Ok , I will explain. I have 2 tasks STR 2838 - "add background colors to Fl_Text_Display (+ underscore + strikethru)" and void getMaxWHXY(int& pPixelW,int& pPixelH,int& pTextX,int& pTextY,int pFont) or "getBestFitFont" using fl_measure().
In Ubuntu 12.04 I have 3 versions of FLTK on my system. (1) the FLTK that came with fluid. (2) ~/fltk-1.3.x-r9359/ which I un tar'd several weeks ago to test STR 2838 (3) ~/fltk-1.3.x-r9625/ which is the latest download. If I test against (1) I try hello.cxx form ~/cpp/ ( of course it "points" to FLTK libs) If I test (2) or (3) (~/fltk-1.3.x-r9359/ or ~/fltk-1.3.x-r9625/) I overwrite test/demo.cxx and make it from each root. The code changes for STR 2838 still had to be embedded into r9625, so I put #define FLTK_ABI_VERSION 10302 into Fl_Text_Display.H (as well as the (FLTK_ABI_VERSION 10302) changes) and got compile errors in some other modules. Some grepping found //#define FLTK_ABI_VERSION 10302 in Enumerations.H, so I uncommented the above line, deleted the line in Fl_Text_Display.H, forced the STR 2838 code in, and clean compiled everything(including STR 2838). I also (as a test) left the comment in (Enumerations.H) The results : NO font problem with 1) OR 2) FONT PROBLEMS WITH 3) with or without #define FLTK_ABI_VERSION 10302 ergo the CHUCK exists in R9625 I can't remember where I looked recently, but consider that I recall an STR with something like "extract redundant (dead) code from fl_measure()". This won't worry me because I'll use R9359 to continue my own testing, but I think this is a problem that someone should look at. Assuming of course that this odd behavoir is not already known. David _______________________________________________ fltk-dev mailing list [email protected] http://lists.easysw.com/mailman/listinfo/fltk-dev
