That's odd...
Pressing close does the same as pressing both disconnect buttons, then quitting...

Glad to hear the configurable prompt works properly... =)

Jazz
(List 112, Mk2 12 gig #40. Mk1 4 gig #30. Mk3 1.6 16v)

Pressing close does the same as pressing both disconnect buttons, then quitting...

Well, I can reproduce it pretty reliably on both my home and work computers. One is '98, the other is '2K.

On both of those systems, there is a long pause as it tries to process both disconnects, and then it never seems to get around to closing the main window. After the two buttons get done disconnecting, the main window responds sluggishly (I can drag its title bar around), but I can't close it. Eventually, it either crashes on its own or I have to task-kill it. Either way, the error is pretty severe- some sort of memory error or page fault.

___________
Tony Fabris

Now that I think about it, that sounds like an old cascading-event problem like one might get in VB. I assume Delphi is similar.

Perhaps it's one of these kind of scenarios:

- You press close button.
- WindowClose code says "If (connected) then (disconnect) and (wait for disconnect) then (try closing again)".
- Disconnect routines take longer than expected.
- WindowClose code tries to terminate the window process but it can't because there's still active code in the disconnect routines.

Depending on how it's written, that might put the code into an infinite re-entrant loop where the WindowClose routine keeps calling itself. Eventually the process crashes when it runs out of stack space.

I don't know if that's what's happening or not, I just thought the behavior looked like that's what was happening. I've seen it in VB too many times now.

___________
Tony Fabris

Hey, Jazzwire. Does EmpegTaxi have a file-size limit?

I just tried to snag a file off the Empeg that was over ten megabytes in size, and it refused to get the file size. The File Size column stayed at "TBA" even though it ran through the motions of getting the file size.

___________
Tony Fabris

It shouldn't have any limit..
The file I was testing the transfer with was over 20 megs...

Can you try getting the file size, then look at the serial console and run the exact ls command from Hyperterminal (without the "> /dev/usb" bit). I'd like to see the result, as it's the only thing I can think of that could be causing a problem...


Jazz
(List 112, Mk2 12 gig #40. Mk1 4 gig #30. Mk3 1.6 16v)

The empeg is complaining that the file doesn't exist. This is what I see in the serial debug window:

ls -lQ /drive1/fids/65c0 > /dev/usb
ls: "/drive1/fids/65c0": No such file or directory

The file in question exists on drive0, not on drive1. Perhaps there's something fishy in the code that decides which drive contains which files?

___________
Tony Fabris

I don't know if this is how EmpegTaxi works, but the *1 file and the *0 file are not necessarily on the same disk.


Roger - not necessarily speaking for empeg

I don't know if this is how EmpegTaxi works, but the *1 file and the *0 file are not necessarily on the same disk.

I hope the info is in the database, so that both drives are tried only if the player rebuilds it, right?

Dragi "Bonzi" Raos
Zagreb, Croatia
Q#5196, MkII#80000376, 18GB green

Oops... Guess who assumed they were... =)

I'll fix my oversight tonight...
(Guess who doesn't have a two drive empeg... ;)

Jazz
(List 112, Mk2 12 gig #40. Mk1 4 gig #30. Mk3 1.6 16v)

ROFL!

Cool, thanks for the help, Roger, and thanks for the great utility, Jazzwire.

___________
Tony Fabris