>>5
UR KOOL I BET U RIDE A MOTORCYCLE 2 UR CLASSES @ MIT
Name:
Anonymous2008-02-15 8:38
Interesting. And does it work?
Name:
Anonymous2008-02-15 9:32
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* *
* I AM CURRENTLY HAVING HOT GAY SEX WITH YOUR MOTHER *
* *
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
Name:
Anonymous2008-02-15 9:38
>>8
I cracked Abelson's password with it instantly.
It takes an xbox hueg amount of disk space though. Considering the disk usage of a rainbow table and the small size of each key/value pair, I'm not sure if it's really a better approach than, say, a btree, but I don't have enough terabytes to test both methods with an equal keyspace. A btree is somewhat quicker to search, though.
By the way if you're only using alphanumeric characters, or if you can figure out how to manipulate the tripcodes, you can ditch the htmlspecialchars stuff and feed the input straight into crypt. It's noticeably faster.
Name:
Abelson!NobodyCracksMyTrip2008-02-15 10:01
Nobody crack my tripcode
Name:
Anonymous2008-02-15 10:52
>>10
You should put up a web frontend. Mainly because I'm a lazy cunt with 3Gbs free space and a Pentium M ;_;
>>13
I'd rather have people work for themselves. If every retard can crack a tripcode it sort of defeats any point to having them. Kinda like if a 'tripcode' field was added to the posting form.
>>18
Too long. I suspect using something like John the Ripper's DES code, or pretty much anything lower-level than fcrypt, would have helped tremendously.
It's not like my computer really does anything else most of the time, anyway. Kind of like seti@home, but with a usable output.
Name:
Anonymous2008-02-15 19:15
Couple of questions:
- what range of characters is valid in tripcode input?
- what range have you actually used in your rainbow table, and how much space does it take?
- what range of characters is valid in tripcode input?
You can use all 128 different values, from 0 to 127 -- you will have to set the eighth bit for some of them though, or they will be replaced/ignored.
Name:
Anonymous2008-02-16 6:39
patching file HashAlgorithm.cpp
Hunk #1 FAILED at 57.
1 out of 1 hunk FAILED -- saving rejects to file HashAlgorithm.cpp.rej
patching file HashAlgorithm.h
Hunk #1 FAILED at 10.
1 out of 1 hunk FAILED -- saving rejects to file HashAlgorithm.h.rej
patching file HashRoutine.cpp
patch unexpectedly ends in middle of line
Hunk #1 FAILED at 20.
1 out of 1 hunk FAILED -- saving rejects to file HashRoutine.cpp.rej
I fail it.
Is rainbowcrack even able to take adjantage of several CPUs? If not, it's not worth more effort to get it to work to me.
Name:
Anonymous2008-02-16 7:24
>>35
The code is right there. Why would you even ask such a question?
Name:
Anonymous2008-02-16 7:26
>>36
Because I'm not going to put more than 10 minutes of effort in something I'll only play with for 5 minutes once it's done.
Name:
Anonymous2008-02-16 8:20
>>37
If it takes you more than ten minutes to find out if a small Sepples application is multi-threaded, maybe you shouldn't be posting on /prog/.
If you think multithreading in the application is the only way to take advantage of multiple processors, maybe you shouldn't be posting on /prog/.
Name:
Anonymous2008-02-16 9:25
>>41
I'm too lazy to start multiple processes or write a script. I would, but not for something I'll only play with for 5 minutes.
Name:
Anonymous2008-02-16 9:32
>>42
If you think you can generate usable rainbow tables in 5 minutes, maybe you should send me your computer.
Name:
Anonymous2008-02-16 10:23
what range of characters is valid in tripcode input?
Really, [\001-\377]. DES is only 7-bit, but remember there are side effects of using high-bit characters in the second and third positions because they potentially affect the salt used.
what range have you actually used in your rainbow table, and how much space does it take?
