偶尔会有
 目前里面还有的addon有: googletools, proxy-manager, super-drag-drop
 要是禁用掉所有的add-on,就感觉失去firefox的特色了
的确,这样的话用起来会缺少很多乐趣。但是暂时禁用 add-on
或许可以帮助你更好地定位问题[1]。=)

--
[1] http://www.mozilla.com/en-US/firefox/2.0/releasenotes/#troubleshooting

-- 
Regards,
Kov Chai
2008.06.02 Mon

--
Let no one ignorant of mathematics enter here.
   --   Plato, Inscription written over the 
entrance to the academy


signature.asc
Description: Digital signature


Re: debian4.0中的menco der该如何安装 ?

2008-06-02 文章 Kov Chai
On Mon, Jun 02, 2008 at 11:56:53AM +0800, [EMAIL PROTECTED] wrote:
 zwang 写道:
  大家好!
  我通过网络安装了debian4.0 lenny, 安装后发现没有mencoder,但mplayer是可以正常使用的!
  我看网上说mencoder应该是与mplayer一起被安装的才对,但现在就是没有,源里也找不到,想下载源码包也没搜索到!
  各位有什么好的建议,请告知,多谢!
 
 这些非自由的软件包, 应该在 non-free 和 contrib目录里.
这个说法有点问题。contrib 里面的软件包应该是符合 DFSG
的。这些软件包之所以被分到 contrib 里,是因为他们直接地或间接地依赖于
non-free 的包,或者它们依赖的包不在 Debian 里面。

而且,mencoder 也不在 contrib 或者 non-free 里面。用 Google 搜索 mencoder 和
debian 第一个结果[1]里面就有 www.debian-multimedia.org
的信息。这和另一位朋友提到的是一个地方。


[1] http://wiki.debian.org/MPlayer


-- 
Regards,
Kov Chai
2008.06.02 Mon

--
A topologist is one who doesn't know the difference between a doughnut
and a coffee cup.
  -- J. Kelley


signature.asc
Description: Digital signature


Re: iceweasel 总出 现 segmentation fault

2008-05-31 文章 Kov Chai
On Fri, May 30, 2008 at 10:51:10PM -0400, Michael Yang wrote:
 经常用到途中突然crash,出现的错误信息只有一条 segmentation fault (从终端启动返回的)
 
 怎么找原因阿?
你为了你的 iceweasel 装了 add-on 或者 plugin 么?会不会是它们造成的?
用 iceweasel -safe-mode 启动 iceweasel,可以让 iceweasel 禁用 add-on。你要不 试试看,会不会还出现 seg 
fault。

-- 
Regards,
Kov Chai
2008.05.31 Sat

--
An engineer, a chemist and a mathematician are staying in three adjoining 
cabins at an old motel. First the engineer's coffee maker catches fire. 
He smells the smoke, wakes up, unplugs the coffee maker, throws it out 
the window, and goes back to sleep. Later that night the chemist smells 
smoke too. He wakes up and sees that a cigarette butt has set the trash 
can on fire. He says to himself, Hmm. How does one put out a fire?
One can reduce the temperature of the fuel below the flash point, isolate 
the burning material from oxygen, or both. This could be accomplished 
by applying water. So he picks up the trash can, puts it in the shower 
stall, turns on the water, and, when the fire is out, goes back to sleep. 
The mathematician, of course, has been watching all this out the window. 
So later, when he finds that his pipe ashes have set the bed-sheet on fire,
he is not in the least taken aback. He says: Aha! A solution exists! 
and goes back to sleep.


signature.asc
Description: Digital signature


Re: 请校对 w.d.o/intro/cn 网页的翻译

2008-05-15 文章 Kov Chai
2008/5/14 Vern Sun [EMAIL PROTECTED]:
 proof listing:
  http://s5unty.googlepages.com/cn.zh-tw.html
  http://s5unty.googlepages.com/cn.zh-cn.html

