After further testing I'm now having the following issue.

The purpose of the app is to test for discrepancies between two folders,
where the second is a direct copy of the first, but in another location. In
theory, the Source and Target folders *should* be identical. The idea of the
app is to confirm that, or to find any discrepancies between the folders if
they exist. 

After each folder has all it's files added to separate listboxes, the Source
and the Target are compared using two dictionaries, and any files found to
not exist in one are marked in red and added to that folder's listbox. Then
any files which are found to be missing on the Target are copied from their
Source location to the Target location, keeping their original file
structure intact (i.e, the same files are copied to the same relative
locations from the Source to the Target).

After seeing that the app was behaving as expecting, I started a more robust
test. Instead of using the same set of 7 missing files in the Target
location that I had been testing with, I tried removing ten additional
files, by deleting the files, which were in sequential order when sorted by
Name, from the Target. Now, although the folder had not changed in any other
way from the number or names of files remaining, the same file set that all
previous tests were run on (i.e., no new files had been added to the
existing structure), I started to get a bunch of nilObjects as items were
added to the Target listbox. I looked at the filenames causing the
nilObjects, but those same files had been tested for a week, in both Source
and Target locations.

Then it got stranger. At first, when I put back the 10 missing files, the
nilobjects went away. But then I tried to track down exactly what was
happening. I removed only four of the same files this time, expecting to
find which of the files' removal was causing the nilObject.

The nilObject did not appear. However, the app was now incorrectly reporting
four additional files to the ones I had removed as missing, although they
were clearly present in the Target folder.

So here's the thing. Virtually all the routines I've been testing and fixing
for the past 2 weeks now, come from either the LR, or a ResExcellence
snippet on recursion, both pretty well-known RB snippets. They now seem, at
least at the moment, to be giving inconsistent results, apparently based on
how many files, and in what order, are in each folder.

So I'm now completely lost.

If anyone would be kind enough to view the project and see if they can see
what I obviously can't, I've uploaded the file to:

http://dhnet.us/compare/Compare_Volumes.rb.sit

Thanks again.

All My Best,
Jeffrey


_______________________________________________
Unsubscribe or switch delivery mode:
<http://www.realsoftware.com/support/listmanager/>

Search the archives of this list here:
<http://support.realsoftware.com/listarchives/lists.html>

Reply via email to