Jean-Marc wrote: URL: http://permalink.gmane.org/gmane.editors.lyx.devel/84923 > Uwe reopened the bug and added the following comment: > > ------- Additional Comments From > [EMAIL PROTECTED] 2007-05-22 17:26 ------- I do > not agree with the reverting. Why was this necessary? This fix works > fine except for read-only documents. So instead of reverting the > patch, the inset state buffer saving should only be active when the > document can be changed. So why not embed the patch to an if > condition?: > > if (Document != Read-only) > cur.bv().buffer()->markDirty(); > --------------------------------------------- > > Well, this is something I proposed in the aforementioned thread and it > was discarded as not needed.
I don't think, I understood this at the time. Uwe's solution is straight forward and fixes both 2993 and 3503. However, I think Abdel points to something interresting, an inconsistency in terms of BufferView and Buffer (if I understand this correctly). Would it matter to know someones states of insets when you share documents or is this something you do not actually want to know. Perhaps there could be really some setting where general inset states (upon opening) could be defined. Cheers, Sam
