« first day (1 day earlier)      last day (122 days later) » 
03:00 - 05:0005:00 - 00:00

Anonymous
03:09
I should probably turn off the rest of the commands, since they are extraneous to the bot's actual function
Anonymous
!stop
@Mego Bot terminated.
Bot started.
Anonymous
!commands
@Mego Command not found.
Anonymous
!listcommands
03:16
Commands:
module, stop, disable, enable, ban, unban, delete, alive, utc, listcommands, help, cat, read, getcurrentusers, ping, eval
2
Anonymous
Usage: !eval language "code" "input"
Anonymous
(quotes required for code and input)
!eval brainfuck "+." ""
@quartata 
Interesting.
03:21
!eval brainfuck "++++++[->++++++++<]>." ""
@ETHproductions 0
!eval rotor "<3s" ""
@quartata 9.0
@quartata Did you mean to attempt to output the byte 0x01 with that?
Anonymous
03:22
Rebooting it real quick
Bot started.
I wanted to see how it handled unprintables
@Data Hello.
It doesn't respond? Bummer.
Anonymous
It doesn't handle non-ASCII yet, so languages like Jelly and Seriously are handicapped
Anonymous
03:23
I'm still working on figuring out why the bot has such an issue with unicode
!eval rotor "NN" ""
:27225590
nailed it
!eval rotor "" "asdf"
@quartata asdf
Anonymous
Another reboot
03:27
Bot started.
Anonymous
This should make it work with Unicode
!eval rotor "" "\"
:27225639
Anonymous
@quartata `\` is used for escaping quotes. You confused it.
Bot started.
Anonymous
03:29
Escaping backslashes should now work as intended
enter PhiNotPi
Anonymous
enter Sandman
Anonymous
!eval seriously "'0,,u@xεjc" "7" "100"
@Mego Something went wrong with your request, sorry! Are you sure that language is on Try It Online?
0
8
Anonymous
derp
Anonymous
03:32
!eval seriously ",x" "[2,101]"
@Mego [2, 101]
!random
Anonymous
...
@PhiNotPi Command not found.
Anonymous
Oh right
03:32
I guess that got taken off the list.
Anonymous
Yeah I removed a bunch of the commands that were spammy and such
Anonymous
!eval seriously "G"
@Mego 0.865593015701
Anonymous
!eval seriously ",x" "2,101"
@Mego [2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100]
Anonymous
03:34
\o/
@Mego I was trying to
Anonymous
!eval seriously ",xεj" "2,101"
iconv: illegal input sequence at position 2
[2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100]
Anonymous
That's new
Are you using ChatExchage or this bot?
Anonymous
03:35
The one ProgramFOX made
also on a scale of one to ten how much is this injection heaven with one being not and ten being Blaze
Anonymous
Injection heaven?
Anonymous
It's just a simple POST request
but if you are running it through iconv
Anonymous
That's just Python3 being Python3
Anonymous
03:37
I think
Anonymous
I'm trying to find the offending code now
oh lal
!eval rotor "NN'a" ""
@quartata a
wat
what happened to the newlines?
Anonymous
Probably sillyness
03:40
Hmm
!eval rotor "<y" ""
:27225880
Shoud probably do something about empty output/stderr output
annoyingly I cannot type special chars on this tablet
Anonymous
Hmm... Unicode should be fine
Anonymous
!eval seriously "['a','b']εj"
iconv: illegal input sequence at position 9
['a', 'b']
03:45
yea unicode no worky
Anonymous
Clearly :P
Bot started.
Anonymous
Let's see what the exact issue is
Anonymous
!eval seriously "['a','b']εj"
iconv: illegal input sequence at position 9
['a', 'b']
Bot started.
03:46
Try !eval pl _ e symbol that I cannot type
Anonymous
!eval seriously "['a','b']εj"
iconv: illegal input sequence at position 9
['a', 'b']
Anonymous
Îľ
Anonymous
It's 2 bytes...
Parsing it as UTF 16?
Anonymous
03:47
No I mean it's reading it as 2 bytes, which is correct for utf-8
Bot restarted after crash.
!eval pl "3¥" ""
Anonymous
I was concerned it might be reading it as latin-1 or someting
Anonymous
!eval seriously "['a','b']εj"
Anonymous
03:49
So that doesn't work
Anonymous
!stop
oops I might have borked it
that should work fine in pl but it might throw a small error because it is not a complete if statement
Anonymous
Nah the borking was unicode
do you have a github for this?
