Re: COBOL
From: Chuck Stevens (charles.stevens_at_unisys.com)
Date: 12/23/04
- Previous message: Howard Brazee: "Re: COBOL"
- In reply to: Robert Wagner: "Re: COBOL"
- Next in thread: Lueko Willms: "Re: COBOL"
- Reply: Lueko Willms: "Re: COBOL"
- Reply: Pete Dashwood: "Re: COBOL"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Thu, 23 Dec 2004 08:53:44 -0800
"Robert Wagner" <spamblocker-robert@wagner.net> wrote in message
news:137ks05k9jnticqq4qihfq5obkmlri1lam@4ax.com...
> Device-specific assignments are obsolete. Nowadays, ASSIGN statements
> take three forms:
>
> 1. SELECT file ASSIGN TO '/path/file'
>
> 2. SELECT file ASSIGN TO data-name (which contains /path/file)
>
> 3. SELECT file ASSIGN TO ddname (environment variable or JCL
> containing /path/file)
Ummm.... In what context? Although exactly *what* a file may be assigned
to is still defined by the implementor, ISO/IEC 1989:2002 specifically
changed the terminology for the non-literal case from "implementor-name-1"
to "device-name".
While IBM may have favored such intutively-clear device-names as
"SYS00C-UR-2540R-S" at one time, Unisys MCP-based systems have gone more
toward such unfathomable mnemonics as "READER" and "DISK" in this context,
believing that more specific assignation than that is rightly more the
purview of the operating environment than of the COBOL program itself.
-Chuck Stevens
- Previous message: Howard Brazee: "Re: COBOL"
- In reply to: Robert Wagner: "Re: COBOL"
- Next in thread: Lueko Willms: "Re: COBOL"
- Reply: Lueko Willms: "Re: COBOL"
- Reply: Pete Dashwood: "Re: COBOL"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]