Replace Indexing Routine Not Starting 2nd Round (AREV Specific)
At 16 APR 1998 05:44:32PM Mark Caldwell wrote:
Using version 3.12 of AREV, I have created a Timeout routine that runs as a Replace-Indexing option. (We have a dedicated indexing machine to handle updates). The Timeout routine is called as a window (Code W) that covers the screen then uses scribe for accepting a password to restore the screen.
If someone restores the screen, and then does NOTHING else, not even a keystroke, then the timeout routine does NOT kick back in even after the time elapses.
How do I "reactivate" the indexing timer without forcing a keystroke? (A keystroke could wipe out a critical MSG from the screen).
Thanks in advance!
At 17 APR 1998 09:27AM Aaron Kaplan wrote:
What are the contents of @INDEX.TIME and @BACKGRND.TIME?
It could be something in the post process of the window.
Maybe you can force in a unused macro key which will eat up the buffer and do whatever reset might be required?
apk@sprezzatura.com
At 17 APR 1998 12:56PM Mark Caldwell wrote:
Here's the results of some testing I've done based on your questions…
1) I'm running my tests with @Index.Time=30 and @Backgrnd.Time==60. However, I just put in a message to display their values at the beginning of my routine, and the @Index.time=0 at that point. I tried setting it myself at the end, to 15, but that didn't seem to make any difference. If I do an Eval after that, it shows @index.time set back to 30.
2) I'm not using any pre or post window processes. Also, I created a different routine for some testing, and it doesn't make any difference whether the Replace Indexing code is a W, S, X, or E for calling my replace routine, I still get the same results.
3) How do you suggest forcing an unused macro key? I played around with Data statements but they didn't make any difference that I noticed.
Thanks again.
At 17 APR 1998 02:21PM Aaron Kaplan wrote:
Not sure why this is happening. Assuming a clean return and no modifcations to system stuff are being made, everything should go back to normal. If I remember right, the code is simple.
If @environ.set then
catalyst...End Else
Indexer.Controler()End
or something like that. It was pretty straight forward.
Anyway, macro kestrokes are \0078\ for alt-1, \0079\ for alt-2 and so forth up through alt 7. Alt-8 is used for some real funky screen stuff, so don't call that.
apk@sprezzatura.com
At 17 APR 1998 05:30PM Mark Caldwell wrote:
Upon further testing, I have found that if I place my routine as a POST indexing process instead of a REPLACE, that it behaves like I want it to (i.e. it reactivates without any keystrokes), so RTI is doing something in their index controller that re-starts the clock… But what?!
Unfortunately, I don't want this workstation to do background indexing, because we have the dedicated indexing machine. So, using it as a post does not really solve my problem.
Mark Caldwell
At 24 APR 1998 11:43AM Mark Caldwell wrote:
I got it! The solution was a combination of ensuring that the routine is called as a subroutine by the indexer (which windows are, but my other test logic was not), AND using an @DATA=\007E\ (which is Alt-7, apparently not used by AREV) to activate a key that does nothing, but resets the timer.
Thanks for your help!
MC
At 25 APR 1998 07:22AM Aaron Kaplan wrote:
And I guess you've tried loading up @DATA with something?
apk@sprezzatura.com