On Mon, 9 Sep 2013 21:03:03 -0400, Shmuel Metz (Seymour J.) wrote: > on 09/09/2013 at 11:14 AM, Paul Gilmartin said: > >>Then what went wrong? > >You used the wrong tool. > >>Try "ALLOCATE DD(SYSIN) DSN(*) > >Why would I do that for an interactive command? Try EDIT foo.text. > <RHETORIC> Why does DSN(*) exist at all? </RHETORIC> What you suggest is more like batch than interactive.
It was a program which issued prompts to DD SYSTERM and read responses from DD SYSIN. It could run either in batch, in which case the programmer needed to anticipate the prompts and EDIT the file to be used as SYSIN and could ignore the prompts in SYSTERM, or interactively, in which case SYSTERM and SYSIN could both be allocated as DSN(*). But data read from DSN(*) were perniciously folded. I could have required that users detect interactive use and code specially for that. But since I controlled the runtime library, I could do better. At the DCB OPEN exit I stole the access method pointer and pointed it at my code that did a TGET; no fold. It all worked. But why did I have to go to that effort? Why isn't there an ASIS option on ALLOCATE? >And what does that have to do with justifying "Ah! Thermal paper? > Merely that it was the first time I saw a computer (it was a PDP-10) writing messages and column headings in mixed case. "Thermal" is irrelevant; merely an exclamation of recognition of the device. >But that provided me the epiphany that computers could deal with >mixed-case text."? The processing of DA(*) is the same whether the >terminal is a TI "silent" 700 or an IBM 2741. > IOW, identically wrong. When I type "a", I want my program to see that, not "A". If I need to I can fold; I can't unambiguously unfold. -- gil ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
