[Search for users]
[Overall Top Noters]
[List of all Conferences]
[Download this site]
Title: | DOCUMENT T1.0 |
Notice: | **New notesfile (DOCUMENT.NOTE) now available (see note 897)** |
Moderator: | CLOSET::ADLER |
|
Created: | Mon Feb 09 1987 |
Last Modified: | Thu Oct 31 1991 |
Last Successful Update: | Fri Jun 06 1997 |
Number of topics: | 897 |
Total number of notes: | 4397 |
642.0. "Troubles with fixed length records" by RTOADC::CAAHANS (Hans Bachner, ADG Munich) Mon Jul 13 1987 05:30
Does DOCUMENT accept only files with variable record length format?
We have a (fast-hack) program that extracts info from a data file,
inserts a few DOCUMENT tags (mainly <p>s) and writes 132 byte records
to a file. If DOCUMENT (FT Update) is invoked with this input file,
we get the following weird messages:
%TAG-F-ERROPEINP, at text on line 0 in file
DISK:[DIRECTORY]JUNK.SDML;
Cannot open input file:
DISK:[DIRECTORY]JUNK.SDML;
-RMS-S-NORMAL, normal successful completion
-SYSTEM-W-IVQUOTAL, invalid quota list
It puzzles me somewhat that
- a RMS success message converts to a DOCUMENT fatal error
- a message code returned by $CREPRC is quoted to be the initial
error
BTW, if the file is NOT created with the fixed record length attribute
(or converted to variable record length), DOCUMENT processes the
file as expected. So this behaviour does not stop our DOCUMENTing;
however, you may add it to your buglist.
Hans.
T.R | Title | User | Personal Name | Date | Lines |
---|
642.1 | yes, just variable length for v1.0. | VAXUUM::KOHLBRENNER | | Mon Jul 13 1987 12:23 | 17 |
| The tag translator looks at the status returned from an open
call. If it is greater than 0, the tag translator puts out the
fatal error message. This is probably too simple an approach.
It appears that opening a fixed length record file results
in some kind of "error" return that may actually be
acceptable, if we bothered to look at it and filter it out of
all the other more serious error returns. You could probably
read a relative file or an indexed sequential file as well.
I'll put this on the list of things to investigate for v1.1.
All the files that we ever tested must have come from editors
or "hack" programs that wrote variable length record sequential
files.
bill
|
642.2 | See Pascal notes too | IJSAPL::KLERK | Theo de Klerk | Wed Jul 15 1987 04:36 | 8 |
| I recall a discussion on problems with Pascal opening files
using fixed length text records. This was in the context of
a question how to make a "general purpose opener" routine
that would accept all sorts of files...
No doubt John Reagan of the Pascal group may shed some light
on this.
Theo
|
642.3 | good pointer | VAXUUM::KOHLBRENNER | | Wed Jul 15 1987 14:26 | 2 |
| thanks, Theo. bill
|