Sign up on the Revelation Software website to have access to the most current content, and to be able to ask questions and get answers from the Revelation community

Window/Table mode broken (AREV Specific)

At 18 AUG 2002 01:24:38PM Gary Gisondi wrote:

Ctrl-F5 Window/Table View key is not working in my ARev 3.12 installation. Since BROWSETABLE never really worked w/o problems, we long ago disabled Ctrl-F5 in all apps. After upgrading to 3.12, I thought it was time to revisit the feature but was stopped short:

Ctrl-F5 is now returning message W559:

"Cannot continue. An attempt was made to execute a null ("") program."

Wondering if BROWSETABLE was updated in 3.12, I found the following, possibly corrupt text at end of $BROWSETABLE code in REVBOOT\SYSOBJ. Note that a ctrl-L (ascii decimal 12) character is embedded in the the reference to DICT.TABLEBROWSER. Ctrl-L won't display here so I show it below as {12}; other control chars in original are removed.

"Copyright (C) 1992,93 Revelation Technologies, Inc. 0299

Reading Dictionary Information For %1% BrowseTable - %1 @CRT @STATUP DICT.VOC FIXATR HF DICT.{12}TABLEBROWSER

AREV.BP

BROWSETABLE

VAREV*3.1.73"

I have no TABLEBROWSER or DICT.TABLEBROWSER files visible in my system. Is the embedded Ctrl-L legitimate? If not, then how to recompile unavailable SYSOBJ source?

Otherwise, can anyone help me re-enable the Window/Table VIEW toggle?

Any help much appreciated,

Gary Gisondi


At 18 AUG 2002 03:35PM Richard Hunt wrote:

Gary,

I can't help you with yur problem.

I do know that the CTRL "L" is correct. You are looking in the "literal string variable" section of the compiled code. The CTRL "L" is actually saying that the string variable (TABLEBROWSER) is 12 characters long.

And that is all I really can say to help you with your problem.


At 19 AUG 2002 02:34PM David Kafka wrote:

Gary,

That looks the same as the object code in my system.

The 3.12 table browser is very picky. For one thing, AFAIK it will break to the debugger if there is not an active select list (e.g. if you use CTL-F5, be sure you have a record loaded). I would suggest that you start by painting a very simple window on a table, then try to run the table browser on that window. If that works, go from there.

David


At 19 AUG 2002 02:58PM [url=http://www.sprezzatura.com]The Sprezzatura Group[/url] wrote:

This error message is actually more typical of a bug in a custom MFS. Is there an MFS on the table?

The Sprezzatura Group

World Leaders in all things RevSoft


At 19 AUG 2002 08:20PM Gary Gisondi wrote:

Thanks for the suggestion, but no MFS's on these tables except SI.MFS for indexing. Getting same results for tables without indexing.

W559 message appears with or without an active select list in the app window. When I acknowledge message, window closes and calling menu redisplays without breaking to debugger.

Since behavior is global in all apps/accounts/ windows/tables, could there be some dependent environment settings, say in Concurrency or Global Environment? My current settings for latter:

Commit Log Volume TRANSACT

Commit Protection (Y/N) Yes

Deadlock Resolution Scheme Least Active

SQL Implicit Locking Record

Coordinated Locking No

Or possibly a dependent subroutine whose cataloged name was updated from a dot to an underscore but whose code/command reference did not? Still, would have to be global, not template specific.

TIA,

Gary


At 19 AUG 2002 08:41PM Gary G. Gisondi wrote:

Well, I created a new basic window using QuickPaint… but got same result after loading some records and pressing CTRL-F5:

'Cannot continue. Attempt was made to execute a null ("") program.

Ok'

Application closed and menu reappeared, as with all previous windows.

Thanks for suggestion. It was worth trying.

Gary


At 19 AUG 2002 08:44PM Gary Gisondi wrote:

Thanks… at least good to know that record's not corrupt.

-gary

View this thread on the forum...