Anonymous
Yeah, though it's not the latest commit
03:52
k
Anonymous
Just pushed latest
Anonymous
!eval jelly "1¬" ""
Bot started.
Anonymous
I keeled it temporarily
Anonymous
03:54
!eval seriously "['a','b']εj"
iconv: illegal input sequence at position 9
['a', 'b']
!eval jelly "1¬" ""
@quartata 0
Anonymous
what
Anonymous
!eval seriously "['a','b']εj" ""
03:55
Whoa. That worked.
iconv: illegal input sequence at position 9
['a', 'b']
!eval pl "1¥a" ""
given is experimental at /opt/pl-lang/pl.pl line 88.
1
lol wat
I think it doesn't care for your language @Mego.
Bot started.
Anonymous
03:57
!eval seriously "['a','b']εj"
iconv: illegal input sequence at position 9
['a', 'b']
oh leI just realized the yen sign is else not if. It has been so long since i have used pl
Anyways it seems to wok for some but not all chars
Is the yen sign one byte?
shouldnt be
Anonymous
Nope
Anonymous
!eval pl "1¥a" ""
given is experimental at /opt/pl-lang/pl.pl line 88.
1
Anonymous
04:00
But it's displaying it correctly
Anonymous
Which means it's not using unicode
Anonymous
Probably Latin-1
Bot restarted after crash.
Anonymous
!eval pl "1¥a" ""
Try the double question mark from CP437. "1?3¥4"
Anonymous
04:03
!eval pl "1¥a" ""
Anonymous
!eval seriously "['a','b']εj"
but with the double question mark in place of the regular one since I cant type it on here
should probably add a !!/alive? like smokedetector
Bot started.
Anonymous
I really need a restart command more than anything :P
Anonymous
!eval seriously "['a','b']εj"
04:04
Bot restarted after crash.
Bot started.
Anonymous
!eval seriously "['a','b']εj"
iconv: illegal input sequence at position 9
['a', 'b']
!eval jelly "1¬" ""
@quartata 0
Bot started.
Anonymous
!eval seriously "['a','b']εj"
04:06
unicode sucks
Anonymous
Agreed
iconv: illegal input sequence at position 9
['a', 'b']
Anonymous
Wait a second
Anonymous
That's not output from the bot
Anonymous
I don't think
04:07
Oh you know what
Seriously's TIO encoding is proably busted
dennis uses iconv for that
Anonymous
Yeah
Anonymous
@Dennis we need your help in here
Ty the doublr question mark from CP437 with pl. 1?3¥4
*try
Anonymous
Double question mark?
yeah
Anonymous
04:09
@Dennis something is breaking with the backend of TIO when the bot does a POST with utf-8 data
here ill get you the codepoint
cannot type it
You do you perform the POST?
Anonymous
Yeah
You -> How
Anonymous
url = u"http://{}.tryitonline.net/cgi-bin/backend".format(lang)
req = u"code={}&input={}&debug=on".format(urllib.parse.quote(code, encoding='utf-8'), urllib.parse.quote(cinput, encoding='utf-8'))
return requests.post(url, data=req).text[33:]
Anonymous
04:11
urllib.parse.quote(s) is equivalent to encodeURIComponent(s) in JS
Apparently not.
:P
sorry not double inverted
168
Anonymous
!eval pl "1¿3¥4"
given is experimental at /opt/pl-lang/pl.pl line 88.
3
Anonymous
That works because it's also in latin-1
Anonymous
04:13
It seems to only like latin-1 chars
Anonymous
Hmm...
Bot started.
Anonymous
!eval python urllib.parse.quote(u'ε',encoding='utf-8')
Bot restarted after crash.
@Mego %CE%B5
Try the fancy e and give it a decimal as input
04:15
Could you check what you are sending to the server? Should be code=%5B'a'%2C'b'%5D%CE%B5j&input=&debug=on for ['a','b']εj.
epsilon right
Anonymous
encodeURIComponent('ε')
"%CE%B5"
Bot started.
Anonymous
!eval seriously "['a','b']εj"
iconv: illegal input sequence at position 9
['a', 'b']
04:16
do !eval pl epsilon "5.0"
Anonymous
code=%5B%27a%27%2C%27b%27%5D%C3%8E%C4%BEj&input=&debug=on
that will confirm whose problem it is
Anonymous
!eval pl "ε" "5.0"
given is experimental at /opt/pl-lang/pl.pl line 88.
5.0
given is experimental at /opt/pl-lang/pl.pl line 88.
5
Anonymous
Huh... Is it an issue with the quotes, then?
