Hey all,
Looking for someone to test the A44 prealpha upgrade.
I finally got the new theme system to workable, at least enough to get a release out (which should also fix the date bug in MUTIL). Keep in mind this has a LOT of changes. The theme editor is only somewhat working.
Happy to see ya back! I did have a bug request possible I posted else where but will post it here as well the acs OV and -V can not be checked on a new user even after doing the verification on their first logon; I
testing this and also turned this off with the same results. Each time the user would need to log out and back onto the BBS to see the -V for email verification to pop up for them.
testing this and also turned this off with the same results. Each ti the user would need to log out and back onto the BBS to see the -V fo email verification to pop up for them.
I think this is fixed now in the latest A44 prealpha. Thank you and let me know if you do end up trying it and you still have issues.
Hey all,
Looking for someone to test the A44 prealpha upgrade.
Hey all,
Looking for someone to test the A44 prealpha upgrade.
release out (which should also fix the date bug in MUTIL). Keep in mind this has a LOT of changes. The theme editor is only somewhat working.
Looking for someone to test the A44 prealpha upgrade.
Thanks! Could you please take a look at a few other issues
(SEEN-BY lines and wrapping of text) that I have been in touch
about by email? It would be great if the next version could
address those.
Looking for someone to test the A44 prealpha upgrade.
Thanks! Could you please take a look at a few other issues (SEEN-BY
lines and wrapping of text) that I have been in touch about by email? It would be great if the next version could address those.
..oh and thanks for the updates level 3 logging in the way tossing and packing is being reported. Very helpful.
Could you also enable the logging menu functionality in this build?
Hey g00r00, testing on Ubuntu 18.04.3 x64. Upgraded from A43 today. No
big deal, the prompts.dat/prompts.txt thing you've likely heard about,
but aside from that things worked out.
I noticed my <bbs>/scripts/ directory was moved but its subdirectories were not. Same with <bbs>/text/.
been neglecting e-mail a lot lately (ie I haven't been reading it since maybe November sometime)
At the current moment I don't know of the SEEN-BY issue but I will of course get it cleared up ASAP once I do!
and maybe all the other FTN bugs:
I ran the update, but the /scripts folder was missing for me.
Hey all,
Looking for someone to test the A44 prealpha upgrade.
I finally got the new theme system to workable, at least enough to get a release out (which should also fix the date bug in MUTIL). Keep in mind this has a LOT of changes. The theme editor is only somewhat working.
At the current moment I don't know of the SEEN-BY issue but I will of course get it cleared up ASAP once I do!
Thank you, Yeah probably best if you just look for the 3-4 emails I sent and that should hopefully sum up quite a bit of what's been picked up.
wrong file timestamps in binkp (offsetted by the timezone)
wrong format in the binkp time field (but that is only informational, afaik)
wrong month in the pkt header
I ran the update, but the /scripts folder was missing for me.
It should now end up in <bbs>/themes/default/scripts
Found one item seems to be limited on the path for the theme location, an example say I wanted to do:
/mystic/data/theme/ansi/menu instead of menus, this could be me or is there a short way to do that from the fallbacks provided?
Can you show this? There were some older versions where something
like this would happen, but it was because of another BBS software
that is no longer in development creating improper fields.
Found one item seems to be limited on the path for the theme location example say I wanted to do:
/mystic/data/theme/ansi/menu instead of menus, this could be me or i there a short way to do that from the fallbacks provided?
You don't have separate paths for those anymore. You have a root theme directory and all themes go in there. So there is only one path now
that is configurable and thats where your themes go (in system paths).
Then the theme directory (say c:\bbsthemes\) has a directory one per theme. Under that is "text" "scripts" and "menus" directories. You
cannot change those names.
Mystic creates packets with the month off by one in the packet header,
for example in January, mystic sets the month as 1, whereas it should be 0.
Great to see you back by the way :)
Mystic creates packets with the month off by one in the packet
header, for example in January, mystic sets the month as 1,
whereas it should be 0.
Thanks, this has been fixed. Its amazing it went like a decade
unnoticed!
Thank you, Yeah probably best if you just look for the 3-4 emails I s and that should hopefully sum up quite a bit of what's been picked up
The SEENBY should hopefully be fixed now!
