Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - adarqui

Pages: 1 ... 987 988 [989] 990 991 ... 1505
14821
Progress Journals & Experimental Routines / Re: ADARQ's journal
« on: April 23, 2012, 07:32:19 am »
damn my neighbor died... nuts.... i been coding all night, watching as cops investigate.


ok so i finished this alloc tracker thing, freakin nuts..

when i type 'alloc', it scrolls for like, literally 30 seconds.. so much stuff allocated/free'd for my bot.. then, say i type 'off', which is supposed to clean everything up, then i type 'alloclist' (only show stuff still allocated), here's the final output:

debug: bot_evhook_console: read [alloclist]
gah
bot_alloc_print: Printing alloc'd objects:
        0x817c0b0 => bot_misc.c str_unite 50 (strdup) ::: active=1 addr=0x817c0b0 size=9 data=[/darqbot/], ctime=Mon Apr 23 11:20:08 2012
        0x817bfd8 => bot_misc.c str_unite 50 (strdup) ::: active=1 addr=0x817bfd8 size=14 data=[/darqbot//conf], ctime=Mon Apr 23 11:20:08 2012
        0x817b8d8 => bot_global.c bot_global_conf_parse 83 (strdup) ::: active=1 addr=0x817b8d8 size=9 data=[/darqbot/], ctime=Mon Apr 23 11:20:08 2012
        0x8191478 => bot_misc.c str_unite 50 (strdup) ::: active=1 addr=0x8191478 size=9 data=[/.darqbot], ctime=Mon Apr 23 11:20:08 2012



sooo..... 1204912409120 lines of allocated data scrolling, then i turn it off, and only ~5 ptr's allocated.

BEAST

MODE

SONSON.

that basically means their are no leaks.. anyway, ill have to stress it more and see what remains, and keep this process going as i add more components/modules etc.. to make sure i have 0 memory leaks.

it took longer than expected because i had to re-organize alot of code, specifically the module interface.. i need to reorganize it further in a few days, to allow for multiple servers.

peace

14822
News, Announcements, & Suggestions / Re: monthly newsletter
« on: April 22, 2012, 03:55:00 am »


 :P


almost done with it, been working on it for a while..

:F

14823
Boxing / Re: good performance by marquez
« on: April 21, 2012, 03:07:56 pm »
ya he's impressive considering his age/how many battles he's been through, but he's not going to last much longer.. not enough power/strength, though his boxing "Skills" are ridicuous.

14824
Progress Journals & Experimental Routines / Re: ADARQ's journal
« on: April 21, 2012, 06:21:58 am »
Only four hours sleep last night, but I felt great all day.

Finished some annoying code, wasn't into it, but I needed to get it done:
- Added smf posting feature from the bot, now I can post into threads from the live chat, or pipe stuff/info to the live chat.. I was stuck for a good while because libcurl's cookie engine was giving me problems.. I still don't know if there's a way to get it working properly using libcurl entirely by itself. Instead, im using libcurl initially to get the cookies, then i am keeping track on my own to supply the updated cookies.. libcurl is receiving those cookies, but it won't supply them in the HTTP headerss for responses.. must be something i'm doing wrong but i havn't figured it out.. anyway, i fixed that by keeping track of the updated cookies and supplying them manually.

other than that, i'm writing this cool color/ascii generator for the bot, which is pretty fun.. really "gays up" the live chat, with rainbow colors everywhere, or colors you specifiy in the trigger:

^rainbow(pink,white,cyan) .. <-- just those ... ^rainbow(def) <-- all colors

anyway, that func is going to be fun, already is actually, color bombing irc is entertaining.. i have to setup the ascii thing tomorrow, which will have a bunch of fonts that i setup for my gridwindow code.. so i plan to make my own figlet/sysvbanner style ascii generator but entirely for irc, using some of my gridwindow code from a month back.

so ya my todo for tomorrow is:

finish ^smf, ^rainbow, script parser

start ^rant, ^grid


this project has a mix of really useless but fun components, and then some really intense/advanced components.. hard to get bored with such a scenario.

i still have absolutely no clue how im going to emulate archs, create little vm's for instructions/testing, for various cpu architectures.. need to find something that already exists and then integrate it.. last resort i'll just do intel support initially.

peace


14825
Progress Journals & Experimental Routines / Re: ADARQ's journal
« on: April 21, 2012, 05:07:25 am »
Why is training Ur mind far more interesting and fun :(

because everything we perceive to be important, or even the ability to perceive, stems from a mass of tissue of which we only utilize at most ~8=10% of it's  total capacity. furthermore; the mind outlives the body, in most cases.

