Robert Roessler: While the 100->101 change was performed on December 4, the mailing list messages about this were in late October when the changes to Editor.cxx were committed. Looks like the corresponding Document.cxx weren't committed when they should have been.
http://mailman.lyra.org/pipermail/scintilla-interest/2005-October/006736.html http://mailman.lyra.org/pipermail/scintilla-interest/2005-October/006731.html > Your second two are potentially more troublesome (and I already had > decided that they should be the way they were). Trivially, they > could/should be done using the sixth parameter mentioned above... but > worse, I am not sure you want to. According to ScintillaDoc.html, > line is "0 if unused" - and I think this is more consistent (even > though you might say that 0 is misleading, being a valid line#). Where these are -1, they are being used as they are indicating that no particular line can be identified as the subject of the modification. The modified meaning should have been made more of and should be documented. > More scarily, and going by your "let's not break existing apps", > people could have written code that "knows" about the 0-value on > SC_MOD_CHANGEMARKER, and *uses* it, knowing it is harmless (I know I did). > > So to sum up, I would be happy seeing this changeset backed out - but > that's just me, and I haven't heard your thoughts on the subject. :) The failure to draw markers needed to be fixed. http://mailman.lyra.org/pipermail/scintilla-interest/2005-October/006684.html > As an aside, since I had worked on this code, I would have liked to > have known about the change when it happened and brought this up then > - I do not suppose there is any mechanism like this in CVS? Something > like "email me when something is checked in that affects this range"? Some people do set up such things but I don't know how to. Neil _______________________________________________ Scintilla-interest mailing list [email protected] http://mailman.lyra.org/mailman/listinfo/scintilla-interest
