>>>>> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
Angus> Andre Poenitz wrote: >> On Fri, Sep 19, 2003 at 11:23:22AM +0000, Angus Leeming wrote: >>> InsetInclude should no longer creak at the joints and is now >>> buffer agnostic, as it should be. However, it strikes me that it >>> performs two separate tasks: >>> >>> 1. Include a LyX file. Lots of code to extract info from this LyX >>> file. >>> >>> 2. Include a LaTeX file. Output 4 different 'flavours' of latex, >>> \include, \input, \verbatimitem, \verbatimitem*. >>> >>> It could be simplified enormously if we moved this latter >>> functionality over to the External inset. Shall I do this? >> I have no clue what exactly you are proposing, but if it >> simplifies things I surely won't try to stop you... >> >> [Don't waste time to try to explain it to me...] Angus> Don't be daft. Of course I explain: Angus> * InsetInclude should include LyX files only. Stored data: the Angus> name of the LyX file. Nothing else. Currently, it can either \input or \include LyX files. Although I understand that many people do not what the difference is, the possibility is here. Concerning the simplification argument, I am not sure that having 20 different external inset in a list is really a better UI than a dialog inset that we can arrange to our taste. You are proposing to add LaTeX File LaTeX File (include) Verbatim File Verbatim File (with visible spaces) to the list (or something like that). Think about the 10 other things that we will want to add if we continue on this route, and I am not sure that it will make the user's life easier. As far as the code reduction is concerned these are a few dozens of lines of trivial code, which do not cause any maintainance headache, as far as I remember... So my vote is on the good UI. If you can prove that it will be simpler, then it could be a good move. JMarc