Quote
Looks like ur denouncing wat u used to do :(

nope, i'm glad i did what i did, but i finally got tired of it.. lost the luv.






Looks like ur denouncing wat u used to do :(

wrong.

correkt.








Thanks Darq. I was online on IRC, but you weren't. Maybe next time we'll sync. My dad is doing SQL Scripts right now and managing the databases of the biggest insurance company in Romania. He's the only man doing that and no one else there know how to do them :)) so he rarely takes holidays. He studied cybernetics in College.

yo i saw you, my nick has "ng" in it.. i change my nick alot.. i was away when you joined.

damn it sounds like your dad has a pretty important gig, nice..

i hate sql, im starting to love this nosql movement, mongodb is DIESEL.






The mind and body feed off of each other for a great deal.
Neglecting either one would be making )#*%& excuses.

true but what is neglect? is the lack of lifting heavy weights neglect? is the lack of jumping/sprinting/etc neglect?

is simply walking or running with my dog enough to 'keep the body happy'?

how much is too much, how much is too little?


Quote
You still eatin' right/anorexic b?

eating decent.. pretty broke but it's all good, once i get a new job & get paid better, im going to pig the fu*k out.

;d



Quote
Keep at it and btw, please don't create any kind of AI that will destroy the world, k?

pce

lmfao, unfortunately i don't need to create a.i that will destroy the world, that is happening already..

however, if i could create a skynet-like system, that gets rid of religion and replaces it with science, that'd be worth it

:F








i been busy last few days trying to figure something out, that's why i didnt post.. will post tonight in a bit.

word :D

14826
LanceSTS's Performance Blog / Re: COME TO IRC LANCE
« on: April 18, 2012, 12:06:45 pm »
he came to irc gutts, i sawr it.

btw, dreyth aint not shit bich relax. if dreyth and yourself entered a physique modeling contest dreyth would win effortlessly relax, glittered up and soap suds, all day, err day.

14827
Progress Journals & Experimental Routines / Re: ADARQ's journal
« on: April 18, 2012, 06:39:29 am »
btw forgot to paste the second half of reply, about your dad's 40 years hardcore coding.. sick man.. what kind of stuff did/does he do?

i don't know anything about fortran/cobol.. even though scheme is "old", it is incredibly powerful still to this day.. alot of android software is written in lisp as a current example.

scheme is ridiculously clean and lacks the bulk of common lisp, that's why i chose to go with it.. also, scheme is pretty much the language of artificial intelligence & genetic programming, so I have to get really sick with scheme, because one of the avenues I want to get really good at, is a.i. , genetic programming, neural networks etc.

this "bot" thing i'm doing, will incorporate alot of that stuff eventually.. plus I want it to become some kind of "social programming environment", where people can mess around and code WHILE chatting together.. it already has that functionality now that gnu guile is incorporated.. i'm pondering whether or not i should incorporate something like golang/python etc, various langs.. the only problem is irc's message size & those languagages don't let themselves to "one liners", well C does.. int main() { int x; for(;;) { puts(""); } } etc.. but python definitely no, goland probably not.. C/lisp can fit nicely on one line..

to incorporate golang/python etc, i'd probably have to do like..

^golang start loop1
... typing code multiple lines....
^golang end

and then you could do something like:

^goland loop1 <args..>

sounds ez actually.. i'll probably incorporate it fairly soon.. not sure about C though, i don't know if i want code compiling on my system.. python/golang could run right from an interpreter, it's definitely cleaner and safer.

peace man

14828
Progress Journals & Experimental Routines / Re: ADARQ's journal
« on: April 18, 2012, 06:26:40 am »
It's nice you started working with Unix, Lisp, GNU and so on. I've been wanting to get into Linux work for some time, but only grazed the basics and never really started to do advanced stuff.

About the fact that Lisp is one of the oldest programming languages, I have two books about Fortran (oldest) and Cobol (third oldest) that belong to my dad who used to code hardcore 40 years ago :))

Keep logging this stuff, it's interesting.