I'm using about 400 GB for tables. This includes all 0-4 character tripcodes, all 5-character [A-Za-z0-9./], and an assortment of a bunch of other tables that don't really go together. e.g. I started doing all 5 and 6 characters, but calculated how much it would take for all of them and realized I don't have the disk space (or enough CPUs to do it in an even remotely decent length of time) so I just made some of the tables for 6-8 characters. So really, I could do a lot better but it'd take much greater resources.
>>45
I think he means DES works on the lower range of bytes [00,7F]
Name:
Anonymous2008-02-16 23:27
>>46
Hence the ``wat''. That doesn't make any sense and isn't true.
Name:
Anonymous2008-02-17 0:25
>>47
DES uses a 56-bit key. crypt() takes the key from the lower 7 bits of each of 8 characters.
Tripcodes, on the other hand, have 58 meaningful bits - the 56 used for the key, and the full bytes in the second and third positions which make up the ``salt'', minus 14 bits of overlap.
Name:
Anonymous2008-02-17 0:28
Who still uses DES? Just calculate the SHA-1 sum of the concatenation of the "password" and some random, secret block of bytes on the server.
Every image/textboard on the internet, because it's the standard.
Name:
Anonymous2008-02-17 6:52
>>48
That makes more sense. It's a quirk specific to crypt, not DES.
>>49
SHA-1 is less secure than DES. DES is considered insecure because its keyspace is small enough to bruteforce in reasonable time, but SHA-1 has feasibly exploitable cryptographic vulnerabilities.
Having said that, using a block cipher as a hashing function is always a bad idea, but unfortunately, it has become the standard.
I'll settle this finger/thumb thing. Historically, the word "finger" has referred to any of the digits on on a hand, OR any of the digits on a hand except the thumb. So a thumb can be a finger or not depending on what you want to say. The definition that includes the thumb is the older, so I, personally, would consider it to be the "more correct" one.
>>62 >>63
"...the ancient Indo-European word for five was 'penkwe,' which became 'pente' in Greek, 'quinque' in Latin and 'panca' in Sanskrit. One can immediately see surface similarities between 'penkwe' and the Indo-European roots for fist, 'pnkwstis' and finger 'penkweros.'"
Not that it's at all hard to patch it by hand. It's nothing but adding a function and inserting it into the list of available hashes. If you're even intending to run this sort of thing in the first place, you also probably want to use a much better tripcode algorithm anyway, which also doesn't waste CPU cycles doing HTML conversions. Theoretically, when building a table of all printable characters, the code will eventually produce the same escape sequences, and by not encoding them explicitly, you also allow for alternative renderings for other boards that don't encode those characters.
Retrospectively speaking, posting it on a pastebin would probably have been a good idea too. For instance: + case ',': strcat(pw, "," ); p += 5; break;
That's practically guaranteed to break in the stupidest way possible. Apparently shitchan encodes some &-escapes... but not all of them. Always nice to have a programming board that hopelessly mangles code.
>>83
The point of tripcode searchers isn't primarily to crack your shitty tripcode. It's to find interesting novelty tripcodes.
Name:
Anonymous2009-02-26 12:30
my CS teacher said multiple returns in a switch statement is bad programming practice.
Name:
Anonymous2009-02-26 12:51
>>86
Anonymous on /prog/ said your CS teacher is a bad programmer.
Name:
Anonymous2009-02-26 13:10
>>86
My CS teacher didn't know you can do a While True loop (problem: infer the language from the code snippet).
Name:
Anonymous2009-02-26 13:35
>>88
Looks like Haskell to me. I don't know how to do a data WhileLoop = While Bool loop though.
Name:
Anonymous2009-02-26 13:45
>>86
Anyone who talks about programming practice like that is usually a total retard when it comes to actually programming in the real world. They're the kind of people who have huge 20000 line classes that take years to run, whereas a elite c coder will be returning all over the place and his code will run in seconds
Name:
Anonymous2009-02-26 14:12
>>90
Anyone who overgeneralizes will miss the point that the code in this particular case is actually shitty, and it'd be much more efficient to use a lookup table instead. >>86's CS teacher has a point.