If you get a chance to try it let me know, and if it works then I will spring on some more changes (the start of the stat tracking) that could totally break everything lol.
The SEENBY should hopefully be fixed now!
If you get a chance to try it let me know, and if it works then I will spring on some more changes (the start of the stat tracking) that could totally break everything lol.
wrong file timestamps in binkp (offsetted by the timezone)
Just double checked and the standard I am looking at says to give the
Unix time of the file. It does not say it should be UTC. But if you
can find something official that clarifies that let me know.
wrong format in the binkp time field (but that is only
informational, afaik)
Can you elaborate? Are you looking at an old version or something?
Here is what it says:
_NUL "TIME remote_date_time"
remote_date_time format is described in [RFC822].
The format given fits that described in RFC822.
wrong month in the pkt header
Can you show this? There were some older versions where something
like this would happen, but it was because of another BBS software
that is no longer in development creating improper fields.
Mystic creates packets with the month off by one in the packet
header, for example in January, mystic sets the month as 1,
whereas it should be 0.
Thanks, this has been fixed. Its amazing it went like a decade
unnoticed!
The SEENBY should hopefully be fixed now!
If you get a chance to try it let me know, and if it works then I will spring on some more changes (the start of the stat tracking) that could totally break everything lol.
So as you can see there's some additional forward slashes showing at the start of the second (and subsequent) SEEN-BY lines which don't look
right to me. But I do think you're heading in the right direction with
Mystics BBS interpretation isn't accepted by the linux date command. I think it was a missing + or space character. Some tiny detail was wrong.
I know that I've got from Avon's 1/100 Hub had a invalid month in December. AFAIK the packets were created by Mystic BBS A43. Fidonet mail packets weirdly use 0 to 11 (0: Jan, 11: Dec). I have no idea why, but that's the way it's specified. Mystic is one month off (1 to 12).
Hi ho... OK day two so I tested the next build but sadly can only report the same issue observed as last night
Mystics BBS interpretation isn't accepted by the linux date
command. I think it was a missing + or space character. Some
tiny detail was wrong.
You'll need to be more specific than that.
I just noticed that the FSX mail doesn't seem to be getting to the BBS I was using (not sure if my replies have gone out either) so I am a bit behind.
I just fixed up the SEEN-BY again and I will make another build shortly, So there should be a build dated Jan 24th shortly that has the fix.
just compare it with the date format that binkd is using. the difference is at the end of the string. i don't think it's important. i only looked into this, because i thought that was the reason for the wrong file timestamps (but it's not).
the line breaking at 80 chars is much more problematic ...
I tested the following and we're getting there but there is still an
issue I think with the logic being applied.
Thanks for the update. This is challenging to test because I don't have any system capable of reproducing the situation at the moment.
Along the echomail lines, I saw your message about potential issues with FLO files so I have introduced FLO.TXT! We got rid of the BUSYLOG and
now we get another one ;)
I made a special build for you (anyone really) who wants to help locate any issues with FLO logs. The builds below will create a "flo.txt" in
the same directory anytime one of the flow files are modified in any
way. If there is some sort of error or issue with FLO it should be captured in that .txt file.
just compare it with the date format that binkd is using. the
difference is at the end of the string. i don't think it's
important. i only looked into this, because i thought that was
the reason for the wrong file timestamps (but it's not).
The string for my timezone is identical to that of binkd, which is I thought what I had already said but maybe I didn't. I literally
looked at the binkd code back in the day because the standard wasn't always clear.
I am willing to bet you saw a + on non-biased time zones (which I
wouldn't see where I live because I am not in Asia/Austrialia).
BinkP references that RFC like I already mentioned, but then the
example they give directly after that doesn't even follow the RFC
format (lol).
The Fidonet standard for TZ specifically says NOT to
include a + but that software should be prepared to handle it if it
does.
So which one do I follow? Mystic can follow the shoddy BinkP specification tha contradicts itself, or the FidoNet standard for supplying the TZ bias? Right now it follows the TZ FidoNet standard
and doesn't include the + but I can change it to show the + just for
BinkP TIME frame if that is what BINKD does.
the line breaking at 80 chars is much more problematic ...
Its not breaking, its word wrapping text.