PS: I still am kinda sad you stopped training.

bleh reposting all messed up, i forget i have to refresh links after each post, or it messes up.. nogui>*

word nightfly, thanks I will keep posting it, glad you find__
                 it somewhat interesting.. and as always, you are always______
                 invited to the live chat & any kind of talk, including_______
                 programming/unix/etc.. more so, I could even give you a shell
                 on one of my machines, if that would help you to better learn
                 nix/coding.. I've actually been into nix/programming for a___
                 long time, i just took 5 years off to persue my HEAD AT RIM__
                 GO GET IT mission.___________________________________________
                 _____________________________________________________________
                 don't be sad that i stopped training, because i didn't.. i'm_
                 training harder than ever, except, now i'm training my mind,_
                 which is far more interesting and fun.. took me a while to___
                 realize that.________________________________________________
                 _____________________________________________________________
                 peace man & stop by IRC some time.. im on there alot coding,_
                 boring people to death ;d____________________________________

ill repost yet again if that came out really wrecked.. heh

14829
Progress Journals & Experimental Routines / Re: ADARQ's journal
« on: April 18, 2012, 06:25:55 am »
It's nice you started working with Unix, Lisp, GNU and so on. I've been wanting to get into Linux work for some time, but only grazed the basics and never really started to do advanced stuff.

About the fact that Lisp is one of the oldest programming languages, I have two books about Fortran (oldest) and Cobol (third oldest) that belong to my dad who used to code hardcore 40 years ago :))

Keep logging this stuff, it's interesting.

PS: I still am kinda sad you stopped training.

word nightfly, thanks I will keep posting it, glad you find it somewhat interesting.. and as always, you are always invited to the live chat & any kind of talk, including programming/unix/etc.. more so, I could even give you a shell on one of my machines, if that would help you to better learn nix/coding.. I've actually been into nix/programming for a long time, i just took 5 years off to persue my HEAD AT RIM GO GET IT mission.

don't be sad that i stopped training, because i didn't.. i'm training harder than ever, except, now i'm training my mind, which is far more interesting and fun.. took me a while to realize that.

peace man & stop by IRC some time.. im on there alot coding, boring people to death ;d

14830
Progress Journals & Experimental Routines / Re: ADARQ's journal
« on: April 18, 2012, 06:22:23 am »
hey rix, fuck the fuck off. thanks.




Today I got the movie-script parser working.. The next step, is to "find names" within each movie script, and replace them with a special tag that I can use to substitute IRC nicknames with, for example:

14:49 < warm-n9> the last step would be to try and extract names
14:49 < warm-n9> and replace them with <name>
14:49 < warm-n9> which is kind of difficult
14:49 < warm-n9> for example
14:49 < warm-n9> COLE: Sup
14:49 < warm-n9> MIKE: Hey james
14:49 < warm-n9> MARRY: That's James Cole

14:50 < warm-n9> would need to become:
14:50 < warm-n9> COLE: Sup
14:50 < warm-n9> MIKE: Hey <name:COLE>
14:50 < warm-n9> MARRY: That's <name:COLE>

So I need to find, based on the CAPITALIZEDNAMES, other variations of the names within the text.. shouldn't be too hard.


Other than that, I got some lisp functions working nicely in the bot's alias modules.. this is pretty sick.. check it, here's the raw bot + guile-scheme code:

dns code:
^a(add) dns \|^e dns result: \|^guile (define p (gethost "$$2")) (define phost (vector-ref p 0)) (define pip (car (vector-ref p 4))) (display phost) (display "=") (display (inet-ntoa pip))

cat1 code:
^a(add) cat \|^e\|^guile (define (read-string char-set input-port) (let ((char (peek-char input-port))) (if (eof-object? char) char (list->string (let loop ((char char)) (if (or (eof-object? char)) '() (begin (read-char input-port) (cons char (loop (peek-char input-port)))))))))) (define p (open-input-file "$$2")) (display (read-string 's p)) \|^multi


here's it being used in an example:

06:00 < warm-n9> best bot ever.
06:03 < warm-n9> ^a cat /etc/motd
06:03 < dko> Linux darqbot 4.3.19.12202011 #2 SMP Thu Dec 25 01:28:15 EST 0BC i686
06:03 < dko> Welcome to the darqbot ircOS.
06:03 < warm-n9> RAPED.
06:03 < warm-n9> AS
06:03 < warm-n9> FUCK
06:03 < warm-n9> ^a cat /etc/passwd
06:03 < dko> root:x:0:0:root:/root:/bin/bash
06:03 < dko> test:test:1000:1000:test,,,:/:/bin/darqbotsh
06:03 < warm-n9> LMFAO
06:07 < warm-n9> ^a dns adarq.org
06:07 < dko> dns result: adarq.org=199.193.251.107


notice the $$2's in the code? args.

