Jean-Marc Lasgouttes wrote:
"Abdelrazak" == Abdelrazak Younes <[EMAIL PROTECTED]> writes:
Abdelrazak> Jürgen Spitzmüller wrote:
Abdelrazak Younes wrote:
I understand but I still don't like the scanning... :-)
I don't like it either. But I don't see an alternative yet.
Abdelrazak> We could just provide the feature if the label is not
Abdelrazak> deeper than one-level. If the user wants to deeply bury
Abdelrazak> the label inside nested insets, then he will miss the
Abdelrazak> feature. In other word:
I really do not understand your point. We have a cursor.
The cursor carries way too much information. This was most certainly a
clear improvement over the old code but at this point it is difficult to
do real cleanup. Look at the flurry of LyXText functions. A lot of them
(most of them) do touch much more than the contents of the _current_
LyXText, this is no good.
We have a
method giving the right information.
I do not see why you want to find
a suboptimal method which only interest is to fit some theory of
yours.
My theory is simple: use what we have access to. Scanning all the
enclosing inset in search for an information that you could have at
first hand _is_ suboptimal in terms of CPU cycle.
Kind of a solution in search for a problem.
No, if you really want to scan through all the enclosing insets it is
still possible with a recursive call to parent(). It will be less code
than the inset scanning and more efficient too.
Abdel.
