I'm hoping someone can help point me in the right direction.

I'm trying to get Python 3 working on Mystic (A48 PreAlpha from 2022-03-11)
without much luck.

When executing even the included test script, the user is immediately
disconnected from the BBS with the below in the error.log:

MYSTIC 001 An error has occured: EAccessViolation Access violation

Under Configuration >> General Settings >> Python3 Library I have set:
/usr/lib/arm-linux-gnueabihf/libpython3.8.so.1.0

I have checked that this file exists. If I leave this blank Mystic BBS fails to
find Python3 at all.

Raspberry Pi OS Bullseye comes with Python 3.9 (3.9.2) by default so I've
installed Python 3.8 (3.8.7). I've also tried pointing to libpython3.9.so.1.0
but the results are the same.

Anyone else have Python 3 working on the Pi? Any suggestions for what I should
try?

Thanks!

.... Error 3032 - Recursion error. See error 3032.

TM> When executing even the included test script, the user is immediately
TM> disconnected from the BBS with the below in the error.log:

Unfortunately I do not have a Pi that I can install the latest OS on at the
moment to test with so it will be a little more difficult to look into this.

If you run a "ldconfig -p | grep libpython" what do you see?

I'll see if I can find a way to get access to Bullseye so I can test and get
the Wiki updated with any findings (I just noticed I haven't updated the Wiki
with notes since the 2.7 release).

.... The caterpillar does all the work but the butterfly gets all the publicity

g0> If you run a "ldconfig -p | grep libpython" what do you see?

libpython3.9.so.1.0 (libc6,hard-float) =>
/lib/arm-linux-gnueabihf/libpython3.9.so.1.0

libpython3.9.so (libc6,hard-float) => /lib/arm-linux-gnueabihf/libpython3.9.so

libpython3.8.so.1.0 (libc6,hard-float) =>
/lib/arm-linux-gnueabihf/libpython3.8.so.1.0

libpython3.8.so (libc6,hard-float) => /lib/arm-linux-gnueabihf/libpython3.8.so

libpython2.7.so.1.0 (libc6,hard-float) =>
/lib/arm-linux-gnueabihf/libpython2.7.so.1.0

g0> I'll see if I can find a way to get access to Bullseye so I can test and
g0> get the Wiki updated with any findings (I just noticed I haven't updated
g0> the Wiki with notes since the 2.7 release).

Any advice you're able to offer is much appreciated. For the time being I'm
able to make do with Python 2.7. I'm just more familiar with 3 so I wanted to
give it a shot.

Thank you for your reply!

.... DOS=HIGH? I knew it was on something...

g0> Unfortunately I do not have a Pi that I can install the latest OS on at
g0> the moment to test with so it will be a little more difficult to look
g0> into this.

I just wanted to thank you again for your response on this and also let you
know that you can probably ignore it.

I'm not sure what has caused things to go sideways but I first tried Mystic
with a fresh install of Raspberry Pi OS Bullseye. Python3 worked out of the
box, no need to point it to libpython3.8.so.1.0.

I then installed mystic into a new directory on the same RPi as my existing
board. Again Python 3 worked out of the box. I'm not sure what is broken in my
main board but I think I'll just try migrating my BBS to a fresh install of
Mystic and see if I can sort things out that way.

It's possible that because I had my board on a drive that was using LVM then
linked the directory back to /mystic that something went awry there. I'll play
around with a few different scenarios and see if I can figure out how I broke
things.

Thanks again!

.... A .GIF is worth a thousand .TXT.

On 31 Mar 2022, Travis Mehrer said the following...
TM> ... A .GIF is worth a thousand .TXT.

LOL, consider this stolen. ;)

Jay

.... New Mail not found. Start whine-pout sequence? (Y/N)

JH> TM> ... A .GIF is worth a thousand .TXT.
JH>
JH> LOL, consider this stolen. ;)

Dude. You need to check your BBS's config. That's actually a pretty common
tagline...

McDoob
SysOp, PiBBS
pibbs.sytes.net

.... This virus requires Microsoft Windows 3.x

TM> g0> Unfortunately I do not have a Pi that I can install the latest OS on
TM> TM> g0>
TM> g0> the moment to test with so it will be a little more difficult to TM>
TM> g0> look
TM> g0> into this.
TM>
TM> I just wanted to thank you again for your response on this and also let
TM> you know that you can probably ignore it.

Thank you for the update.

I don't know if this would be a factor in your case, but in the past I have
struggled having multiple Python installations in the same Linux VMware, and I
have completed destroyed installations by trying to install different versions
of Python on top of or to replace what the OS ships with.

I wouldn't consider myself an expert in that area though, but I have found it
to be tricky to touch Python's default installations without blowing up
things badly enough that I end up completely reinstalling Ubuntu.

.... If at first you don't succeed, blame your parents!

g0> I don't know if this would be a factor in your case, but in the past I
g0> have struggled having multiple Python installations in the same Linux
g0> VMware, and I have completed destroyed installations by trying to
g0> install different versions of Python on top of or to replace what the OS
g0> ships with.
g0>
g0> I wouldn't consider myself an expert in that area though, but I have
g0> found it to be tricky to touch Python's default installations without
g0> blowing up things badly enough that I end up completely reinstalling
g0> Ubuntu.

Thank you again, for your reply g00r00! So, FINALLY after a little pulling out
of hair, I managed to figure out what is going on.

After a new install on the same Pi started out working, and then failed after I
started migrating my menus to the new board; it struck me.

In my prelogin menu I have a simple python2 script running. And it was the
mixture of Python2 and 3 that was breaking things for me. After fixing a text
file my script was importing (had some sort of broken UTF-8 enconding in it) I
was able to run this script as Python3 and then continue to use Python 3.

