Re: [Fwd: gnubol: Record delimiter clause and parse order]

[fneale@tss.com.au]

Appendix 'I' of my trusty MicroFocus System Reference manual contains a LOT
of information regarding the structure of the variable length file header
block and of the record prefix (2 bytes for records < 4096 bytes and 4
bytes for longer records).

If you feel this will be of assistance I am happy to take it home this
weekend and type it up for you. Let me know.

Fred Neale





Tim Josling <tej@melbpc.org.au>@wallace.lusars.net on 26/05/2000 07:09:33
AM

Please respond to gnu-cobol@lusars.net

Sent by:  owner-gnu-cobol@wallace.lusars.net


To:   "gnu-cobol@lusars.net" <gnu-cobol@lusars.net>
cc:

Subject:  [Fwd: gnubol: Record delimiter clause and parse order]



Content-Transfer-Encoding: 7bit
X-Mozilla-Status2: 00000000
Message-ID: <392D86E9.E2406665@melbpc.org.au>
Date: Fri, 26 May 2000 06:02:49 +1000
From: Tim Josling <tej@melbpc.org.au>
Organization: Melbourne PC User Group
X-Mailer: Mozilla 4.61 [en] (X11; I; Linux 2.2.12-20 i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Matthew Vanecek <linuxguy@directlink.net>
Subject: Re: gnubol: Record delimiter clause and  parse order
References: <39284A8B.759FE5FF@directlink.net>
<392903C8.9F7D851F@melbpc.org.au> <3929FE98.C448C3DF@directlink.net>
<392BB2C0.BFE22245@melbpc.org.au> <392C7B67.4C63D6D0@directlink.net>
Content-Type: text/plain; charset=us-ascii

Matthew,

Good. So

- line sequential = records followed by "end of line delimiter" =
LF, CRLF, CR depending on the platform. No extra padding at end.
- fixed length = records stacked one after another, no record
header, no delimiter.
- variable - 128 bytes of slop (including date/time stamps by the
look) followed by the records. The record format looks like this

x'40' byte(length) data
The data is padded to make the record length a multiple of 4
bytes. The padding seems to vary, sometimes spaces, sometimes
x'00'.

It would be useful to try variable length non line sequential
with record length

> 127

> 256

>32k

>64k

Without the doc of the 128 bytes we cannot create a file that MF
will read (other than by just copying one of their headers byte
for byte), though we can read their files. I will have a look for
conversion routines.

Maybe we should just use s 4 byte word as the record header (at
least as one option).

Tim Josling





--
This message was sent through the gnu-cobol mailing list.  To remove yourself
from this mailing list, send a message to majordomo@lusars.net with the
words "unsubscribe gnu-cobol" in the message body.  For more information on
the GNU COBOL project, send mail to gnu-cobol-owner@lusars.net.