Linux Input Method settings

There are many articles on Chinese input on the Internet. Here I just want to streamline some concepts.

1. Input Method platform:
In Windows, the input method and the input method platform seem to be unclear. In Linux, their concepts are well separated. The input method platform is the basis of the input method. Sometimes you have installed an input method, but it cannot be called, it is possible that the platform it needs has not yet been set up.
Input Methods platforms:
Xcin (X-chineseinput );
Iiimf (Intranet/Internet Input Method Framework );
Chinaput;
Ami;
Xim (X-inputmethod );
Currently, many Linux systems use xcin, but unfortunately, fedora uses iiimf. You can add an intuitive concept: Start gedit in fedora and right-click the file to select the input framework, some friends say that in fedora, CTRL + space cannot call the input method. You can select iiimf here. I think the reason is that some software changes the default iiimf used by fedora to xcin.
2. Input Method:
In the http://www.linuxfans.org has a Chinese Input Method software column, well received: fcitx, scim, skim, Red Flag input method and so on.
Fcitx: not based on the iiimf platform, so Ctrl + space in fedora cannot be called unless you modify the configuration file (/etc/gtk-2.0/GTK. immodules), or set the shell variable "gtk_im_module = xim" to start your input method platform as xim by default;
37c37
<"Xim" "X Input Method" "gtk20" "/usr/share/locale" "Ko: Ja: ZH"
---
> "Xim" "X Input Method" "gtk20" "/usr/share/locale" "EN: Ko: Ja: ZH"

Then modify/etc/X11/xinit/xinitrc. in the D/xinput file, configure the default startup input method. Alternatively, you can only modify the two shell variables "xmodifiers" and "xim" and enter (or put them in your own ~) in the terminal ~ /. Bashrc ):
Export xmodifiers = @ im = fcitx
Export xim = fcitx

Unfortunately, several other input methods, like fcitx, are not based on the iiimf platform;
Also, scim input method also needs their own scim platform support, corresponding files: scim-chinese-0.4.1-1fc2.i386.rpm, scim-1.0.0-1fc2.i386.rpm;
It is even more regrettable to tell you that currently only the built-in Pinyin is supported for the Chinese Input of iiimf, which is called iiimf-le-inpinyin. Therefore, if you want to use the preceding nice input methods, make preparations for changing the platform.
If you want to track RH tests on iiimf: http://www.apac.redhat.com/iiimftest/
A must-read article about Chinese input under fedora: http://www.erning.net/zh_CN/archives/000002.html

3. Character Set)
Character Set and font are also a pair of concepts that must be clarified in Linux. In short, charset is responsible for interpreting the 1 and 0 sequences in bytes in the file according to the agreed standard, and the font is responsible for displaying the beautiful shape of a word to the user.
Fedora uses en_US.UTF-8 (English), zh_CN.UTF-8 (Chinese) as charset by default; use the "locale" command to see the variables related to this, of course, they can be changed; I have encountered the following problems:
After eclipse3.0 is installed, the default charset is changed to zh_cn.gb01830. Of course, this is a very good character set. However, many default character sets in ora, such as the toolbar, window, title, and dialog box, are not, garbled characters are displayed.
Changing the default charset is not difficult, yesterday I only used two commands, changed from zh_cn.gb01830 to zh_CN.UTF-8:
<1> terminal: locale >> ~ /. Bashrc
<2> open the. bashrc file with gedit, replace all zh_cn.gb01830 as zh_CN.UTF-8, such:
......
Export lang = "zh_CN.UTF-8"
Export lang = "zh_CN.UTF-8"
Export lc_ctype = "zh_CN.UTF-8"
Export lc_numeric = "zh_CN.UTF-8"
Export lc_time = "zh_CN.UTF-8"
Export lc_collate = "zh_CN.UTF-8"
Export lc_monetary = "zh_CN.UTF-8"
Export lc_messages = "zh_CN.UTF-8"
Export lc_paper = "zh_CN.UTF-8"
Export lc_name = "zh_CN.UTF-8"
Export lc_address = "zh_CN.UTF-8"
Export lc_telephone = "zh_CN.UTF-8"
Export lc_measurement = "zh_CN.UTF-8"
Export lc_identification = "zh_CN.UTF-8"
Export lc_all = "zh_CN.UTF-8"
......
Of course, this is reckless, but it is very simple and effective.