Hi Mathias,
Mathias Bauer wrote:
Hi Sophie,
Sophie wrote:
Hi Mathias,
Mathias Bauer wrote:
Sophie wrote:
Hi Eike,
In an exchange with Goran this summer, you discussed the support of
RFC4646 tags and ODF 1.2 attribute enhancements to add more than one
language variant or dialects. This is not supported currently in OOo.
You said also that you need time to workout a concept to pass tags and
script to API structure.
Do you have an idea of when this support will be added to OOo? It's not
to put pressure on your shoulders, only to have an idea :)
The issue we face in FR project is different than Goran one, this is we
have more than one spelling allowed for a same word and need to let the
choice of two spellcheckers to the user.
That looks as if it was something completely different. Maybe Metadata
would be a solution for that problem?
How would it work? For me spell checkers are dependent from the locale
and OOo does not allow for more that one language variant per locale.
Yes, in OOo the spellchecker (or the dictionary) depends on the locale
only and that is the problem for e.g. Serbian or Mongolian where it also
should depend on the script type (that OOo does not support at all
except for some hard coded scripts with an also hardcoded table which
locale uses which script type). This problem could be solved by
supporting script types and including script types in the selection of
the applied dictionary.
It wouldn't help for your problem with using different French
Dictionaries for different parts of the text that both have the same
locale and the same script type but some different semantics, history or
whatever. This might be solvable with RFC 4646. I didn't read it but I
take your word for that. :-)
That's why I meant that Goran's problem is something different than yours.
Choosing the spell checker based on e.g. semantic or style
considerations is a special "application logic" or "user logic" that is
applied on the text, it's a kind of markup for the text. If OOo allowed
to save this markup somewhere and had an API to access it, a
spellchecker component could use this to make the dictionary that shall
be applied to the text a persistent information. Using ODF Metadata
could be an option as applying semantic information to document content
is one of its primary motivations.
Thank you for your detailed explanation, it's very interesting and help
me to understand how it could runf. Reading the 'if' and 'could' in your
last paragraph about metadata, does it mean that the API and the process
to save the markup somewhere do not exist currently ?
Kind regards
Sophie
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