04:17
k its dennis's fault
5 is the correct output it mwans that the input was floating point
Bot started.
Anonymous
!eval seriously "['a','b']εj"
iconv: illegal input sequence at position 9
['a', 'b']
so seriously has something going wrong
on tios end
!eval cjam "'4"
04:18
@Dennis 4
Quote works.
Anonymous
I made ' a safe character in the quoting now
Anonymous
...it's because the POST string should be CP437-encoded, not UTF-8, right?
But epsilon worked for pl...
epsilon = looks_like_number(
Anonymous
%C3%8E%C4%BE isn't the same as %CE%B5
04:21
@Mego That's not the right string. %C3%8E%C4%BE should be %CE%B5.
ninja'd
But why did it work for pl then?
Anonymous
I dunno
Anonymous
!eval python urllib.parse.quote('ε',encoding='utf-8')
Bot restarted after crash.
Anonymous
!eval python urllib.parse.quote(u'ε',encoding='utf-8')
04:23
@Dennis How exactly does tio do the encoding conversion?
Does seriously do something different?
encodeURIComponent(code.value)
Anonymous
!alive
Anonymous
!stop
Anonymous
The bot seems to be suffering from critical existence failure
Bot started.
Anonymous
04:24
!eval python urllib.parse.quote(u'ε',encoding='utf-8')
@Mego %C3%8E%C4%BE
Anonymous
!eval python urllib.parse.quote(u'ε')
@Dennis I mean the step where it goes from unicode to cp437
@Mego %C3%8E%C4%BE
Anonymous
!eval python urllib.parse.quote(u'ε',encoding='cp437')
04:25
Bot restarted after crash.
This seems to be an issue afflicting only Seriously.
Anonymous
!eval python urllib.parse.quote(u'ε',encoding='utf-16')
@Mego %FF%FE%CE%00%3E%01
Anonymous
!eval python urllib.parse.quote(u'ε',encoding='ucs2')
Bot restarted after crash.
04:25
This has nothing to do with Seriously. %C3%8E%C4%BE is the UTF-8 encoding of the UTF-8 encoding of epsilon.
If that was the case it wouldnt have worked for pl.
That is the case.
Anonymous
!stop
@Mego Bot terminated.
encodeURIComponent(unescape('%CE%B5'))
"%C3%8E%C2%B5"
04:26
@Mego try gs2
Bot started.
Anonymous
!eval python urllib.parse.quote(u'ε',encoding='utf-8')
@Mego %C3%8E%C4%BE
Anonymous
I don't get why it's doing the encoding twice...
Anonymous
!eval seriously "['a','b']εj"
04:28
iconv: illegal input sequence at position 9
['a', 'b']
Anonymous
@quartata I have no idea how to use gs2
What happens when you take the u off the string
@Mego just do eval gs2 epsilon
Anonymous
@quartata It crashes
Anonymous
!eval python urllib.parse.quote(u'ε')
@Mego %C3%8E%C4%BE
Anonymous
04:30
!eval gs2 "ε"
iconv: illegal input sequence at position 0
Traceback (most recent call last):
File "/opt/gs2/gs2.py", line 299, in run
self.evaluate(tokenize(self.code))
File "/opt/gs2/gs2.py", line 169, in tokenize
if prog[0] in '\x30\x31\x32': # set mode
IndexError: string index out of range
Why does my Python not have urllib.parse?
Anonymous
Because that's python3
Anonymous
For python2 it's just urllib
I tried 2 and 3. Neither works.
04:30
Wait what if this is happening in programs with no input
Anonymous
So python3 urllib.parse.quote(s) -> python2 urllib.quote(s)
Anonymous
!eval seriously ",εj" "['a','b']"
iconv: illegal input sequence at position 1
['a', 'b']
How does your bot communicate with chat?
@quartata That has nothing to do with it. It is an encoding issue.
Anonymous
Using the ChatExchange thingy
Anonymous
04:32
It's ProgramFOX's bot, with this module added
ChatExchange probably sends UTF-8.
If it worked in pl then I don't understand why it doesn't work for the others
Anonymous
Yeah, but that should be good - there isn't a double-encoding going on that I can see
Can you set up an echo command?
04:34
@Mego My Python refuses to cooperate. Try encoding='iso-8859-1' or whatever it's called in the module.
So we can see what the strings getting returned from chatexchange
Anonymous
!eval python urllib.parse.quote(u'ε',encoding='iso-8859-1')
Bot restarted after crash.
