Closed AR1972 closed 1 year ago
I can confirm this. I spent about one hour trying to compile my code with fill character $FF, nothing worked. After patching with the above code it works if I enter "FILCHR TEXT $FF" in the the code I try to compile. It will however not work if I try to add -dFILCHR=$FF to the command line
OK, I got some time now to look at it.
You're right, the last commit introduced a problem here.
I won't include your solution above, however; the better solution (both for space and execution time) is to change
if (!strcmp(namebuf, "FILCHR")) /* special for FILCHR : set internal */
(some lines down from your fix) to
if (!strcmp(namebuf, "&FILCHR")) /* special for FILCHR : set internal */
... that's what the next version (due maybe today, maybe tomorrow) will contain. It will create a little clash with your solution, however, so watch out.
Oh, wait a moment ... that's what your pull request does anyway. So, no clash, all good :grin:
OK, that should be fixed.
I'm using "FILCHR TEXT $3F" in my project, last commit always uses 0x00 as the FILCHR adding this hack to settext() returns previous behavior.
if (memcmp(namebuf, "FILCHR", 6) != 0) { / to keep texts separate from the normal labels' way, prepend them with & / if (namebuf != '&') { sprintf(iname, "&%.s", (int)(sizeof(iname) - 2), namebuf); namebuf = iname; } }