 在原翻译[1]的基础上,参考了与最新版本的差异[2],请大家校对。
- 世界上并非人人都使用同样的语言 改为 显而易见,世界上并不是所有的人都使用同种语言
- 属性协商我想还是译为内容协商更合适一些,这是普遍采用的翻译。
- 原始的语言 改为 原来的语言
- 设定是不是改成设置更符合我们大陆使用者的习惯?
- as explained below 翻译成 例如下面的说明 不是很妥当,不如改成下面将会作具体的解释。
- 映射网站 依照大陆习惯应为镜像站点
- They sit in the middle between users and real web
servers改译为它们介于用户和真正的网络服务器之间
- 本地端 改为 本地
- 存放一个缓存的拷贝,以便往后使用 改为 保存一个拷贝作为缓存,以备以后取用
- 网络交通负担减少很多改为网络带宽的压力大大减少
- 大部份时间来说,这是一个很好的方式,但是当缓存有问题时就麻烦了改为大多数时候,这是个好主意,但是如果缓存出问题的话,这个办法就不行了。
- 此时唯一的方式就是用力向您的 ISP 业者抱怨,要求他们优化或更新它们的缓存软件 改为 这时唯一的办法就是向您的 ISP
提供商提意见,要求他们升级或更新他们的缓存软件 似乎更妥
- people only used a proxy when they configured their browser to use
one. 改译为只有用户设定让他的浏览器使用代理的时候,才会用到代理服务器
- 将所有的 HTTP 连线要求导向到一个代理服务器 改为将所有的 HTTP 请求重定向到一个透明的代理上
- some older proxy servers do not understand content negotiation 译为
有些较旧的代理服务器不支持内容协商机制
- For example, if you are a native French speaker, you would want to
set your language variable to include first the French language (with
the language code 'fr'), followed by English (with the language code
'en'). 原译为 例如,您说的是中文,您可能希望将您的语言第一个设定成繁体中文(语言代码为 'zh_TW' 或
'zh_TW.Big5'),然后第二个设定成英文(语言代码为 'en')。 
我想,按照简体中文的情况,是否可以改成例如,如果您的母语是中文,您可能希望将您的语言设置中的第一个设置成简体中文(语言代码为
'zh_CN' 或 'zh_TW.GBK'),并把第二个设置成英文(语言代码为 'en'), 并且后面的例子也应该相应地改为
zh_CN, en 而非繁体中文中保留的fr, en. 类似的还有 en-GB, fr、fr;q=1.0,en;q=0.5
和法文,都最好一并改掉。
- 及以后的版本 改为 及更高的版本 或者 及后续版本


Re: Fwd: Bug#478811: ITP: sunpinyin -- An input method engine based on SLM

2008-05-09 文章 Kov Chai
我用的是 i386 的机器。
2008/5/7 ZhengPeng Hou [EMAIL PROTECTED]:

 attached the patch

太谢谢了。 我回家编译个 gcc-4.3 试试看。


 -BEGIN PGP SIGNATURE-
 Version: GnuPG v1.4.6 (GNU/Linux)

 iD8DBQFIISL1JrhDLBNiZtwRAnkiAJkBIh8uadCkixVK7zvYpWc4Rp3oSQCgif2H
 3twkt9eSRBjRV3im79ZJVHE=
 =vT1H
 -END PGP SIGNATURE-




-- 
Regards
Kov Chai


Re: Bug#478811: ITP: sunpinyin -- An input method engine based on SLM

2008-05-09 文章 Kov Chai
Xiyue,我把在 debian-develop 上的信和 Julien 的回覆被放到信尾了。他和你的意见是一致的。:)

谢谢你的建议。我现在就快把包分好了,周末估计就能弄好。虽然还有些问题没有解决,比如说源代码包过大的问题。

2008/5/5 Deng Xiyue [EMAIL PROTECTED]:

 CC 回 debian-chinese-gb@

 在 2008-05-05一的 23:38 +0800,Kov Chai写道:
  -BEGIN PGP SIGNED MESSAGE-
  Hash: SHA1
 
 
 
  2008/5/5 Deng Xiyue :
  
 
   如果是与架构无关的话,单独提取到 -data 里可以节约仓库的空间。不过目前看
   起来并非如此。
 放到 -data 里的确可以节约仓库的空间啊,如果 mirror 同时提供同种 endian
 的几种架构,那么就可以节省一些空间。而且如果用户同时使用 iiimf-le-sunpinyin-chinese 和
 scim-sunpinyin,那么用户没有必要同时保存两份数据。尽管这个可能性不大。
  
确实,可以将两个都打包成 arch: all 的包,然后在不同的架构上依赖不同的
  
   包。
  既然哪个 arch 是 big endian,哪个 arch 是 small endian 都是已知的了,能不能让 i386, alpha,
  amd64 用 small endian 的数据文件;而让 sparc, m68k 它们用 big endian 的数据文件呢?
 
  也就是按照目标体系架构直接生成对应的 deb 包,包的名字就叫 sunpinyin-data,只不过每个 arch 只有对应自己 endian
 的数据。
  而不是两个 arch:all 的包,即 sunpinyin-le-data 和 sunpinyin-be-data。并且 arch:all
  包应该是 arch 中立的才对,不是么?
 
  这样不是更干净么?我猜 Debian 是不是已经有这样的机制了?

