Hi, Christian,

2013/7/9 Christian Wittern <cwittern@gmail.com>

Hi Bastien, Yujie

On 2013-07-08 17:22, Bastien wrote:

Hi Yujie,

Yujie Wen <yjwen.ty@gmail.com> writes:

I am wondering whether there is a way to forbid such spaces, or if
no such a way, whether it is possible to work out a patch to forbid
it.

I would try M-x visual-line-mode RET and see if you like it.

Well, this will not solve the problem for the OP, which occurs during export.

What has to be done, I think, is to remove the CR/LF characters for Chinese (and Japanese, Korean etc.) text when exporting to HTML, because they will otherwise show up in the displayed HTML as single space characters.

I used to thinking of removing CR/LF at the parsing or exporting stage. But now I think it is not a perfect solution since it will remove by mistake some meaningful spaces. For example, in an English/Chinese mixed texts whereas lines are sometimes broken at between two English words, which is often the case when we fill graphs with m-q.

Discerning meaningful and meaningless spaces for Chinese and Japanese is rather too complicated a topic. Chinese rarely use spaces for separating words, Japanese sometimes use spaces for separating syntax elements. Korean, by my instinct at staring on Korean scripts, does use spaces to separate words.

I guess, this requires either a derived custom export engine or a internationalization framework that allows the specification of the text language.

Christian

--
Christian Wittern, Kyoto

Regards,

Yujie