I just checked using Virtual-T. The buffer always ends at F307 with a 1A(EOF). You need to find the start of the buffer and then read it in.
On Fri, Jul 19, 2019, at 11:22 AM, Dan Higdon wrote: > Peeking the paste buffer and scanning for EOF would definitely work for > reading the buffer. The writing of data into the paste buffer seems like a > harder problem. > > Might using the "make room" function (or whatever its real name is, I'm no > near my docs right now) be capable of resizing the fake file to hold new data? > I'm imagining a machine language routine that can replace/append a string > onto that file. > > On Fri, Jul 19, 2019 at 1:11 PM John R. Hogerhuis <jho...@pobox.com> wrote: >> >> >> On Fri, Jul 19, 2019 at 10:15 AM Dan Higdon <therealh...@gmail.com> wrote: >>> So maybe it's impossible to access the paste buffer from inside a basic >>> program after all, except trivially when the user presses <paste> during an >>> input statement. >>> Well, if I figure something out one day, I'll share it here. >>>> >> >> Maybe that's your solution. Poke PASTE into the keyboard buffer and run an >> INPUT command. >> >> Though you have a phasing issue, since INPUT wants to block. If you do the >> INPUT command first, you'll never get to the POKEs. If you POKE first, the >> paste will happen while INPUT isn't started. Maybe the characters just sit >> in the keyboard buffer until an input command runs in which case it would >> work. >> >> I don't know much about the PASTE buffer. According to Inside the Model 100, >> it is the hidden "Hayashi" file. That's more than 6 characters so I guess, >> HAYASH >> >> But Tandy removed the names in some version of the Rom. >> >> paste buffer pointer is at F88C >> >> So you cannot open the file by name. >> >> You might be able to pass the address to a low level part of the file open >> command and call it. Not sure. >> >> -- John.