It was a requested feature
both at the echomail and NNTP levels because of content getting lost.
OK ta... so to clarify this build looks at files created by MUTIL for a BSO mailer weather it be Mystic's inbuilt BinkP server or the one I am currently using to send traffic around (BinkD)? Am I understanding that right?
I need to ensure SEEN-BY's are sorted before I drop it into NET 1 HUB though.. else I create more issues for nodes while trying to test for
this next thing. Make sense?
Just follow the RFC 822 standard as binkd does and binkp requires.
And it's really annoying. Just because it happens to have some
advantages for Mystic BBS users on a 80x25 terminal doesn't mean
There is also not one technical reason to add <CR> characters for
exported mail or to mails imported to the JAM base. Word wrapping can be dynamically applied in the message reader or the NNTP server.
Please disable it ASAP. It introduces unnecessary problems and we wasted
No, whenever MIS or MUTIL or FidoPoll touches a ?lo file it just logs in detail what its doing and if there was any issues. So if you're saying file references are getting removed from flo files, this should show us when/how that is happening.
I could be misunderstanding the issue too, I've been trying to do a lot quickly lately.
I need to ensure SEEN-BY's are sorted before I drop it into NET 1 HUB though.. else I create more issues for nodes while trying to test for this next thing. Make sense?
Yep makes sense. Unfortunately I was not able to get my laptop
yesterday so I could test it and make sure its fixed. I'll try to
figure something out today. I do have the Pi3 that I use to compile
out today. I do have the Pi3 that I use to compile Mystic maybe I can set my BBS up on that too for testing, but testing on a Pi is very very time consuming.
With my development I use CI, and have it automating the builds of apps, app testing and docker images. EG: When I commit code that is building a docker container, CI automates the build of that container - that I can then use a few mins later. As an example, my mystic container build is
out today. I do have the Pi3 that I use to compile Mystic maybe I ca my BBS up on that too for testing, but testing on a Pi is very very t consuming.
So testing on any platform shouldnt be time consuming - and if you are
up for it, I'd be willing to help you make it easier.
And it's really annoying. Just because it happens to have
some advantages for Mystic BBS users on a 80x25 terminal
doesn't mean
It has nothing to do with Mystic BBS. It was requested by
people who saw problems with legacy software that can't
handle text longer than 80 characters, offline readers,
the mess NNTP was making (there was a huge push for NNTP
gatin back in the day).
You seem to have almost no context in history or the
reason things are done the way they are.
There is also not one technical reason to add <CR>
characters for exported mail or to mails imported to the
JAM base. Word wrapping can be dynamically applied in the
message reader or the NNTP server.
Sure they can. Now go back and tell the 50,000 DOS,
Atari, Commodore, etc packages that haven't had updates in
25-30 years so they can release new version that can
handle it. :)
Please disable it ASAP. It introduces unnecessary
problems and we wasted
Its such a huge problem that no one has brought it up in a
decade. The "problems" it introduced was that message
content doesn't get lost at the expense of a line being
wrapped. One is horrible,
and one is a visual line wra
that is hardly a big deal given how few people are using
more than 80 column displays.
The latter was universally agreed to be the best approach
by people actually using software.
Its interesting you're pushing so hard on this but (for
example) you don't seem to care about software using
things like MSGID that are not compatible with old
software despite the FTN people themselves saying its not
intended to be used that way.
I agree its time to revisit it, but you seem to have a
knack at trying to paint every little issue as the end of
the world. Its not.
NNTP gating and offline readers are seperate issues. They are not fidonet technology. Nobody said you should not prewrap messages exported for offline reading (QWK).
I don't know about what history you are refering to. I joined Fidonet in 1993, first as a point than as node running a BBS and serving several
anymore. Let's face it, closed source communication software that
doesn't evolve with the ecosystem will wither away at some point.
don't be a drama queen.
and you are still ignoring the real dupes it produces in fsxnet.
Can you (or someone else provide) some real world examples instead of
some imaginary number? I wonder how these programs handled fidonet
You obviously live in another bubble. I discovered all the bugs I
reported by actually using common FTN software. Were are the people and voices that universally agreed on this? I didn't hear any in the last couple of weeks and months
Not a programmer or know the specifics of how this works the way it does...but is there a way to add being able to validate a new user while they are online without needing them to log off & back on?
