On Thu, May 03, 2018 at 10:14:32PM -0400, Mario Beaulieu wrote: > Hi, > > The problem I'm having might not have anything to do with GForth, but let's > see. I'm running GForth on a Raspberry Pi (Linux). > > I have a GForth program (notification.fs) that sends notifications to my cell > phone. The problem I have is that it replaces accented characters (à, è, é, > etc.) with question marks but only if I run it from the command line , like: > GForth notification.fs > > What the file notification.fs does, is compile the necessary words, the main > one being SendNotification, and the last line sends a message, like: > s" àéèeeeee" SendNotification > > If I run it from the command line, it replaces the accented characters, but > if I start GForth first and paste the content of notification.fs, it works > perfectly. > > Even weirder, if I 'redo' the last line right after running it from the > command line, it works! > > The character encoding is UTF-8. I tried with ISO8859-1, with the same result. > > Any idea?
I have no idea what's going on, but you can see what's going on by doing a 2DUP DUMP before sending the string. Then you see the bytes that are sent. A wild guess is that the string is encoded in some other encoding in the file, which the editor recognizes, and there is some conversion going on in the cut-and-paste process. I am not sure what you mean with 'redo'. - anton
