Subj : Problem with filenames containing spaces
To   : James Coyle
From : Oli
Date : Tue Jan 18 2022 08:44 pm

James wrote (2022-01-18):

Ol>> This is not what FTS-1026 recommends. It also doesn't work with
Ol>> every legacy mailer. What you don't understand is that FSP-1011
Ol>> specified the wrong escape method.

JC> FTS-1026 recommends it to be a per-connection configurable option as I
JC> have shown here, which is implemented in Mystic.  It is not an option in
JC> BINKD and if it were, this would not be a problem for Paul.  Those are
JC> the facts.

Do you really believe everyone is too stupid to recognize your discussion strategy?

You are trying to make non-standard behavior (bug) of Mystic a feature and explaining that every other mailer should implement a workaround for Mystic's problems (or it's not your problem anymore). You also do your best to not mention the fact, that some Mystic versions were not able to process filenames with standard compliant escaping (\x##).

Again, look at the old source code of binkd, which was and still is the reference implementation. Where is the code that sends \## escape sequences?

ftp://cvs.happy.kiev.ua/pub/fidosoft/mailer/binkd/

It always was and still is \x##.

Ol>> This option was meant for mailers that uses the incorrect escape
Ol>> sequence \## (like Mystic does).

JC> Again, you're literally quoting a message that says Mystic can operate in
JC> either way, which is the FTS recommendation.

That is not the point. Why would anyone who is using a standard compliant mailer care, that Mystic can be configured to send incorrect escape codes? (or has to be explicitly configured for sending the correct escape code)

JC> Grow up and stop being so disingenious all of the time.

Seriously?

Ol>> So you expect that binkd implement a feature that is not
Ol>> implemented in a

JC> No, as I have said in messages prior: I don't care if BINKD implements it
JC> as an option or not.  But the only way for Paul's problem to be solved is
JC> for BINKD to implement the FTS recommendation on how to handle escaping.
JC> I cannot help him.

And we cannot help Mystic users with their broken software.

JC> Stop trying twist everything to create confusion, and just let it go.

Me?

Ol>> I would also recommend upgrading from the Mystic mailer to some FTS
Ol>> compliant mailer and to one that supports the faster binkp/1.1
Ol>> protocol.

JC> As with every single thing you've said in this message, none of this is
JC> true.

No matter what you say, it is still true that I would recommend to use another mailer.

JC> The difference between BINKP 1.0/NR and BINKP 1.1 has nothing to do with
JC> performance.  It simply sends an extra round of EOB for handling file
JC> requests differently.

Here I was indeed wrong, binkp/1.0 can be as fast as binkp/1.1 (for stable connections). The slowness of Mystic's mailer must had some other reasons (maybe it is faster nowadays, but performance was not great in 2020. I switched to another uplink and Paul switched to binkd since then. So I don't know if Mystic's mailer has improved).

JC> You're saying the \## was a typo and not used anywhere, which is
JC> absolutely false.  The FTS documentation on the subject specifically says
JC> mailers DO use that escaping.  In fact, older mailers either did no
JC> escaping at all or only use \##.

JC> There are no FTS compliance issues with Mystic's echomail, and it
JC> probably transmits more FTN-style mail than any other software.

What the fuck are you talking about now? This was never about echomail.

JC> If there was an actual problem then speak up.

I did speak up some time ago, when Mystic used to mangle in-transit echomail. Your response was to throw a fit and react aggressively against the messenger.

I warned about that it could produce dupes. And it did. Mystic was THE software that produced most of the problems with echomail in Fsxnet.

JC> And Paul has worked with
JC> me enough to know if there was a problem I would fix it for him quickly,
JC> and he wouldn't have to go through this drama-filled daycare to get help.

And now he uses binkd and hpt.

JC> It doesn't matter how many more years you go on trying to harass me and
JC> spew a bunch of misinformation about me or Mystic, reality will never
JC> agree with you.

I know Mystic is your baby, but since when is harassing a thing a thing?

I'm really not interested to harass you. But I'm not the only one who is sick of your behavior. It doesn't matter how many more years you go on trying to harass others and spew a bunch of misinformation, reality will never agree with you.

JC> I am sure everyone here is more than tired of this, so just move on.

Yes, we are tired of your bullshit too. Fix your software and don't burden everyone else with the problems your software creates. It's as simple as that.


Let's wrap it up:
- You implemented Mystic binkp mailer on base of FSP-1011 (http://ftsc.org/docs/old/fsp-1011.003), which was only a proposal and was never released as a standard (right or wrong?).
- Now Mystic is sending incorrect escape sequences (if not configured otherwise in the newest (alpha) version?).
- You are convinced that sending an incorrect escape (\##) is the most compatible way.
- Binkd (and every other mailer) should do the same or implement workarounds for non-compliant software.
- Also: not understanding the correct escape sequences is not the problem of the software that doesn't understand it.

So Mystic is fighting to save the incorrect \## forever, because of some old legacy software (whichever that is) or something (whatever) ...

.... meanwhile, Binkd, qico, BinkIt and others are doing just fine with \x##

---
* Origin: Birds aren't real (2:280/464.47)