Lars Gullik Bjønnes a écrit :
Abdelrazak Younes <[EMAIL PROTECTED]> writes:
[...]
| IMHO, this is because the code is not object oriented. In this context | docIterator is an advantage. The code is not 'not object oriented' either.
IMHO, it is not _much_ object oriented. That's just an opinion, not meant as a critic. I know it was re-factored a lot in recent years. Beside code evolves, nobody should take personally critics on his code (I don't).
| When/if we move to a truly object oriented way of dealing with insets | I believe that most of the its use will disappear. Beside updating | _all_ the counters of the document (including inside insets!) because | of one new paragraph seems wrong to me. It sound like you are thinking of something we rejected quite some time ago. But I look forward to see your ideas.
Within this context, no this is my first attempt to optimize updateCounters. You did refuse some other things though. I am fine with that as long as you tell me in advance that my doing things is not acceptable.
[...]
| This is something I don't understand, sorry. IMHO, rewriting is a sane | thing to do in any case if you want more people to understand the | code. Not if this ends up making the people that already understand the code loose that understanding.
Well, if you understand better your code than my patch, I give up.
| There are really too much code of that sort in lyx. Please find | attached a patch that improve the situation dramatically. The | LABEL_ENUMERATE case optimization is not yet enabled but is almost | ready. Passing (Buffer const & buf) is not nice from an OO point of | view but this can be solved later. With a multi paradigm language like c++, pure OO is not the goal.
Did I said that? Genericity is great concept also.
But surely LyX can be improved in a lot of ways.
Indeed. Abdel.
