Hans-Christoph Steiner wrote:
On Dec 13, 2005, at 10:37 PM, Hans-Christoph Steiner wrote:
On Dec 13, 2005, at 10:53 AM, geiger wrote:
On Tue, 13 Dec 2005, vincent rioux wrote:
i really love utf (far better than iso-8859 and the like), but
outside OpenOffice it is still often unusable (at least on linux; i recon
that w32 works fine with utf8 and osX might do as well)mfg.asdr. IOhannes
ok, then i totally agree with you, the tricky part being /can unicode strings be integrated into a .pd (text file) relatively simply even if not directly editable with usual text editors? / after all, unicode characters can be represented by escape codes
(ascii characters preceded by ), isnt' it?by the way i would be interested to know how many people are editing their patchs without the patch/cord gui and for what reasons? for my part, i do it very seldom, only when i need to correct a patch that cannot be opened any more.
An easy thing that one can do is replacing the "isprint(n)" by "n > 31" in g_rtext.c and then recompile pd. On linux this
produces extended ascii characters. Don' t know what it does on windows and OSX, maybe someone is willing to try.Works for me on Mac OS X! Could this simple patch cause any bugs?
Why was isprint() there in the first place?So far no crashes, tho once I typed a character and it hung for a few
seconds, and then recovered and typed an empty box. HmmHere's a little snapshop of the test:
http://eds.org/~hans/Pu%cc%88reDa%cc%88ta%cc%88.pdf
Comments work fine, you can type special characters into object boxes, but you can't load objects or abstractions using special characters. And [print] replaces them with a ?.
I am going to include this in the next Pd-extended, we'll see then what the problems are.
.hc
impressive! one pb i can see is that you might not be able to reproduce it on linux or windows just as well and that it is definetely better to write "espanol", or "francais" when things like "espa?ol" or "fran?ais" (or worse) might come up when you swap platforms.
anyway, this is already a very nice result.
Also, we don't really need to call files (abstractions, libraries, objects) with special characters, this will never really work cross-platform i believe. i sometimes think that a subset of english could work just as unicode does, the difference being that we will never have to learn unicode... or?
vincent