On 2015-09-13, Georg Baum wrote:
> Enrico Forestieri wrote:
> I like it very much. I did not test it, but if the autocompletion
> problem does still exist after applying it, then we should implement a
> general mechanism to restrict autocompletion in mathed to the currently
> active mode: If inside a text box, only propose text commands, and if
> inside a math box, only propose math commands.
Not only autocompletion, but also the GUI replacement are still active for
the text mode commands in mathematical mode.
The conversion of text commands to symbols should be:
* not restricted to the 20 symbols now defined in lib/symbols
(out of ca. 1500 text commands defined in lib/unicodesymbols).
* not be restricted to text-in-math but also be accessible in text mode
without the result beeing exported as $\text{...}$.
* not double the effort from lib/unicodesymbols in lib/symbols.
This already works in some limited cases:
a) Edit>Special-Insert>from LaTeX
b) text-in-math -- however with a strange limitation:
* open a new document
* open a math box (Ctrl-M)
* open a text in math box (\text)
* insert a text command, e.g. \"A or \ss
The command appears as ERT.
* save
The command is converted to the symbol.
Commands inserted after the first save are only converted after closing and
re-opening the file.
A proper support for text commands (LICR macros) should focus on these
features already in place, not on the 20 text commands defined in lib/symbols.
Günter
