Re: [Texmacs-dev] string encoding

[Joris van der Hoeven]

> > What we should start with is building dictionaries between the current
> > TeXmacs encoding and unicode and Html. Would you be willing to help on this?
> 
> I had a look at the dictool. How can I build an encoding dictionary,
> i.e. how to enter character codes? Should I escape them using \ as in
> "\032"? Which file format is best for this: .dic, .scm?

We currently still use the .enc format (see fonts/enc),
but we should move towards the .scm format. It would be good to add
support for the .scm format in src/Resources/Translators,
as well as routines for passing from one encoding to another.

The next step is the support of most standard encodings.
The most important ones are Html, ISO-8859-* and Unicode.

> Am I right in assuming that the dictionary should map texmacs encoding
> -> unicode and we then encode the unicode characters of the form
> 0x******** to utf-8 sequences by means of some c-function? Does the \
> escape sequence work for values > 0xff ?

You can use #hexcode.

Andrey: I think that the hexcode should not be a string,
as is currently the case, but rather a symbol or a number.
Could you think about this? When loading a translator,
we should be able to interpret a character like #8f
as a string of length 1.