 有的。但是这样每个架构都要存储这几十MB的东西, ftp-master 未必会同意,而
 arch: all 可以多个架构共用一个,在仓库里也只有一个副本,相对应该好接受一
 些。可以通过指定 scim-sunpinyin 在不同架构下的依赖关系来选择使用哪个数
 据。当然咨询下 debian-devel@ 更好 :)


On Wed, May 7, 2008 at 1:14 AM, Julien BLACHE [EMAIL PROTECTED] wrote:
Kov Chai [EMAIL PROTECTED] wrote:

Hi,

 best approach in my mind is to create two other data packages named
 sunpinyin-le-data and sunpinyin-ge-data, and make the binary packages
depend
 on the data package by its endianness accordingly.

 Is this solution feasible, or is there any better way to solve this
problem?

Yes, that makes sense. At least as long as your data files are
compatible between 32bit and 64bit hosts.

(and make that -data-le and -data-be, IMHO)

 And moreover, I am afraid that sunpinyin-{le,ge}-data are not architecture
 neutral. So is it okay to make these two packages Arch:all?

Yes.

 At last, Is there any way to specify the package dependency by
architecture?

Use a substitution variable in debian/control and determine the
correct value in debian/rules before calling dh_gencontrol.

JB.


-- 
Regards
Kov Chai


Re: Bug#478811: ITP: sunpinyin -- An input method engine based on SLM

2008-05-09 文章 Kov Chai
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1



2008/5/9 Kov Chai [EMAIL PROTECTED]:

Xiyue,我把在 debian-develop 上的信和 Julien 的回覆被放到信尾了。他和你的意见是一致的。:)

谢谢你的建议。我现在就快把包分好了,周末估计就能弄好。虽然还有些问题没有解决,比如说源代码包过大的问题。
已经改好了。新的包在
dget
http://mentors.debian.net/debian/pool/main/s/sunpinyin/sunpinyin_1.0~hg20080501-1.dsc
可以取到。

- --
Regards
Kov Chai
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFIJG3gC8hyx6EmTc0RAmF1AJ9vNE14hHHLijSMKvtGyGRqEH31wgCgg33O
0dUU1LJtdLDWIiBHTQbD7kE=
=fPYr
-END PGP SIGNATURE-


how to split architecture dependent data files into separate package

2008-05-06 文章 Kov Chai
Hi all,

I am packaging [1] sunpinyin. Sunpinyin is a Chinese input method engine.

As you may know, this source package builds two binary packages for two
input method platform (SCIM and IIIMF) respectively. Both of these binary
packages contain a copy of the lexicon and its index data files. The total
size of the data files is about 19M in each binary package. And what makes
things more complicate is that there are actually two copies of data in
source package. One is in big-endian format, the other is in small-endian
format. So the source package is around 38M.

As the size of data files in the package is considerable, I think it is
necessary to split the data out into separate package(s) [2]. So far, the
best approach in my mind is to create two other data packages named
sunpinyin-le-data and sunpinyin-ge-data, and make the binary packages depend
on the data package by its endianness accordingly.

Is this solution feasible, or is there any better way to solve this problem?


And moreover, I am afraid that sunpinyin-{le,ge}-data are not architecture
neutral. So is it okay to make these two packages Arch:all?

At last, Is there any way to specify the package dependency by architecture?

Thanks in advance.

--
[1] Bug#478811: ITP: sunpinyin -- An input method engine based on SLM
[2]
http://www.debian.org/doc/developers-reference/ch-best-pkging-practices#s-bpp-archindepdata

-- 
Regards
Kov Chai


how to split architecture dependent data files into separate package

2008-05-06 文章 Kov Chai
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Hi all,

I am packaging [1] sunpinyin. Sunpinyin is a Chinese input method engine.

As you may know, this source package builds two binary packages for
two input method platform (SCIM and IIIMF) respectively. Both of these
binary packages contain a copy of the lexicon and its index data
files. The total size of the data files is about 19M in each binary
package. And what makes things more complicate is that there are
actually two copies of data in source package. One is in big-endian
format, the other is in small-endian format. So the source package is
around 38M.

As the size of data files in the package is considerable, I think it
is necessary to split the data out into separate package(s) [2]. So
far, the best approach in my mind is to create two other data packages
named sunpinyin-le-data and sunpinyin-ge-data, and make the binary
packages depend on the data package by its endianness accordingly.

Is this solution feasible, or is there any better way to solve this problem?


And moreover, I am afraid that sunpinyin-{le,ge}-data are not
architecture neutral. So is it okay to make these two packages
Arch:all?

At last, Is there any way to specify the package dependency by architecture?

Thanks in advance.