Thank you.
Not a programmer or know the specifics of how this works the way it does...but is there a way to add being able to validate a new user while they are online without needing them to log off & back on?
--- Mystic BBS v1.12 A44 2020/01/27 (Windows/32)
email address and enter the code and the BBS ups their access to a
access level of your choice.. I do believe it gives them the new access immediately. If you go back a month or so I explained in detail how I do
Not a programmer or know the specifics of how this works the way it does...but is there a way to add being able to validate a new user while they are online without needing them to log off & back on?
Not a programmer or know the specifics of how this works the way it does...but is there a way to add being able to validate a new user wh they are online without needing them to log off & back on?
There's the email verification that's in there, and I *think* you don't have to be logged off and back on to work it. Or you could just set everyone's access to normal user access on signup. =)
You can do a Email Validation for them, they get a code sent to their email address and enter the code and the BBS ups their access to a
access level of your choice.. I do believe it gives them the new access immediately. If you go back a month or so I explained in detail how I do it.. IT works great. If you can't find it let me know and I will repost how I do it..
If you are using Windows then you can go into their node window where the user is logged in and use Alt-E to open the user editor
There's the email verification that's in there, and I *think* you don't have to be logged off and back on to work it. Or you could just set everyone's access to normal user access on signup. =)
I have a fun little auto validator script I found somewhere that just
kind of makes it look it's doing something and then flips their access level to a higher level. This isn't a proper Mystic Validator email routine but I had it for 8 years and it was cool. No logoff and back on.
email address and enter the code and the BBS ups their access to a access level of your choice.. I do believe it gives them the new acce immediately. If you go back a month or so I explained in detail how I
I'm using it as well. I'm not 100% sure if it bumps them up immediately though. Will have to test that out.
There's the email verification that's in there, and I *think* you don't have to be logged off and back on to work it. Or you could just set everyone's access to normal user access on signup. =)
I have a fun little auto validator script I found somewhere that just kind of makes it look it's doing something and then flips their acces level to a higher level. This isn't a proper Mystic Validator email routine but I had it for 8 years and it was cool. No logoff and back
Fantastic! Kick it up here or over here & will get it running.
Mystic already does e-mail auto-validation. You can also automatically upgrade users using a menu command too.I have a fun little auto validator script I found somewhere thatFantastic! Kick it up here or over here & will get it running.
I think its obvious which one of us has the knowledge and which
one of us has an obscure use-case that they can't let go.
Its an option to have it upgrade your access, and it should happen immediately.
Don't remember it...but please repost it. Will certainly do it.
Thank you again!!!
Its an option to have it upgrade your access, and it should happen immediately.
Ok, cool. I've only had a couple of new users since I added it and I wasn't around when they went through the process so wasn't sure if it
had been immediate or not.
Mystic already does e-mail auto-validation. You can also automatical upgrade users using a menu command too.
Have today off...so off to the dox I go.
Thank you again!
Don't remember it...but please repost it. Will certainly do it.No problem I will do it later tonight when i get into my "cave"...
If you have troubles let me know, the documentation on some areas is bad or at least not well organized.Mystic already does e-mail auto-validation. You can also automa upgrade users using a menu command too.
I only noticed because I've been hex editing packets to / from mystic. Trying to figure out why mystic won't route my netmails, yet both HPT
and crashmail do. Mystic routes other netmails fine, but I can't for the life of me figure out what I am doing wrong.
Found one item seems to be limited on the path for the theme location example say I wanted to do:
/mystic/data/theme/ansi/menu instead of menus, this could be me or i there a short way to do that from the fallbacks provided?
You don't have separate paths for those anymore. You have a root theme directory and all themes go in there. So there is only one path now
that is configurable and thats where your themes go (in system paths).
I still would like to be able to customize the names of the folders regardless if possible as a request :)
| Sysop: | Eric Oulashin |
|---|---|
| Location: | Beaverton, Oregon, USA |
| Users: | 128 |
| Nodes: | 16 (0 / 16) |
| Uptime: | 38:46:36 |
| Calls: | 1,546 |
| Files: | 2,153 |
| Messages: | 313,323 |