g00r00 --

I've added ANSI FSE support to ENiGMA 1/2 recently. When replying to ANSI messages coming from Mystic, the reply works well in enig, but shows up corrupted in Mystic (and maybe other ANSI supporting boards... not yet sure).

Can you take a look & see what I'm doing wrong or could do better? Here is a sample .pkt file:

http://s000.tinyupload.com/?file_id=85046989931615034381

Thanks!
Nu





--- ENiGMA 1/2 v0.0.7-alpha (linux; x64; 6.10.3)
* Origin: Xibalba -+- xibalba.l33t.codes:44510 (21:1/121)

On 08/19/17, NuSkooler said the following...

I've added ANSI FSE support to ENiGMA 1/2 recently. When replying to ANSI messages coming from Mystic, the reply works well in enig, but shows up corrupted in Mystic (and maybe other ANSI supporting boards... not yet sure).

It's hard to do ANSI correctly, even Mystic doesn't handle it well
with examples like yours which was quoting an ANSI logo. Mystic does
it better because it has a clue about ESC sequences and what you can
get rid of color codes and attributes. They all take the same form:
ESC[#(;#)m. Nothing else starts with ESC[ & ends in m, but everything
starts with ESC[ (1BH 5BH) making it harder to deal with.
The 'rules' are shown at:
http://bluesock.org/~willkg/dev/ansi.html

ESC[37m is white text. Some text follows and it ends with [5C or right 5 spaces, then two CRLF pairs (newlines).
00000000 1B 5B 33 37 6D 68 74 74 70 3A 2F 2F 62 63 77 31
.[37mhttp://bcw1
00000010 34 32 2E 7A 61 70 74 6F 2E 6F 72 67 1B 5B 35 43
42.zapto.org.[5C
00000020 0D 0A 0D 0A ....

The true result of all that is white text that shows:
http://bcw142.zapto.org

(ends here from 2 CRLF pairs)
Mystic strips the color codes to put a color in for sender and receiver.

--- Mystic BBS v1.12 A33 (Raspberry Pi/32)
* Origin: Mystic Pi BBS bcw142.zapto.org (21:1/145)

On Saturday, August 19th bcw142 was heard saying...

It's hard to do ANSI correctly, even Mystic doesn't handle it well with examples like yours which was quoting an ANSI logo. Mystic does it better because it has a clue about ESC sequences and what you can get rid of color codes and attributes. They all take the same form: ESC[#(;#)m. Nothing else starts with ESC[ & ends in m, but everything starts with ESC[ (1BH 5BH) making it harder to deal with. The 'rules' are shown at: http://bluesock.org/~willkg/dev/ansi.html

Thanks for the response - I don't need help with ESC codes though. ENiG fully supports them (hence the ANSI in FSE and ANSI quote builder). I have valid ANSI/ESC in a message and Mystic isn't showing it correctly.

Showing ANSI in FSE can be a real pain. I imagine Mystic just doesn't like something particular about how I'm building the message.

For ANSI in FSE, enigma pre-processes the original input into a canvas -- normalizing movement, line wrapping, etc. into essentially a grid containing either blank space or SGR + character. When you quote, lines from this buffer are appended to the message. This displays just fine elsewhere, but Mystic seems to wrap the lines / garble it. ...probably just doesn't expect something I'm doing (and could perhaps do differnet to be more "BBS compatible")





--- ENiGMA 1/2 v0.0.7-alpha (linux; x64; 6.10.3)
* Origin: Xibalba -+- xibalba.l33t.codes:44510 (21:1/121)

On 08/19/17, NuSkooler said the following...

g00r00 --

I've added ANSI FSE support to ENiGMA 1/2 recently. When replying to ANSI messages coming from Mystic, the reply works well in enig, but shows up corrupted in Mystic (and maybe other ANSI supporting boards... not yet sure).

Can you take a look & see what I'm doing wrong or could do better? Here
is a sample .pkt file:

One thing to note make sure the max width is 79 that is what fixed it when working with Mystic and possibly other systems.

Cheers!
Pequito

--- Mystic BBS v1.12 A34 (Linux/32)
* Origin: Twinkle BBS # (21:1/126)

On Aug 20th 2:19 am NuSkooler said...

I've added ANSI FSE support to ENiGMA 1/2 recently. When replying to ANSI messages coming from Mystic, the reply works well in enig, but shows up corrupted in Mystic (and maybe other ANSI supporting boards... not yet sure).

Just so you know, gooroo said something about he's moving house (or painting house or something i forget.. :) but it might be a little while before he answers.