So, as you can see, ^a dns <args> AND ^a cat <arg> now run lisp code with the arguments i specify.

I need to develop a bunch of alias codes for things like:
 - ls/uname/stat/find/grep/rm/mkdir/etc
 - adding "quotes" to flat file db's, such as ^a addquote <a funny quote> .... ^a getquote

I have a pretty crazy idea i'm going to implement soon, mod_exp, which runs vulnerable functions in the child.. such as:
 ^exp(fmtstring) <data>
 ^exp(offby1) <data>

HEH.. once i program the unicode funcs, people will be able to pass null bytes to it, and really muck around, trying to gain shell access to the bot.. i want to put a bunch of vulns in that module for fun.. only thing I need to figure out is how to possibly compile without canary/ASLR etc.

Finally, if I am boring you, I am so very sorry. Seriously, I will go do some dunk on 9' and act stupid on camera, if that will make it up to you.

pC folks.

14831
Progress Journals & Experimental Routines / Re: ADARQ's journal
« on: April 17, 2012, 05:14:31 am »
"UNIX is a companion which no future can depress, no crime can destroy, no enemy can alienate it and no nepotism can enslave." Ropo Oguntimehin

"GNU's purpose is to replace a proprietary mind with an open one." Malcolm Forbes

"Develop a passion for coding. If you do, you will never cease to grow the stack." Anthony J. D'Angelo, The DoD Blue Book

I was reminiscing of those quotes as i installed gimp.tbz and it's dependencies manually, behind a firewall + proxy, and two routers (wifi + cable), much like when I installed netbsd 1.3.1 from 31 floppies.

Next time someone starts blabbing about the "college experience", proceed to int 0x80 them & go on about the unix experience (TM).

relax.

14832
Progress Journals & Experimental Routines / Re: ADARQ's journal
« on: April 17, 2012, 05:10:52 am »
Spent a few hours debugging, trying to figure out why my bot would zombie, even though I had a SIGALRM hook & alarm() set, also waitpid'n etc.. it was blocking in write(), because it writes to a pipe that was getting full on certain tests.

Cleaned environ when passed to ^guile, was resulting in info leaks, need to protect LD_PRELOAD/LD_LIBRARY_PATH variables.

Added fake environment variable, bot_t { dlist_t *dl_environ; }

Added something to build "argv[]" for my ^alias trigger, for example:


"dns" alias example:
05:03 < n9> ^a(get) dns|^e
05:03 < vkx> ^e|^guile (display (gethost "$$2"))

using it:
05:03 < n9> ^a dns adarq.org
05:03 < vkx> #(adarq.org () 2 4 (3351378795))

$$0-$$9 = argv[0]=argv[9] .. argv[0] = ^a, argv[1] = dns, argv[2] = first argument

Tomorrow I need to make it parse strings properly, instead of just tokenizing everything into words based on spaces.. for eample
:

^a blah one two three four <-- it only works on this
^a blah one "two three" four <-- it doesn't work on this, $$3 would be "two, $$4 would be three".. need to parse it into one arg
v, ie, $$3 = two three

Finally I need to add $$@0-9, ie, $$@0 = every argument.. $$@1 = every argument except the 0th. and so on

im crazy tired today, but got alot done.. 1 week on 5 hours sleep per night, max.

pC

14833
Progress Journals & Experimental Routines / Re: ADARQ's journal
« on: April 16, 2012, 06:59:10 am »

oh one more thing, this is pretty cool.. using the "nop input module", I can turn an irc/live chat session into a guile session.
. ^nop causes the bot to "add stuff" to each line it receives, so normal chat text, can become processed using my triggers..

for example

^nop(ina) \|^e\|^guile <-- this causes guile to append \|^e\|^guile to each line received, then processes it.. so..

typing:

(display 1) turns into (display 1) |^e |^guile, which causes guile to parse (display 1) resulting in 1 being outputted..

here's an example:

04:40 <@z> (display (list "hey"))
04:40 < xdd> (hey)
04:41 <@z> (display (length (list "hey")))
04:41 < xdd> 1
04:41 <@z> (display (%global-site-dir))
04:41 < xdd> /usr/local/share/guile/site
04:43 <@z> (display (remainder 10 4))
04:43 < xdd> 2
04:47 <@z> (number->string (+ 1 2))
04:47 <@z> (display (number->string (+ 1 2)))
04:47 < xdd> 3
05:01 <@z> (procedure-documentation display)
05:01 <@z> (display (procedure-documentation display))
05:01 < xdd> #f
05:01 <@z> (display (procedure-documentation 'display))
05:01 <@z> (display (procedure-documentation sort))
05:01 < xdd> #f
05:02 <@z> (display (random))
05:02 <@z> (display (random 4))
05:02 < xdd> 3
05:02 <@z> (display (random 100000))
05:02 < xdd> 50621
05:02 <@z> (display (random 10000000000000))
05:02 < xdd> 6312491691827
05:02 <@z> (display (random 100000000000000000000000000000000000000000000))
05:02 < xdd> 78306622436512418053130718978773471728041129

pretty cool.. ^nop is real powerful.. ^modfiy will be insanely powerful once i code it.
~

14834
This.. I was actually serious about reading this post....
(while my office blocked youtube...)
so i sent this video to my iphone, hide my self in the toilet to watch this thread.... and...

lol.... :uhhhfacepalm:


BTW, if i am not mistaken.. Adarq re posted in the program review section stating that, BOINvert is actually legit now right?

All this making fun bointard post is messing with people mind...

i never said it was legit, i just said it had more potential now that shawn was on board.. i can't say something is legit or not unless i research it extensively, which i didnt with boingvert.. i havnt been following what happened with boingvert, so no clue how that turned out.. from what i've seen, it seems to be getting some lackluster feedback. please correct me if i'm wrong.

pC

14835
Progress Journals & Experimental Routines / Re: ADARQ's journal
« on: April 16, 2012, 06:44:56 am »
hmm.. so what did I do today.

I modded darqbot to run in a chroot jail, this actually took some work;
- chroot/setre[U,G]id etc
- modify a bunch of paths in darqbot
- chdir twice to clean envs (didn't want to have env variables null, but wanted to wash them a bit)
- mount binded fs's inside the chroot jail, read only, ie:
   mount -B /lib /darqbot/lib
   mount -o remount,ro /darqbot/lib

   had to do that with dev, usr, lib.. copied over etc manually, only a few files

   so this means, darqbot now has a little fs people can play with through whatever mechanism, most likely the ^guile interface (scheme lisp interpreter).. since that scheme code can modify files/do all sorts of stuff, it needs to be jailed for basic protection.

- modify guile source slightly, to get rid of (system*) & (kill), as well as fork/exec etc..
- lisp processing and my pipe processing now goes hand in hang, you can do things like:

  ^e |^guile (display (gethost "aol.com"))) |^caps|^fgcolor(pink)

  so lisp and my stuff can all be processed on the same line..

  next step is to integrated my modules INTO guile, that will be very nice.. then i can do things like:

  ^e |^guile (^fgcolor pink) (gethost "aol.com")

  plus it needs to understand my bot_t structure, so that power/admin users can actually use (system* ..) etc

  etc

- having a slight problem with zombie procs in guile.. reduced 99% of the zombies by redoing my sigchld/sigpipe/sigsegv signal handlers & set up an alarm to kill any guile modules after 5 seconds, but 1 scenario is still causing a problem...




here's an example loop being processed: (dko is the bot)

02:56 < n9> ^e|^guile  (let loop ((n 1)) (if (<= n 10) (begin (display n)(newline) (loop (+ n 1))))) |^multi
02:56 < dko> 1
02:56 < dko> 2
02:56 < dko> 3
02:56 < dko> 4
02:56 < dko> 5
...

without ^multi, it would print on the same line.. beast.

there's all sorts of limits and flood protection mechanisms in place, all of which can be set with the ^var trigger.




tomorrow i need to work on:

- adding arguments to my alias function

  my idea so far is, a very basic argument structure, shell-like:


^alias test ^e $1 | ^rot13 | | hi $3 | yo $2
^a test one two three
 -> xg hi three yo two
etc..


another example, incorporating a lisp func:
^a(add) dns ^e | ^guile (display (gethost "$1")) <-- set the alias
^a dns adarq.org <-- resolve adarq.org

the alias would expand to

^e | ^guile (display (gethost "adarq.org"))




really happy about implementing guile.. darqbot is now 1000000x more powerful.


damn this is my longest post in a while.. i write stuff like this down every day in my code-journal thing, so im going to move it into my actual journal..

;d

Pages: 1 ... 987 988 [989] 990 991 ... 1505