- --
[1] Bug#478811: ITP: sunpinyin -- An input method engine based on SLM
[2]
http://www.debian.org/doc/developers-reference/ch-best-pkging-practices#s-bpp-archindepdata

- --
Regards
Kov Chai


-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFIIIksC8hyx6EmTc0RAnWLAKC9A61mIwKO1o+2bJgOiwXtfvH7cQCbB3SK
1CwRJxuMlhItBZiSeRKpw/w=
=PtK2
-END PGP SIGNATURE-


how to split architecture dependent data files into separate package

2008-05-06 文章 Kov Chai
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Hi all,

I am packaging [1] sunpinyin. Sunpinyin is a Chinese input method engine.

As you may know, this source package builds two binary packages for
two input method platform (SCIM and IIIMF) respectively. Both of these
binary packages contain a copy of the lexicon and its index data
files. The total size of the data files is about 19M in each binary
package. And what makes things more complicate is that there are
actually two copies of data in source package. One is in big-endian
format, the other is in small-endian format. So the source package is
around 38M.

As the size of data files in the package is considerable, I think it
is necessary to split the data out into separate package(s) [2]. So
far, the best approach in my mind is to create two other data packages
named sunpinyin-le-data and sunpinyin-ge-data, and make the binary
packages depend on the data package by its endianness accordingly.

Is this solution feasible, or is there any better way to solve this problem?


And moreover, I am afraid that sunpinyin-{le,ge}-data are not
architecture neutral. So is it okay to make these two packages
Arch:all?

At last, Is there any way to specify the package dependency by architecture?

Thanks in advance.

- --
[1] Bug#478811: ITP: sunpinyin -- An input method engine based on SLM
[2]
http://www.debian.org/doc/developers-reference/ch-best-pkging-practices#s-bpp-archindepdata

- --
Regards
Kov Chai

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFIIIx4C8hyx6EmTc0RAvkFAKCZVahdf+l+DrKdajaqfHJP43pGBQCfXRO9
e0sFQ6C/yQU1if2G2NGZIqY=
=YzYo
-END PGP SIGNATURE-


how to split architecture dependent data files into separate package

2008-05-06 文章 Kov Chai
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Hi all,

I am packaging [1] sunpinyin. Sunpinyin is a Chinese input method engine.

As you may know, this source package builds two binary packages for
two input method platform (SCIM and IIIMF) respectively. Both of these
binary packages contain a copy of the lexicon and its index data
files. The total size of the data files is about 19M in each binary
package. And what makes things more complicate is that there are
actually two copies of data in source package. One is in big-endian
format, the other is in small-endian format. So the source package is
around 38M.

As the size of data files in the package is considerable, I think it
is necessary to split the data out into separate package(s) [2]. So
far, the best approach in my mind is to create two other data packages
named sunpinyin-le-data and sunpinyin-ge-data, and make the binary
packages depend on the data package by its endianness accordingly.

Is this solution feasible, or is there any better way to solve this problem?


And moreover, I am afraid that sunpinyin-{le,ge}-data are not
architecture neutral. So is it okay to make these two packages
Arch:all?

At last, Is there any way to specify the package dependency by architecture?

Thanks in advance.

- --
[1] Bug#478811: ITP: sunpinyin -- An input method engine based on SLM
[2]
http://www.debian.org/doc/developers-reference/ch-best-pkging-practices#s-bpp-archindepdata

- --
Regards
Kov Chai


-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFIIIY3C8hyx6EmTc0RAu58AKDCFdJXRcfHGiB8D2l+wGlzhm91/QCgiJRv
Vnboqj6CWShU955yAwFqvIs=
=Jzsw
-END PGP SIGNATURE-


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Bug#478811: ITP: sunpinyin -- An input method engine based on SLM

2008-05-05 文章 Kov Chai
2008/5/5 ZhengPeng Hou [EMAIL PROTECTED]:

 --12:25:37--

 http://mentors.debian.net/debian/pool/main/s/sunpinyin/sunpinyin_1.0.orig.tar.gz
   = `sunpinyin_1.0.orig.tar.gz'
   Resolving mentors.debian.net... 64.79.197.109
   Connecting to mentors.debian.net|64.79.197.109|:80...
   connected.
   HTTP request sent, awaiting response... 200 OK
   Length: 39,363,185 (38M) [application/x-gzip]
 38M? 都是什么啊


38M 是有点大。主要是 data 目录下的四个文件。 lm_sc.t3g.{sparc,i386} (6727K*2),
pydict_sc.bin.{sparc,i386}
(23M*2)。前者是线索化后的语言模型数据[1],目的是加速查找速度和压缩数据,差不多就是建立索引的效果,有了它能较快地计算 n-gram
语言模型里一串单字 (S = {W_1,W_2, W_3, ..., W_n}) 成为该语言模型里一个句子的概率
P(S)。后者则是词表(lexicon),或者说是大家常说的输入法的词库,这个词表支持不完全拼音和词到词的转换。由于要支持 big endian 和
small endian 的体系架构,所以干脆就把两种情况下的数据文件都弄上来了。

有没有必要把数据文件单独作为一个 package,放到 sunpinyin-data 里面去呢?还是有更好的办法呢?

