Msg: 5985 *Conference*

08-23-95 17:46:13

From: RON WIESEN

To : RON WIESEN

Subj: REPLY TO MSG #5983 (UNDELETE -- COLD START RECOVERY)

The RAM hook for WIDTH, which I mentioned in the prior message, is not related
to the "Width:" prompt the OS gives in TEXT response to SHIFT+PRINT.  The WIDTH
hook relates to an OS extension to Disk-BASIC (Disk/Video Interface) command
WIDTH which is used to set the screen width.  So "Width override" code can't be
linked; so Width can't be restricted; so immunity from encroachment into the
"save area" can't be guarenteed.  Bummer!
 
There is a RAM hook for TEXT and it's a post-invocation that's linked to F8
(Menu).  If "establish Width" code is linked to it, the Width default (ASCII
digit string at F64A) can be set; so immunity from encroachment into the "save
area" can be established each time TEXT is exited.  This is not a guarentee
against encroachment, but it would be rather certain that a cold-start would
not occur while TEXT was active.  With "establish Width" code, an overlay of
the directory info in the "safe area" due to PRINT+SHIFT Width in excess of 80
persists to complicate recovery from only a cold-start that occurs after the
exit from TEXT but before the next update of the "safe area" with new directory
info.  Such "establish WIDTH" code uses about 12 bytes of the 16-byte code
region of the "safe area" (F910 through F91F).
 
By the way, experiments show that the "safe area" holds line_strings the OS
builds for general output: LCD and LPT.  Merely viewing an ASCII file alters
content in the "safe area".  Not to worry, LCD output uses 40-character
line_strings and the Disk/Video Interface uses either 40-
or 80-character line_strings.