Anyway, I just wanted to point out that I don't think mystic quotes ansi at all. Every time I've seen someone quote an ansi that is using a mystic system, it's stripped out all the colour and appears to be plain ascii (but with block characters or whatever).

Now, why it wouldn't take your perfectly fine ansi with quotes infront of it and render it as ansi is a mystery. Perhaps it's because of the way it handles quotes, I know when reading on agency quotes appear coloured....

Either way, if you want to delve into it more before gooroo gets back to you there's always the old opensource mystic you could peek at, though I understand
why you might not want to do that :)

Andrew



--- ENiGMA 1/2 v0.0.7-alpha (linux; x64; 6.11.2)
* Origin: Exotica - exoticabbs.com:8888 (21:1/125)

On Sunday, August 20th apam muttered...

Either way, if you want to delve into it more before gooroo gets back to you there's always the old opensource mystic you could peek at, though I understand why you might not want to do that :)

Yeah, went that path, but I think the source is too old (or I'm missing something).

Maybe I'll add the option to allow quoting ANSI as ASCII later if it's not going to be supported by other systems.




--- ENiGMA 1/2 v0.0.7-alpha (linux; x64; 6.10.3)
* Origin: Xibalba -+- xibalba.l33t.codes:44510 (21:1/121)

On 08/21/17, apam said the following... ap> Jus
t so you know, gooroo said something about he's moving house (or

painting house or something i forget.. :) but it might be a little whil

e ap> before he answers. ap>Yes
, moving and has a job that should last till Octob
er before he has muchtime. ap>
Anyway, I just wanted to point out that I don't think mystic quotes a
nsi ap> at all. Every time I've seen someone quote an ansi t
hat is using a ap> mystic system, it's stripped out all the
colour and appears to be plain ap> ascii (but with block cha
racters or whatever). ap> It does,
but strips the color to color it like it does all other email
. Ifyou want the full exact copy you have to
save it (X) with .ans and put it
back with (U) upload. Then edit it as y
ou want in (D) draw mode.

