RS> And in some ways it makes more sense to piss off PKTJOIN and use
RS> Daves message sorter to combine the PKTs anyway.

PE> Paul Markham's, you mean I presume.

Yeah, probably, brain fart on that one.

PE>> Nope, a perfectly working set.

RS> You dont know that, you arent trying to read the QWK with OLX.

PE> True, I use "offline".  However, the index files produced by
PE> reprocessing were different from his.

Well, by definition they must be different if he gets a usable QWK
by rebuilding the NDXs.

PE>> I'll let you into a secret. Once the thing is a single PKT file,
PE>> it's as clean as it can be, just like the original PKTs I sent you.

RS> Nice theory Paul. That doesnt explain how processing just one
RS> archived file of PKTs at a time makes the problem go away.

PE> You'll have to show that to me before I believe it.

I dont give a damn what you believe.

PE> That code is very simple, that combines packets.

Yep, and it wouldnt be the first time that some code which looks extremely
simple turns out to bite someone on the bum. Thats computing for you.

PE> Of course, like the code says, don't go "pktjoin *.pkt temp.pkt", as
PE> that will probably give you a nasty result.  Is that what your batch
PE> file does?

Nar, checked that. It doesnt put the created PKT in the same subdirectory
as the source PKTs. That appears to be how Brentons original did it.

PE>> Since everyone else is using PKT2QWK and indexes quite happily,

RS> Only two are using OLX.

PE> I think two for 1 year is sufficient.

PE>> it means you are doing something unusual.

RS> Nope. Lousy logic.

PE> Nope, you're imagining bugs in PKTJOIN which are more likely
PE> to be you stuffing up and leaving old NDX files lying around.

Nope, cant be that. I did indeed get a situation where it duplicated
every message, because the NDXs were literally duplicated. But that
gives a different symptom and I checked that that wasnt happening
anyway, coz I had seen it before.

The other problem with your theory is that it doesnt explain the
sequence where processing more than one archived set of PKTs at a
time produced a dud QWK and processing each one individually did not.
With absolutely no changes to the BAT files etc between those runs.

Anyway, not much point in speculating, that wont fine the fault.
The answer is to carefully preserve the starting files which produce
a dud QWK and carefully analyse whats going on. I cant do that coz I
havent had a dud QWK for many months now.

--- PQWK202