谢谢啦。

-- 
[1]
http://blogs.sun.com/yongsun/entry/sunpinyin%E4%BB%A3%E7%A0%81%E5%AF%BC%E8%AF%BB_%E4%BA%94
[2]
http://blogs.sun.com/yongsun/entry/sunpinyin%E4%BB%A3%E7%A0%81%E5%AF%BC%E8%AF%BB_%E4%B8%83

-- 
Regards
Kov Chai


Fwd: Bug#478811: ITP: sunpinyin -- An input method engine based on SLM

2008-05-05 文章 Kov Chai
On Fri, May 2, 2008 at 8:17 AM, li xin [EMAIL PROTECTED] wrote:

 sunpinyin从solaris移植过来了?


其实,SunPinyin 在 IIIMF 上一直就是可以用的。移植到 SCIM 平台的工作在去年底也已经完成了。:)


-- 
Regards
Kov Chai


Re: Bug#478811: ITP: sunpinyin -- An input method engine based on SLM

2008-05-05 文章 Kov Chai
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1



2008/5/5 ZhengPeng Hou [EMAIL PROTECTED]:
On Sat, May 03, 2008 at 09:44:40PM +0800, Jun Ma wrote:
 debian etch编译安装:
 1. sudo apt-get install devscripts
 2. dget
http://mentors.debian.net/debian/pool/main/s/sunpinyin/sunpinyin_1.0-1.dsc
 3. dpkg-source -x sunpinyi_1.0-1.dsc
 4. sudo vi /usr/lib/pkgconfig/scim.pc
 修改moduledir为moduledir=${exec_prefix}/lib/scim-1.0/1.4.0
你应该去patch sunpinyin.我觉得,这个其实是 etch 中 libscim-dev 的问题。etch 中
libscim-dev 的版本是 1.4.4-7。同时, 在 etch 里,scim 版本是 1.4.4-7。自从 1.4.1-1,SCIM
上输入法的所有模块就被放到了 /usr/lib/scim-1.0/1.4.0/ 而不是以前的 /usr/lib/scim-1.0。但是
libscim-dev_1.4.4-7 中指定的 moduledir 却是
${exec_prefix}/lib/scim-1.0。很明显,如果不在输入法模块的 package 里 hardcode 或者修改
libscim-dev 是无法让 etch 里 SCIM 输入法模块正常编译或者工作的。bug#422739 虽然是针对 lenny
的,但是这个问题在 etch 中也存在。etch 中的 scim-pinyin 是通过在其 Makefile.am 中指定
moduledir 路径为 @SCIM_MODULEDIR@/1.4.0/IMEngine
解决的。很明显,这个办法只能适用于老版本的  libscim-dev,对于新版的 libscim-dev (比如
1.4.7-3)就不能用了。我们可以知道,sid 中的 scim-pinyin build-depends on libscim-dev
(= 1.4.6-1)。这是为什么 testing 和 sid 中的 scim-pinyin 都能正常编译的原因。

由于 sunpinyin 的 deb 包才刚刚准备进入 Debian,我想如果成功 upload 的话,它应该先会进入
sid。所以现在恐怕无法兼顾 testing 中的编译问题。如果将来进入 testing 时 libscim-dev 还是 1.4.4-7
的话,也许要专门 backport 一个版本,让它的 Makefile.am 中路径像 scim-pinyin 一样写死。


 5. cd sunpinyin-1.0  sudo dpkg-buildpackage
s/sudo/fakeroot
 6. sudo dpkg -i ../scim-sunpinyin_1.0-1_i386.deb
 7. 重启scim

 2008/5/3 Jun Ma [EMAIL PROTECTED]:
  On Fri, May 2, 2008 at 10:02 AM, Yong Sun [EMAIL PROTECTED] wrote:
Yes, SunPinyin had been ported to SCIM framework. And now it supports
IIIMF,
SCIM, BeCJK, and the Mac OS porting (leopard) is on going.
   
 Regards,
   
 
   在kov的帮助下,我顺利的切换到了SCIM-PINYIN。
   感觉不错,大家可以测试下。
 
 
 
   --
 
 
  FIXME if it is wrong.
 



 --
 FIXME if it is wrong.

