I'm a Japanese localizing engineer who now tries to generate
a WebHelp with a nicely sorted index in Japanese.
As I assume that this has been a known issue for double-byte
languages for a long time, a Japanese index cannot be sorted
perfectly in a compiled chm or WebHelp. In Japanese, there are
several types of Japanese characters (Kanji, Katakana, Hiragana).
Regardless of what type of character a string is typed in, Japanese
strings should be sorted according to <i>yomi-gana</i>,
the way each Japanese string pronounces but currently an index gets
sorted according to the following order (in ASCII code):
*kanji... and so on.
So here I'm trying to do the followings:
1. In a HHK file, I put <so>...</so> in front of
every entry where I spell out the pronounciation in each <so>
2. Open the HHK with HTML Workshop, sort the file, and save
it (I'll get the file sorted according to what I have put in
3. Open it with a Text Editor and remove all the
4. Put the HHK file in a build folder and generate a
chm/WebHelp in RoboHelp X3.
In the 4th step, I don't want RoboHelp to re-sort the HHK but
it does it automatically. If I can disable the index sorting
functionality in RoboHelp X3 (the latest version in Japanese) but
looks like there is no way to do it. If anyone is sure that it's
not possible to disable the auto-sorting functionality in RoboHelp,
please let me know so that I can give up witout a sweat.
By the way, I have tried the alternative for WebHelp that
skips the 3rd step and removes all the <so>...</so> in
the files that RoboHelp creates. The result is that everything got
messed up and some of the contents in the Index couldn't be viewed
in a browser.