[Date Prev][Date Next][Subject Prev][Subject Next][
Date Index][
Subject Index]
Re: Wildcard error message
- Subject: Re: Wildcard error message
- From: "Robert Holmgren" holmgren@xxxxxxxx
- Date: Tue, 21 Jan 2003 21:24:16 -0500
** Reply to message from Patricia M Godfrey on Tue, 21 Jan
2003 13:02:40 -0500
Patricia:
> Carl--is there any way to make DREF run with CI, rather than CV? It took
> me a half an hour, and my index finger was lame from pressing Y, to get
> that file cleaned up. I realize CV is safer, but since I always keep the
> original file, both in its original format and in the
> fresh-from-conversion Xy form for comparison, there are frequently
> occasions when I'm willing to take a chance with CI.
DREF doesn't use CI or CV or any such broad-brush whole-hog slap-dash approach.
In fact, given what it is trying to accomplish, to CHange everything it finds
would be madness. DREF *carefully* works its way through the document, one
format/MoDe statement at a time, and *compares* the current environment to the
effect of each individual command (format/MoDe) that it finds. If that command
would not change the environment, i.e. the prevailing condition and the effect
of the command are identical, then that command is deemed "redundant", and user
is offered an opportunity to delete it. One needs to be cautious about hitting
"Y": the current environment is often determined by "presets" that are not
part of the document, but rather of startup.int or settings.dfl, i.e. they are
personal choices and peculiarities. If you take the supposedly "redundant"
format out of the document, you *may* affect its portability to an entirely
different installation, which doesn't share the same presets.
Example: The document states, at TOF, something like .
Your default, established in SETTINGS.DFL, happens also to be
. Ergo, DREF considers that format redundant, and you
zap it. You then give the file to somebody with default
. Goodbye meticulous tab formatting.
That said, a revised DREF with a /NoVerify switch is in the pipeline.
The truth is, Carl and I vet each other's work in almost every instance, and
authorship gets a bit muddy (or muddled) sometimes; so, no, I don't at all mind
being called "Carl" from time to time.
-----------------------------
Robert Holmgren
holmgren@xxxxxxxx
-----------------------------