Re: [O] [PATCH] Table continuation strings

[Yasushi SHOJI]

HI,

At Sat, 21 Dec 2013 10:05:35 +0100,
Nicolas Goaziou wrote:
> 
> Yasushi SHOJI <address@hidden> writes:
> 
> > The thing I don't understand is the reason all Japanese entries have
> > `:utf-8'. Would you kindly enlighten me the relationship among the
> > followings:
> >
> >  - transtion coding key (ie :utf-8, :default, :html)
> >  - your current buffer coding system
> >  - `buffer-file-coding-system' and friends
> 
> Coding keys are related to export back-ends. Therefore :latex entry will
> be used for `latex' export, :html for `html' export, `:utf-8' for both
> text (utf-8) and odt export, and so on.
> 
> As its name suggests, :default key is used as a fallback value when no
> appropriate property is found. It makes up for a handy shortcut when
> some strings are identical.

Ah, OK.  Those coding keys are for the back-ends to select proper
strings, not for the string encoding.

Then, is there any restriction with HTML back-ends? Why does it need
numeric character reference instead of just plain characters, if the
coding system is not a concern?

> Coding system is a different thing. When `org-export-coding-system' is
> non-nil, it will be used as the coding system for output (note that some
> export back-ends override this behaviour). Otherwise, output will have
> the same encoding as the source buffer.

Correct me if I'm wrong.  My understainding is as follows: All
translation strings is in `emacs-internal' coding system, since it is
defined in .el.  A org file ready to be exported has a coding system
specific to the buffer, ie. utf-8, iso-latin-1, euc-jp, etc.

Org export back-ends get a strings for the back-ends from the
translation table when appropriate.  At that time Emacs converts the
strings encoding system to match the buffer encoding system (or does
Emacs convert all encoding when it writes to file?).  Back-ends uses
`org-export-coding-system' if set, otherwise use the current buffer
coding system.


If my understanding is ok, all entries of Japanese translation should
have :default instead of :utf-8.

Thanks,
-- 
            yashi