Actually I came across something today that might apply... Trace _appeared_ to drop out... but I don't really think it was that simple. What was really going on was two vars were typed as Currency in the command file and Double on the form. Trace appearing to stop was just one of several subtle hints that something was wrong.
Ben Petersen On 19 Feb 2003, at 20:28, Dan wrote: > > Hi, > Ok, this is a cool new thing, adding trace way down the line to start > it later... Great, that saves lots of time. However, something is > still stopping it from tracing, I don't know yet what. Is there any > known problems that cause trace to stop tracing? Dan > > > > > At 12:23 AM 2/20/2003 +0000, you wrote: > > > >I guess I should have known to try the obvious but there is nothing > >in the documentation _that I've seen_ to suggest that you could do > >anything other than "TRACE Command.fle (in ...)" > > > >I suppose something like: DEBUG TRACE works as well? > > > >Thanks Ben. > > > >Regards, > >Alastair. > > > > > > > >----- Original Message ----- > >From: "Ben Petersen" <[EMAIL PROTECTED]> > >To: "RBASE-L Mailing List" <[EMAIL PROTECTED]> > >Sent: Wednesday, February 19, 2003 2:20 PM > >Subject: [RBASE-L] - Re: Crashing > > > > > >> You don't have to start the app with trace... just place the word > >> "trace" at the point that you want it to start. Sorry it didn't > >> occur to me before :( > >> > >> Ben Petersen > >> > >> > >> On 19 Feb 2003, at 21:46, Alastair Burr wrote: > >> > >> > I've not had a problem with TRACE not going "deep enough" but it > >> > might be nice to be able to invoke TRACE at some point in a > >> > command file rather than having to start with it. > >> > > >> > What I have done, sometimes, is put "Trace_Break" on a line where > >> > I want TRACE to revert from running the file to single step mode > >> > but this won't help Dan if TRACE never gets to the point in his > >> > file where the problem occurs. "Trace_Break" just throws an error > >> > message and I find it easier than trying to use the [F9] Add > >> > Break option when running files within files and it can easily be > >> > remmed out or moved. > >> > > >> > Regards, > >> > Alastair. > >> > > >> > > >> > > >> > ----- Original Message ----- > >> > From: "Dan Champion" <[EMAIL PROTECTED]> > >> > To: "RBASE-L Mailing List" <[EMAIL PROTECTED]> > >> > Sent: Wednesday, February 19, 2003 7:48 PM > >> > Subject: [RBASE-L] - Re: Crashing > >> > > >> > > >> > > > >> > > No, thats not it. I can make it exit abnormally by following a > >> > > direct > >> > route > >> > > through the form, to another then closing them. If only I could > >> > > run trace that deep I might find it. I am surprised not to > >> > > hear any mention of why trace can't go that far. I have files > >> > > set to 60 and trace just stops tracing at about the fourth form > >> > > that gets opened. Then the program runs as though trace was > >> > > never called. (very strange indeed) > >> > > > >> > > > >> > > At 02:41 PM 2/19/03 -0500, you wrote: > >> > > > >> > > >The RBase files are not on a box that the disk has spooled > >> > > >down (sleep because of disk inactivity) is it? > >> > > > > >> > > > >can be open for hours at a time, > >> > > > > >> > > > > >> > > > > >> > > > > >> > > >----- Original Message ----- > >> > > >From: "Dan Champion" <[EMAIL PROTECTED]> > >> > > >To: "RBASE-L Mailing List" <[EMAIL PROTECTED]> > >> > > >Sent: Wednesday, February 19, 2003 2:14 PM > >> > > >Subject: [RBASE-L] - Re: Crashing > >> > > > > >> > > > > >> > > > > > >> > > > > Allister, > >> > > > > The trouble you had, was it with the form not exiting > >> > > > > normally? The > >> > > >form > >> > > > > I am having trouble with can be open for hours at a time, > >> > > > > the when it comest time to close it, it does not return to > >> > > > > the calling program, so Rbase dumps me at the prompt. > >> > > > > I can try adding a pause before hand and see what it > >> > > > > does. > >> > > > > Thanks, > >> > > > > Oh, also I will examine how many run/returns I have nesting > >> > > > > and see > >> > if it > >> > > > > is possible to use quit to instead. > >> > > > > Dan > >> > > > > > >> > > > > > >> > > > > At 06:45 PM 2/19/03 +0000, you wrote: > >> > > > > > >> > > > > >Dan > >> > > > > > > >> > > > > >I have had (still do have) problems with forms where I > >> > > > > >have solved it > >> > by > >> > > > > >putting: > >> > > > > > > > >> > > > > >Pause for 2 > >> > > > > > > >> > > > > >immediately BEFORE calling the form. Intentionally there > >> > > > > >is no > >> > message > >> > > > > >because the screen doesn't need it so that nothing shows > >> > > > > >that there > >> > is a > >> > > > > >delay and there is nothing to do to clear the message. > >> > > > > > > >> > > > > >Anything less than 2 seconds and the problem remains. > >> > > > > > > >> > > > > >It might be worth your giving it a try... > >> > > > > > > >> > > > > >Regards, > >> > > > > >Alastair. > >> > > > > > > >> > > > > > > >> > > > > >----- Original Message ----- > >> > > > > >From: "Dan" <[EMAIL PROTECTED]> > >> > > > > >To: "RBASE-L Mailing List" <[EMAIL PROTECTED]> > >> > > > > >Sent: Wednesday, February 19, 2003 5:57 AM > >> > > > > >Subject: [RBASE-L] - Re: Crashing > >> > > > > > > >> > > > > > > >> > > > > > > enter using dispatch > >> > > > > > > pause 2 using "test" > >> > > > > > > return > >> > > > > > >> > > > > > >> > > > > > >> > > > > Dan Champion > >> > > > > P.O. Box 223 > >> > > > > Grandville, MI. 49428-0223 > >> > > > > www.championsolutions.net > >> > > > > > >> > > > > > >> > > > >> > > > >> > > > >> > > Dan Champion > >> > > P.O. Box 223 > >> > > Grandville, MI. 49428-0223 > >> > > www.championsolutions.net > >> > > > >> > > >> > > >