- -BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFIHrGZJrhDLBNiZtwRApxeAJ9YLxRPCr4/99PGymfF7K8FmzIbPACgjOld
bxeo5aMg9B2P3phoTNDJuAo=
=kmFq
- -END PGP SIGNATURE-




- --
Regards
Kov Chai
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: http://getfiregpg.org

iD8DBQFIHwMWC8hyx6EmTc0RAuadAJ0bsFJL7oLdr5TCLfJdpawKnRZ+3wCgnTwO
PbfQvl9TEMZ037DvdqmbWEk=
=d+wp
-END PGP SIGNATURE-


Re: Bug#478811: ITP: sunpinyin -- An input method engine based on SLM

2008-05-05 文章 Kov Chai
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1



2008/5/5 Deng Xiyue :
 在 2008-05-05一的 14:04 +0800,Kov Chai写道:

 
  
   2008/5/5 ZhengPeng Hou :
   --12:25:37--
   
 http://mentors.debian.net/debian/pool/main/s/sunpinyin/sunpinyin_1.0.orig.tar.gz
 = `sunpinyin_1.0.orig.tar.gz'
 Resolving mentors.debian.net... 64.79.197.109
 Connecting to mentors.debian.net|
   64.79.197.109|:80...
 connected.
 HTTP request sent, awaiting response... 200 OK
 Length: 39,363,185 (38M) [application/x-gzip]
   38M? 都是什么啊
  
   38M 是有点大。主要是 data 目录下的四个文件。 lm_sc.t3g.{sparc,i386}

  (6727K*2), pydict_sc.bin.{sparc,i386} (23M*2)。前者是线索化后的语言模
   型数据[1],目的是加速查找速度和压缩数据,差不多就是建立索引的效果,有

  了它能较快地计算 n-gram 语言模型里一串单字 (S = {W_1,W_2, W_3, ...,
   W_n}) 成为该语言模型里一个句子的概率 P(S)。后者则是词表(lexicon),或
   者说是大家常说的输入法的词库,这个词表支持不完全拼音和词到词的转换。由
   于要支持 big endian 和 small endian 的体系架构,所以干脆就把两种情况下
   的数据文件都弄上来了。

  听起来像是应该与架构无关的数据, big endian/small endian 的转换不是在程
  序中进行而是提供不同的数据?感觉应该可以避免。
我想 big endian 和 small endian 就是架构不同的体现吧? 虽然不是一一对应的关系,而是 n 对 2 的关系。
big endian 和 small endian
的确不是在程序中进行的。我想这是基于性能和易于编程的考虑。可以避免什么呢?避免同时提供两份事实上是同样的数据么?


  
   有没有必要把数据文件单独作为一个 package,放到 sunpinyin-data 里面去
   呢?还是有更好的办法呢?

  如果是与架构无关的话,单独提取到 -data 里可以节约仓库的空间。不过目前看
  起来并非如此。
放到 -data 里的确可以节约仓库的空间啊,如果 mirror 同时提供同种 endian
的几种架构,那么就可以节省一些空间。而且如果用户同时使用 iiimf-le-sunpinyin-chinese 和
scim-sunpinyin,那么用户没有必要同时保存两份数据。尽管这个可能性不大。

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFIHw1PC8hyx6EmTc0RApiQAKCl4LqzihlObmuaKXSfv4mmH29nogCghP2x
oOvIJJXAJBuIxSArmghbZR8=
=bJ0e
-END PGP SIGNATURE-


Re: Fwd: Bug#478811: ITP: sunpinyin -- An input method engine based on SLM

2008-05-05 文章 Kov Chai
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1



2008/5/5 ZhengPeng Hou :
  config.status: linking ./data/lm_sc.t3g. to data/lm_sc.t3g
  config.status: error: ./data/lm_sc.t3g.: file not found
  Debian sid

我的机器是 i386。
config.status: creating config.h
config.status: linking ./data/lm_sc.t3g.i386 to data/lm_sc.t3g
config.status: linking ./data/pydict_sc.bin.i386 to data/pydict_sc.bin
config.status: linking ./src/config/linux.h to src/host_os.h
config.status: linking ./src/config/i386.h to src/host_cpu.h

ZhengPeng,谢谢你告诉我这个问题。我猜你的机器很可能是 ia64。如果是的话,这是由于我们的 configure.ac
没有考虑这种情况造成的,我们仅仅对 i*86* 和 *sparc* 进行了处理。我正在上传新的 sunpinyin package
[1],在它里面改正了这个错误。就是把 i*86* 改成了 *86*,这样就能匹配 x86_64 了。

[1] dget
http://mentors.debian.net/debian/pool/main/s/sunpinyin/sunpinyin_1.0~hg20080501-1.dsc