Now, why it wouldn't take your perfectly fine ansi with quote

s infront ap> of it and render it as ansi is a
mystery. Perhaps it's becauseof the ap> way it handles quot
es, I know when reading on agency quotesappear ap> coloured
.... ap> ap> Either way, if you want to delve into
it more before gooroo gets back to ap> you there's always th
e old opensource mystic you could peek at, though I ap> unde
rstand why you might not want to do that :) ap> ap>
AndrewLike other BBS it tends to refo
rmat quoted stuff in a bad way with long linesas above
. All BBS I've see have that issue
. That will really mess with thelook of an
ANSI.Come play one of the seldom 3D du
ngeon games now available on Twinkle withmany different sele
ctions to choose from!Subj: [ANSI] Dungeon Master
Date: 08J o h1 nD a i l e y ' sBa
se: FSÜÜÜÜÜÜÜ ÜÜÜ ÜÜÜ ÜÜÜ ÜÜ
Ü ÜÜÜÜÜÜÜÜ ÜÜÜÜÜÜÜ ÜÜÜÜÜÜÜÜ
ÜÜÜ ÜÜÜÛÜÜÜÜÜÛ Û
Û Û Û ÛÛÛßÜ ÛÛ
Û ÛÜÜÜÜÜÜÛ ÛÜÜÜÜÜÛ Û
ÜÜÜÜÜÜÛ ÛÛÛßÜ ÛÛÛ
ÜÜÜ ÜÜÜ Û Û Û Û
ÛÛÛÜÞÝÛÛÛ ÜÜÜ
ÜÜÜÜÜÜ ÜÜÜ ÜÜÜ ÛÛ
ÛÜÞÝÛÛÛ ±
ÛÛÛ ÛÛÛ ÛÛÛ
ÛÛÛ ÛÛÛ ß ÛÛÛ ÛÛ
Û ÛßßÛ ÛÛÜÜÜÛ
ÛÛÛ ÛÛÛ ÛÛÛ ß
ÛÛÛ ° °± ÛÛÛ
ÛÛÛ ÛÛÛ ÛÛÛ
ÛÛÛ ² ÛÛÛ ÛÛ
Û ßÛÛÛ ÛÛÛÛÛÛ
ÛÛÛ ÛÛÛ ÛÛÛ
² ß±° ÛÛßßßßÛÛ ÛÛßß
ßßÛÛ ÛÛÛ ² ÛÛÛ
ÛÛßßßßÛÛ ÛÛßßßßÛ
ÛÛßßßßÛÛ ÛÛÛ ° Û
ÛÛ ² ß±° ßßßßßßßß ßßßßßßßß ßßß
° ßßß ßßßßßßßß ßßßßßßß ßßßßßßßß ßßß ßßß
ßÛ ß±°ß ßßß °°°°±±±±±°° °± °
Ý°°±±° °°°°°°ÜÜÜ
° Ü Ü Ü°±²ÜÜÜÜßßß°°°°°°
°±²ÝÞ °°°°Þ±°°°° °
±°°°°°Üܱ±²²±° It
Û Û Þ°±²ÛÛÛß ÜÜ ßßßß°±±ßß
²±°°±² ²±±±±°°°±°°°ÜÜßÛÛÛÛÛÛÛ²±
° ° Û° Ý °±²ÛÛÛÛ Û
ßÛ Þ²±°Üß²²±±²
ß ß²²²²ß°±°ÜÜ Û
ß Û ÛÛÛÛÛ²±±° is Û±Û
Þ ß°°±±²²Ü ßß ÜÛÛÛÛ
²±°Ü ßß ²²²²ßß ±° ÜÜ
°±²ÛÛÜßßÜÛÛÛÛ²²±±°ß
°°±±ÜßÜÜ ßßß°±²Û²±°
ßßßßß ÜÜÜ Ü²²²²²±±°°²²²Ü ßß°°±²²²
ÛÛÛÛÛÛ²±°ßßß here. ° °°
°±±ÜÜßßÜÜÜÜÜÜÜÜÜܱ²±°
²±°ÛÛ±±° ° °±±±Ü ²±° ßßßßßß
Ü °°°°±²² ÜÜÜ ß ÜÜÜ ÜÜÜÜÜÜÜ ÜÜÜÜÜ
ÜÜ ÜÜÜÜÜÜÜÜÜ ÜÜÜÜÜÜÜ ÜÜÜÜÜÜÜÜ
Ü ß°°°° ÛÛÛßÜßÛÛÛ Û
ÜÜÜÜÜÛ ÛÜÜÜÜÜÛ ÛÜÜÜÜÜÜÜÛ
ÛÜÜÜÜÜÛ ÛÜÜÜÜÜÜÛ ²²±°
ÜÜÜÜÛÛÛÜÛÜÛÛ
Û ÜÜÜÜÜÜÜ ÜÜÜÜÜÜÜ ÜÜÜ
Ü ÜÜÜÜÜÜ ÜÜÜÜÜÜÜܲ±°° °±±
±±²²²²² ÛÛÛ ß ÛÛÛ ÛÛ
ÜÜÜÛÛ ÛÜÜÜÜÛÛ Û
ÛÛ ±± ÛÛÜÜÜÛ Û
ÛÜÜÜÛÜÛ ²²²±±±°° °
ÛÛÛ ÛÛÛ ÛÛÛ
ÛÛÛÛÛÛ ÛÛÛ
ÛÛÛÛÛÛ ÛÛÛ
ÛÛÛ ÛÛÛ ÛÛÛ Û
ÛÛ ÛßßßßÛÛ ÛÛÛ
ÛÛßßßßÛ ÛÛÛ ÛÛ
ßÛßßß ßßß ßßß ßßß ßßßßßßß ßßß ßß
ßßßßß ßßß ßßßß--- Mystic BBS v1.12 A34 (Linux/32)

* Origin: Twinkle BBS # (21:1/126)
--- Mystic BBS v1.12 A33 (Raspberry Pi/32)
* Origin: Mystic Pi BBS bcw142.zapto.org (21:1/145)

On 08/21/17, apam pondered and said...

Anyway, I just wanted to point out that I don't think mystic quotes ansi at all. Every time I've seen someone quote an ansi that is using a
mystic system, it's stripped out all the colour and appears to be plain ascii (but with block characters or whatever).

I think it may be if the base is set to allow ANSI... just a guess..

--- Mystic BBS v1.12 A34 (Windows/32)
* Origin: Agency BBS | telnet://agency.bbs.geek.nz (21:1/101)

On Monday, August 21st bcw142 muttered...

It does,but strips the color to color

it like it does all other email. If

you want the full exact copy you have to save it

(X) with .ans and put it

back with (U) upload.

Then edit it as you want in (D) draw
mode.

Ah that helps, thanks. Is WWIV the same way? Both display ANSI just fine, neither seem to display ANSI when it's qutoed (which is odd... cuz it's still just ANSI...)

If I could ge them to detect that it's "just an ANSI image", they'd prob work better -- but I imagine the word wrapping wouldn't work -- enig does some hackery to get that to work.

In the end maybe I'll have to just quote as ASCII by default... ugg.



