Hi, Bob.

BL>  RS> Pity that if the SOT/EOT has some value with original creators
BL>  RS> of PKTs, your 'isnt necessary' has imploded tho. Its not about
BL>  RS> QWKs.

BL>  FM> I know that, it's about PKTs.

BL>   That's not true.

What's not true? It *is* about PKTs!

BL> IF the *original* creator adds SOT/EOT, then why
BL> not add a Tearline and Origin line instead? Paul's logic is flawed. In
BL> fact, Paul's logic isn't.

I agree about ^aEOT:, not about ^aSOT:

BL>   Have you actually written your reader yet? How do you do it? You

I hadn't then but I have now. Not a reader in the sense of something to
read and write/reply to messages (yet) but a reader in the sense of
reading PKTs.

BL> just put EOT ahead of the Tearline... right? So what's EOT doing? If
BL> you *always* put an Origin line on the end of your message (like you
BL> do with EOT), then the LAST origin line in the PKT message is *always*
BL> the correct one. The fact that EOT is inside that is inconsequential.

I have come to the same conclusion as you are saying in this paragraph.

BL> If the orginal creator is wrong, then it's worng. Why is the EOT
BL> right and the Origin wrong? Where's the logic?

BL>   The only way EOT can work to corect a faulty message, is if someone
BL> like Paul in qwk2pkt adds his own EOT to a faulty message. I actually
BL> thought he was doing that. It doesn't make sense, otherwise.

And you can't to that. Not reliably. In I'm sure a vast majority of
messages you can do it *correctly*, but you can't do it *reliably*. And
where you can do it, you don't need to.

BL>   If you want to write a PKT reader and are tempted to add SOT/EOT,
BL> just add a Tearline and Origin line instead.... or sweet nothing at
BL> all in Netmail.

Yes. As the originator of a message that's all you need to do.

BL> You need an Origin line anyway if you are a point, or
BL> there's no return address in the message.

BL>  FM> Is it possible, in general, to scan a PKT and work out where to
BL>  FM> put SOT/EOT? If not why not, what stuffs you up from doing
BL>  FM> that?

BL>   Yes, you can insert your own SOT/EOT.

BL>                             SOT:

BL>   Read the 4 null-terminated strings that make up the header fields,
BL> read the next line. If it is "AREA:" read the message until you run
BL> out of consecutive #1 lines and add SOT in the next line. If you find
BL> a SOT line, remove it. It may be in the wrong place. SOT: is the last
BL> #1 kludge line unless someone has added one after it, in which case
BL> it's in the wrong place so you may as well remove it anyway.

That case is flawed in the case of no 0x01 lines, a netmail message,
and a user who has typed "AREA: a lot of garbage" as his first line. You
will add the ^aSOT in the wrong place and you CAN'T DO ANYTHING ELSE.

BL>   If the first line does not begin "AREA:" read consecutive #1 lines
BL> as before and insert SOT as before.

That case is safe.

BL>   Or put the fucking thing anywhere; it doesn't matter. It's only
BL> function is to make sure there is at least *one* #1 kludge line at the
BL> top of a message in Netmail.

Yes, and for that reason is worth including if you are writing the
message. Certainly in that netmail/no 0x01/user "AREA:" case. And won't
hurt in any other case.

BL> If you use SOT it may confuse someone's
BL> reader, so just add "#1IllogicalTwit" and it'll do the same thing
BL> anyway.


BL>                             EOT:

BL> 1. NETMAIL.

BL>   To insert EOT in Netmail... find the end of the message null. Read
BL> backwards to find EOT. If found remove it. Add your own on the true

You're still on about this reading backwards thing. Why?

BL> end of the message. Don't trust some other bastard's EOT. This gives
BL> you an idea of how important it is, btw.

I agree, at least until I see Paul's rebuttal to my long message.

BL> 2.  E-MAIL

BL>   In e-mail, find the end, read backwards to find the first line of
BL> the SEEN-BYs. If not found the message is dangerously faulty, so log
BL> it and remove it. Read backwards to find Origin line. This should be
BL> close to the SEEN-BY (120 bytes max). If not found you have a message
BL> without an address but send it anyway. Look for an EOT. If found
BL> remove it and write your own in front of the SEEN-BY line. You still
BL> have a message without an address, but you have an EOT. Who cares?

I still agree (except that you're still reading backwards).

BL>   If you find the Origin line, look for the Tearline within 80 bytes.
BL> If found look for an EOT. If found remove it and add your own in front
BL> of the Tearline. The other dopey bastard may have put his EOT in the
BL> wrong place. If no Tearline found, look for EOT. If found remove it
BL> and put your own EOT in front of the Origin line.

I'm afraid I have to question this 80 bytes, too. Do you mean ONE LINE
(a thing ending with a CR or CRLF)? If so *that's* what you look for -
tear line on THE LINE before origin line. Or are blank lines allowed to
intervene? if so look for *them*. What are you looking for the tear line
for anyway?

BL>   You will observe that when you find the EOT, the best thing to do is
BL> remove it.

BL>   And the mere fact that you can add SOT/EOT to a PKT proves that it
BL> has no function. It's a worry when a "programmer" like
Paul can't use
BL> logic.

That was my proposition a couple of weeks ago. I now know that it's not
possible to reliably add SOT to a packet unless you're the originator,
although I believe it is to add EOT (pending the discussion which will
inevitably ensue. :-)

Regards, fIM.

 * * You smash it - and I'll build around it.
@EOT:

---