Anonymous
Stupid spaces
Anonymous
!eval python urllib.parse.quote(u'ε',encoding='iso-8859-1')
Anonymous
04:36
UnicodeEncodeError: 'latin-1' codec can't encode character '\u013e' in position 1: ordinal not in range(256)
er do you need the python code in quotes?
Anonymous
!eval python '\u013e'
Anonymous
Not for python; it just directly eval's it
Anonymous
!eval python u'\u013e'
Anonymous
!alive
04:37
That's ľ. WTF?
Anonymous
stupid freaking bot...
Bot started.
Anonymous
!eval python '\u013e'
@Mego ľ
Anonymous
!eval python urllib.parse.quote_from_bytes('ε')
04:39
Bot restarted after crash.
Anonymous
!eval python urllib.parse.quote_from_bytes(b'ε')
Bot restarted after crash.
Anonymous
>_<
Anonymous
!eval python u'ε'.encode()
Bot restarted after crash.
04:40
ord(epsilon)?
!eval python (╯°□°)╯︵ ┻━┻
Anonymous
!eval python u'ε'.encode(encoding='utf-8')
Can you do ord(epsilon)?
Anonymous
!eval python ord(u'ε')
Crash?
Anonymous
04:42
what in the everloving noodles is going on
Looks like my table flip killed Data... :(
python 3 that's what
Anonymous
!eval python u'\u03B5'
I have an idea I'd like to test.
As soon as you resurrect it.
Bot restarted after crash.
@Mego ε
Anonymous
04:43
!stop
Anonymous
\o/
Anonymous
It sort of worked
Bot started.
!eval gs2 ε
Traceback (most recent call last):
File "/opt/gs2/gs2.py", line 299, in run
self.evaluate(tokenize(self.code))
File "/opt/gs2/gs2.py", line 169, in tokenize
if prog[0] in '\x30\x31\x32': # set mode
IndexError: string index out of range
Anonymous
04:44
quotes
Why does that need quotes? Your Python command don't have any.
Anonymous
Because Python is special
!eval gs2 "ε"
iconv: illegal input sequence at position 0
Traceback (most recent call last):
File "/opt/gs2/gs2.py", line 299, in run
self.evaluate(tokenize(self.code))
File "/opt/gs2/gs2.py", line 169, in tokenize
if prog[0] in '\x30\x31\x32': # set mode
IndexError: string index out of range
Anonymous
I made it handle Python separately, without needing to parse quotes with regex
04:46
Ah, OK.
FYI gs2 always throws that error with blank programs
so it's dropping the unicode char
Yeah, because iconv cannot handle it.
It should though...
!eval cjam "ε"
Exception in thread "main" java.lang.IllegalArgumentException: Î
at net.aditsu.cjam.Ops.get(Ops.java:51)
at net.aditsu.cjam.Block.parseOp(Block.java:290)
at net.aditsu.cjam.Block.parse(Block.java:158)
at net.aditsu.cjam.Block.parse(Block.java:90)
at net.aditsu.cjam.CJam.main(CJam.java:238)
Anonymous
04:48
code=%C3%8E%C4%BE&input=&debug=on
Anonymous
It's still double-encoding it, I think
It is.
Anonymous
But I don't see where
wait so is the iconv error from tio or the bot
The iconv error is from TIO, because the bot double-encodes the string.
04:49
that is what I thought.
Bot started.
Anonymous
!eval seriously "ε"
@Mego iconv: illegal input sequence at position 0
Anonymous
It's happening in the ChatExchange part of the bot
Anonymous
The code the module is getting is '"Îľ"'
Anonymous
04:52
!eval python urllib.parse.unquote('%CE%B5')
@Mego ε
Anonymous
!eval python urllib.parse.unquote('%C3%8E%C4%BE')
@Mego Îľ
Anonymous
!eval python '%{}%{}'.format(map(ord,urllib.parse.unquote('%C3%8E%C4%BE')))
Bot restarted after crash.
04:54
Is ChatExchange using utf-16?
also was it made for python 3
Anonymous
!eval python '%{}%{}'.format(map(lambda x:hex(ord(x))[2:],urllib.parse.unquote('%C3%8E%C4%BE')))
Bot restarted after crash.
Anonymous
utf-8, python3
Anonymous
Err
Do you think it is something that SE is doing then?
Anonymous
04:56
!eval python list(map(ord,urllib.parse.unquote('%C3%8E%C4%BE')))
Bot restarted after crash.
Coud be getting encoded funny when sent down the websocket
Anonymous
Right now I'm just working on undoing it
03:00 - 05:0005:00 - 00:00

« first day (1 day earlier)      last day (122 days later) »