--- ENiGMA 1/2 v0.0.7-alpha (linux; x64; 6.10.3)
* Origin: Xibalba -+- xibalba.l33t.codes:44510 (21:1/121)

On 08/19/17, NuSkooler said the following...

g00r00 --
I've added ANSI FSE support to ENiGMA 1/2 recently. When replying to ANSI messages coming from Mystic, the reply works well in enig, but shows up corrupted in Mystic (and maybe other ANSI supporting boards... not yet sure).
Can you take a look & see what I'm doing wrong or could do better? Here
is a sample .pkt file:

save it at 79 columns

|08 .|05ú|13ù|15Dr|07e|08am Ma|07st|15er|13ù|05ú|08.
|08 øù|05ú|13ùø |13øù|05ú|08ùø
|11 DoRE|03!|11ACiDiC|03!|11Demonic |08[|15dreamland|09.|15darktech|09.|15org|08]

--- Mystic BBS v1.12 A34 (Windows/64)
* Origin: |08--[|15!|07dreamland BBS dreamland.darktech.org (21:1/163)

On Monday, August 21st dream master was heard saying...

save it at 79 columns

The sample in that pkt file is already < 79 cols per line. Updated examples are
the same other than I added a little more to the prefix of the message (ANSI reset/etc. similar to original ANSI postings from Mystic) and encode explicitly
with CP437 as boards were choaking on ANSI as UTF-8 =(





--- ENiGMA 1/2 v0.0.7-alpha (linux; x64; 6.10.3)
* Origin: Xibalba -+- xibalba.l33t.codes:44510 (21:1/121)

I've added ANSI FSE support to ENiGMA 1/2 recently. When replying to ANSI messages coming from Mystic, the reply works well in enig, but shows up corrupted in Mystic (and maybe other ANSI supporting boards... not yet sure).

Are you post-processing the ANSI so its saved at no more than 79 columns, and if you are, does that 79 character saved ANSI properly open up in TheDraw, AcidDraw, etc? You can just truncate lines at 79 characters you have to totally post-process it and rebuild the ANSI.

That was the most annoying part of adding in that ANSI stuff when I did it.

What I did was I had it save an .ANS for each message at 79 characters whenever I saved the message, so basically the .ANS contained the identical content that was saved in the database.

Then I would load it all up in various viewers and editors to make sure the ANSI is properly reproduced by everything, as well as viewable in terminals.

It took me a bit to get it right and after I did, I tested it in some BBS software like Telegard and WWIV which doesn't directly support ANSI, but will spit out the raw 79 characters per line, meaning it actually views properly by default because it just dumps message content.

--- Mystic BBS v1.12 A35 (Windows/32)
* Origin: Sector 7 [Mystic BBS WHQ] (21:1/108)

On Friday, August 25th g00r00 was heard saying...

Are you post-processing the ANSI so its saved at no more than 79 columns,

See prev reply in other general

On Friday, August 25th g00r00 muttered...

It took me a bit to get it right and after I did, I tested it in some BBS software like Telegard and WWIV which doesn't directly support ANSI, but will spit out the raw 79 characters per line, meaning it actually views properly by default because it just dumps message content.

Currently I don't do the < 79 thing. I do normalize the ANSI though, so I can add that to the alg.

Do you mind sharing some code or at least any tips on what "getting it right" entails? What I'm currently doing works as far as displaying properly in terminals, Pablo, etc. but I'll be making quite a few adjustments to fit within
< 79.




--- ENiGMA 1/2 v0.0.7-alpha (linux; x64; 6.10.3)
* Origin: Xibalba -+- xibalba.l33t.codes:44510 (21:1/121)

Do you mind sharing some code or at least any tips on what "getting it right" entails? What I'm currently doing works as far as displaying properly in terminals, Pablo, etc. but I'll be making quite a few adjustments to fit within < 79.

What I did was to take some ANSIs and load them up in TheDraw inside of a DosBox. I saved them with a line length of 79 characters and then looked at the ANSI in a text editor to see what it looks like.

ANSI saved in this mode is what your ANSI should look like after you save it.

Basically the idea is to use a down and up cursor movement if you ever hit 79 characters in the output of the ANSI (but aren't at the end of the line from an ART perspective).

So I think what happens is the "new line" actually goes through, but then your cursor location code pops the cursor right back up to the previous line where you were before the 79 character wrap came into play. At this point the cursor is in the right spot and you're on a "new line" as far as the output goes.

I think just looking at a TheDraw ANSI saved in 79 line length will provide
the best examples of what I mean.

--- Mystic BBS v1.12 A35 (Windows/32)
* Origin: Sector 7 [Mystic BBS WHQ] (21:1/108)