>>>>> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
Angus> Jean-Marc Lasgouttes wrote: >> Concerning the signature, we should leverage boost::crc32. Two >> ideas of text to compute against: >> >> - the easy one: take the text of the row (including META_INSET >> characters) as a string to compute the crc on, and add the y >> information somehow (we could maybe keep a <crc,y> pair as >> signature). This is crude, but uniformly better than what we have. >> >> - the expensive one: generate the .lyx format code to the range in >> the row (the changes to lyx writing code to make this happen are >> hopefully small) and compute crc on this. This may prove less >> expensive than it seems :) Of course, in a world where we hope to >> be able to update insets without updating the hosting row, it will >> redraw too much. Angus> why not add an Inset::crc member function that would recurse Angus> through nested subinsets? I am not sure it would be better than outputting the .lyx code for the document (this has the added bonus of taking fonts in account, for ex.). But my problem is that I am not sure we want to recurse into insets to decide whether the current row shall be repainted (especially in the future when we will ask the inner insets to handle themselves). So for now, I suspect that something simple will be enough. JMarc