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFIHxo+C8hyx6EmTc0RAslyAKC5Up9sMH1OirLGtbiGVZ1PTrVPAwCgxTHh
EhSNDtIrmukI25x+ePWGULw=
=D/Ij
-END PGP SIGNATURE-


Re: Bug#478811: ITP: sunpinyin -- An input method engine based on SLM

2008-05-05 文章 Kov Chai
2008/5/5 ZhengPeng Hou [EMAIL PROTECTED]:
 On Mon, May 05, 2008 at 02:04:31PM +0800, Kov Chai wrote:
  
  
   2008/5/5 ZhengPeng Hou [EMAIL PROTECTED]:
  
   --12:25:37--
   http://mentors.debian.net/debian/pool/main/s/sunpinyin/
   sunpinyin_1.0.orig.tar.gz
 = `sunpinyin_1.0.orig.tar.gz'
 Resolving mentors.debian.net... 64.79.197.109
 Connecting to mentors.debian.net|64.79.197.109|:80...
 connected.
 HTTP request sent, awaiting response... 200 OK
 Length: 39,363,185 (38M) [application/x-gzip]
   38M? 都是什么啊
  
  
   38M 是有点大。主要是 data 目录下的四个文件。 lm_sc.t3g.{sparc,i386} (6727K*2),
   pydict_sc.bin.{sparc,i386} (23M*2)。前者是线索化后的语言模型数据[1],目的是加速

  查找速度和压缩数据,差不多就是建立索引的效果,有了它能较快地计算 n-gram 语言模

  型里一串单字 (S = {W_1,W_2, W_3, ..., W_n}) 成为该语言模型里一个句子的概率 P(S)
   。后者则是词表(lexicon),或者说是大家常说的输入法的词库,这个词表支持不完全拼
   音和词到词的转换。由于要支持 big endian 和 small endian 的体系架构,所以干脆就
   把两种情况下的数据文件都弄上来了。
  这些data也是在LGPL and CDDL下发布的?
是的,这个问题我向 upstream 咨询过。本来很早就想把 SunPinyin 加入到 Debian 里去,就是因为 CDDL 的问题迟迟没有动手。

  有没有必要把数据文件单独作为一个 package,放到 sunpinyin-data 里面去呢?还是有
   更好的办法呢?
  
  建议拆分
谢谢你的建议,我周末看看怎么做。这是我第一次做这么复杂的 deb 包。:)


  侯正鹏

 -BEGIN PGP SIGNATURE-
  Version: GnuPG v1.4.6 (GNU/Linux)

  iD8DBQFIHrAbJrhDLBNiZtwRAjWdAJ4uoJ5Q9K9EJiHE+CL1LCHJmtCMwwCff0K6
  S/nXet9Ad67xjoTmgjrdIXQ=
  =wAW/
  -END PGP SIGNATURE-




subscribe

2008-05-05 文章 Kov Chai

-- 
Regards,
Kov Chai
2008.05.05 Mon

--
The real mathematician is an enthusiast per se. Without enthusiasm
no mathematics.
-- Novalis


signature.asc
Description: Digital signature


Bug#478811: ITP: sunpinyin -- An input method engine based on SLM

2008-05-01 文章 Kov Chai

Package: wnpp
Severity: wishlist
Owner: Kov Chai [EMAIL PROTECTED]


* Package name: sunpinyin
  Version : 1.0
  Upstream Author : Lei Zhang [EMAIL PROTECTED] and others
* URL : http://www.opensolaris.org/os/project/input-method
* License : LGPL/CDDL Dual license
  Programming Lang: C++
  Description : An input method engine based on SLM

SunPinyin is an Chinese input method engine based on SLM (Statistical
Language Model). 
.
SunPinyin now provides input modules for IIIMF (Internet/Intranet
Input Method Framework), SCIM (Smart Common Input Method) and BeCJK.

Homepage: http://www.opensolaris.org/os/project/input-method

-- System Information:
Debian Release: lenny/sid
  APT prefers unstable
  APT policy: (500, 'unstable'), (1, 'experimental')
Architecture: i386 (i686)

Kernel: Linux 2.6.24-1-686 (SMP w/2 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash

-- 
Regards,
Kov Chai
2008.05.01 Thu


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



RFS: sunpinyin -- An input method engine based on SLM

2008-05-01 文章 Kov Chai

Dear mentors,

I am looking for a sponsor for my package sunpinyin.