TL;DR: Mixing Python2 and Python3 in Mystic (RPi OS Bullseye) can break things.
Lesson learned, now I can finally move on with Python3 and Mystic.

blade

.... User Error: Replace user and hit any key to continue...

TM> TL;DR: Mixing Python2 and Python3 in Mystic (RPi OS Bullseye) can break
TM> things. Lesson learned, now I can finally move on with Python3 and
TM> Mystic.

Thanks for the follow up. Once I have a chance to get back working in a Pi
install again I will be sure to test this and see what I can figure out!

.... Consultant: A person who makes good on a salesman's promises!

Hello and thank you for creating this wonderful BBS!

I'm in the process of building a front end to a Node.js API for keeping track of all of my ICs and other active electronics. I want to be able to access it from my Amiga 1000, and so I've decided that Mystic BBS is the best option for this. I already use Mystic for moving text files back and forth from WP 4.1 on my Amiga to NeoOffice on my Mac and it works great.

Anyway, I am using the Python3 module 'requests' for handling HTTP requests to the Node.js API on an Ubuntu machine on the interwebs.

However, simply importing this module into a Mystic Python3 script causes an error to show up in the error.log:
EAccessViolation Access violation

I am running Mystic version 1.12 A48
It's running on a Raspi 4 with the latest Raspbian
It's running Python 3.9.2

The python script is simply:
import mystic_bbs as bbs
import requests
bbs.writeln("Hello from Python 3")
bbs.writeln("We have imported requests module")
bbs.writeln("|PA")

The script runs fine until it ends at which point it abruptly ends the connection. It even properly executes the |PA pause; but as soon as a key is pressed the connection terminates.

If I comment out the "import requests" line, the script runs without terminating the connection and there is no error in the log file.

Any help would be most appreciated.

ps: I did do a fresh install of everything and tried with Python 2.7. The same script did not cause the connection to terminate, however it did give me a RunTimeWarning of an API version mismatch for module mystic_bbs.

Thank you!

Hello, same problem here with my VM under headless Ubuntu server.
Here only Py3 ai used but same error, i don't know how to solve, do you have
news?

-------------------------------------------------
Violence is the last refuge of the incompetent!

On Sunday, February 12, 2023 at 5:20:23 PM UTC-5, Nicola Mettimano wrote:
> Hello, same problem here with my VM under headless Ubuntu server.
> Here only Py3 ai used but same error, i don't know how to solve, do you have
> news?

No, I’ve not found a solution. I reinstalled everything from the OS up and only installed Python 3.9 and I still get the error: EAccessViolation Access violation

And while my script runs perfectly, as soon as it ends the telnet session closes,and the error above appears. It’s really quite strange.

DS> No, I’ve not found a solution. I reinstalled everything from the OS up
DS> and only installed Python 3.9 and I still get the error:
DS> EAccessViolation Access violation

that's what I was about to try...

DS> And while my script runs perfectly, as soon as it ends the telnet session
DS> closes,and the error above appears. It’s really quite strange.

weird....can't use python with mystic, this is terrible news...

-------------------------------------------------
Violence is the last refuge of the incompetent!

DS> > Hello, same problem here with my VM under headless Ubuntu server.
DS> > Here only Py3 ai used but same error, i don't know how to solve, do you
DS> > news?
DS>
DS> No, I’ve not found a solution. I reinstalled everything from the OS up
DS> and only installed Python 3.9 and I still get the error:
DS> EAccessViolation Access violation

Their problem does not appear to be the same issue, just off the little
knowledge I have. Their setup sounds like an environment issue and yours I
don't think is.

Since your script is running and throwing an access issue when closing, its
more likely to be a bug. As you may or may not know,
the current version for 64-bit ARM is the first release of it, so that makes it
even more likely to be an issue with Mystic. I can only assume when you say
latest Pi OS that means you're using 64-bit OS and Mystic 64-bit for ARM.

The reason I hadn't responded yet is because I haven't had time to get my
64-bit Pi out to see if I can reproduce the issue on my own.

.... Read messages, not taglines

I'm not certain this is related but I've noticed similar issues with Python 3
in Mystic. Particularly when my script is making any sort of external http(s)
connection.

A while back I was writing my own "This Day in History" type script. Everything
worked fine until the script finished and then the connection would get dropped
from the BBS. In the end it was easier to make cron job that downloaded the
history to a file and then open it via open().

More recently while trying to use feedparser to read RSS I've noticed the same
behavior: once the script finished, my session gets disconnected.

I haven't had time to dig into it more but I suspect if I make a cronjob to
save the feed as an XML file and then open() the file and work with it, I
suspect it will work. It's not the cleanest way and perhaps not a final
solution it's a possible work-around for the time being.

blade

.... Error, no Keyboard - Press F1 to Continue.

g00r00, thank you for the response! I know how busy you are.

I'm running the following Raspbian version:
Release date: September 22nd 2022
System: 32-bit
Kernel version: 5.15

So it's 32 bit Raspbian on a Raspi 4 and 32 bit Mystic with Python 3.9.

If I run the following script:
---------------------------------------
import mystic_bbs as bbs
import requests

bbs.writeln( "Testing importing requests" )
bbs.writeln( "All Done |CR|PA" )
-------------------------------------

I get the "EAccessViolation Access violation" error right after pressing any key at the very end of the script and the telnet session closes.

If I run the same script without importing requests, I get no error. Just importing requests causes the error to happen at the end of the script - I don't even have to use it.

But if I do use it, the script runs fine all the way to the end and then throws the error and closes the telnet session.

Very strange indeed. And thank you for helping with this!