* Package name: sunpinyin
  Version : 1.0-1
  Upstream Author : Lei Zhang [EMAIL PROTECTED] and others
* URL : http://www.opensolaris.org/os/project/input-method
* License : LGPL/CDDL Dual license
  Programming Lang: C++
  Description : An input method engine based on SLM

It builds these binary packages:
iiimf-le-sunpinyin-chinese - Sunpinyin input method of Chinese Language Engine 
for iiimf-server
scim-sunpinyin - Sunpinyin input method of Chinese Language Engine for SCIM

The package appears to be lintian clean.

The upload would fix these bugs: 478811

The package can be found on mentors.debian.net:
- URL: http://mentors.debian.net/debian/pool/main/s/sunpinyin/
- Source repository: deb-src http://mentors.debian.net/debian unstable main 
contrib non-free
- dget 
http://mentors.debian.net/debian/pool/main/s/sunpinyin/sunpinyin_1.0-1.dsc

I would be glad if someone uploaded this package for me.

SunPinyin is an Chinese input method engine based on SLM (Statistical
Language Model). 
.
SunPinyin now provides input modules for IIIMF (Internet/Intranet
Input Method Framework), SCIM (Smart Common Input Method) and BeCJK.
.
SunPinyin is the only free Chinese input method engine using a mature SLM 
of back-off tri-gram language model. The upstream also provides a suite 
of utilities to train the SLM using raw corpus.

Kind regards

-- 
Regards,
Kov Chai
2008.05.02 Fri



signature.asc
Description: Digital signature


Re: fortune-mod 软件包介绍的翻译

2006-08-13 文章 Kov Chai

On 8/11/06, Carlos Liu [EMAIL PROTECTED] wrote:

I think fortune cookie can be translated as 签语饼. Please have a look
on fortunes-it [1]. My one isn't a good translation, but we can
improve it later.

签语饼不错。谢谢你给的链接。我又 google 了一下,看来大家都是这样用的。


警句和俏皮话 isn't enough to express the meaning, fortunes-xx also include
many offensive messages.

如果要根据 fortunes-off 和其他 fortunes 包的内容来翻译的确比较难。我查了一下 Merriam-Webster
Online Dictionary 和手头的大部头牛津英汉双解辞典,它们的释义都没有提到 epigram
有侮辱性词语的意味,而多是指短小精悍、机智、讥讽的只言片语。
我想就从字面上的意思翻译即可。


On 8/10/06, Save Chai [EMAIL PROTECTED] wrote:
 大家好,

 有这么几句,我打算这么翻译:
 * Description: provides fortune cookies on demand
 根据需要提供好运饼(fortune cookies)
 * The fortune package displays epigrams ...
 fortune 软件包能显示一些名言警句和俏皮话

 fortune cookie [0] 是北美中餐馆常提供的点心,但是在国内我没有看到有类似或者能对应上的食品。我觉得应该保留原文,大家的意思呢?

 [0] http://en.wikipedia.org/wiki/Fortune_cookie


[1] http://ddtp.debian.net/ddt.cgi?desc_id=2575
--
 Best Regards
 Carlos



使用 udev 后,/dev/rt c 会无端消失

2005-09-04 文章 Kov Chai
大家好,
这两天在使用 mplayer 的时候发现 /dev/rtc 不能读了。而我的 rtc 模块是放在 /etc/modules 
中让系统加载的。她的确也加载了,如下:
$ lsmod|grep rtc
rtc10316  0

$find /dev -name rtc
find: /dev/.static: Permission denied
## 不过 /dev/.static 下面是有 rtc 设备文件的,如果不用 udev 的话,/dev/rtc 就能用。

$fgrep -irn rtc /etc/udev
/etc/udev/rules.d/020_permissions.rules:55:KERNEL==rtc,
MODE=0660, GROUP=audio
/etc/udev/compat-full.rules:24:KERNEL==rtc,   SYMLINK+=%k
/etc/udev/compat.rules:24:KERNEL==rtc,SYMLINK+=%k
/etc/udev/devfs.rules:83:KERNEL==rtc, NAME=misc/%k
/etc/udev/permissions.rules:55:KERNEL==rtc,   MODE=0660,
GROUP=audio

但是如果
# rmmod rtc  modprobe rtc

$ ls /dev/rtc
crw-rw  1 root audio 10, 135 2005-09-04 23:08 /dev/rtc

$ cat /proc/sys/dev/rtc/max-user-freq
64

这下子 /dev/rtc 就出来了。请问这是怎么回事儿呢?我拜了一下 google,找到的大多是说权限的问题,没啥关系。希望大家能帮我瞧瞧,指点一二。
谢谢啦。

-- 
Regards,
Tchaikov
2005.09.04 Sun

--
The mathematician has reached the highest rung on the ladder of human 
thought. 
   --   Havelock Ellis


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]