| |
| |
| |
|
Page: 1 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
<syscode> its hard to look this up but for pointers what is the difference between type* var1 and type** var2? I have been searching but hard to figure out the search term for **
<syscode> oh never mind it is just double pointer
<NKD> glad I could help.
<NKD> /bow
<NKD> lol
<syscode> lol
<Bunkai-> i am sorry if this question is tupid: but will this occur once or twice? for (int iLoop = 0; iLoop < 1; iLoop++) Nesuhlas_poskyt_udajov[iLoop] = 0;
<_m_> once
<Bunkai-> _m_ thank you..
<dizzi> Bunkai-: takze nesuhlas poskytovanych udajov hej? ;)
<threat> with sockets (TCP/IP), is there a way to send data across of an unknown size?
<dizzi> threat: what do you think?
<_m_> Is it possible to eat a bread of unknown size?
<NKD> ...
<NKD> that analogy amuses me
<NKD> You would not know when to open and close your mouth :P
<Jey> why do you need to?
<Jey> just keep putting hte bread in your mouth and biting until you run out of bread
<dizzi> Jey: but its not system solution... in real life you usualy need know how this end
<dizzi> you can be ok, and you can puke all the day
<Trazer> threat: Through a TCP stream you can send any data you want.. But other end would probably like some way of knowing what's incoming..
<Trazer> An IP packet has to be of a known size though..
<Jey> dizzi: so shove down a potato afterwards as an end of bread marker
<_m_> It's amazing how similar the discussions on Undernet/#C++ and Efnet/#C++ evolve.
<dizzi> yea, but you can do it only if you are lucky, and youve eat all bread already, in other case (worse), you will puke before you finish bread job :)
<_m_> Yes, just like your adsl router will puke when you start uploading.
<_m_> Oh man, the bread analogy really works.
<dizzi> :))
<Genia4> _m_: I don't think I've ever heard you use "man" before...
<Bunkai-> ha, this is miracle: I have generated console application in MS VS 2005 Beta 2, however, I can run on my computer, but if I want to run it in another computer, I get a message "can not execute the file specified" from a MSDos console window. The OS is WinXP on both machines..
<NKD> Sir I believe you have my stapler
<llah> Staplers are staples of an office environment. The irony is dizzying.
<dizzi> Bunkai-: are you sure that you have all required .dlls on 2nd machine?
<shaneC> http://instant-e-gold.com/visit.php?id=43&refid=2516772 get free e-gold for like 10 seconds of work
<mauke> what is e-gold?
<NKD> probably some scam
<sculptor> hey
<Bunkai-> dizzi, no I dont have the runtime installed in target machine..
<Bunkai-> I just figured it out..
<llah> Oh boy, e-Gold
<dizzi> :))
<Bunkai-> dizzi, when I used VS 6, I did not need anything like that
<dizzi> its all about your project settings
<Bunkai-> ok, I am going to find it and set it up
<lck> what does ascii[i] >>= 1; do?
<dizzi> shift value of ascii[1] one bit right
<mauke> same as ascii[i] /= 2;
<sculptor> C++ lacks rotating shift operators
<lck> so ascii[i] <<= 1; moves one bit left?
<Jey> nothing two shifts and an or can't handle
<binrapt> Indeed
<dizzi> lck: yes, or its multiply value by 2
<lck> ok
<binrapt> But rotations aren't supported natively on most CPUs anyway, are they not
<sculptor> hm, two shifts, 2 ands with mask, and an or
<binrapt> Huh? Why and?
<sculptor> to remove unecessary bits
<sculptor> and, rotations are natively supported by almost all CPUs
<Bunkai-> dizzi, the console runtime library, do you know please where is its settings in project properties?
<sculptor> even C64s 6510 and ZX Siclairs Z80A
<binrapt> What about PPC
<sculptor> sure
<sculptor> IA32 supports rotating shift operations
<binrapt> Yes
<dizzi> Bunkai-: i really dont know, im using mingw+eclipse, its only what i remember from old times
<sculptor> through Carry flag, or not through it
<dizzi> binrapt: rotating and shifting is so basic and easy to realize operations
<Bunkai-> dizzi, ok.. thanks..
<dizzi> that i think that its support almost all cpus
<sculptor> just realized - no ands operations are necessary - zero bits are padded
<mauke> n >> k | n << 32 - k
<sculptor> smart copmiler could convert that to native ***ebler instruction
<binrapt> Why - k?
<_m_> binrapt: what else would you suggest?
<binrapt> Hmm two shifts and an or, as Jey already mentioned?
<_m_> isn't that two shifts and an or?
<binrapt> Two shifts an or and a subtraction..?
<_m_> ok. How many bits would you like to shift then?
<dizzi> uh
<binrapt> Hmm well, depending on left/right, sizeof(type) * bits_per_byte - k and k
<dizzi> 1110 ror 2 -> 0011 | 1000
<snibril> (x << 1) | (x > (~0 >> 1))
<lck> for (int i=0; i<sLen; i++) { ascii[i] = (int)p***wd[i]; ascii[i] >>= 1; codedPw += (char)ascii[i];} how do i reverse that
<binrapt> Reverse the process :)
<upb> eee you cant :P
<sculptor> revese engineer it :)
<upb> its more like a hash
<lck> upd: i can't?
<binrapt> Ah, of course, didn't even read it :(
<binrapt> I thought it manipulated a string or something
<lck> it does
<lck> i didn't past the whole program
<binrapt> Well, codedPw is what it's about, isn't it?
<upb> heh its clear from this part
<snibril> since a biot is thrown away, there is no way to reverse it
<binrapt> It's a sum apparently
<snibril> binrapt: += works on strings
<sculptor> there is no single solution
<snibril> and he did not state types
<binrapt> Hmm oh well
<lck> cout << "enter pw: "; cin >> p***wd; sLen = p***wd.length(); int ascii[sLen];
<upb> but if you know codedPw and want to generate p***words
<upb> its very easy ;)
<binrapt> >>= 1, a bit is lost, hmm
<lck> i know codedPw and wanna decoded it or what ever you call it
<binrapt> Well, there are multiple possible solutions
<snibril> you cannot
<snibril> btw, coded is an exaggeration
<snibril> it'S slightly obscured at most
<binrapt> And that is...2^(sLen) possibilities or so? :)
<lck> i choose to think i can, since it's a ex. from my teacher...
<snibril> reinventing scurity wheels is even more dangerous than normal wheels
<upb> lol
<upb> take 30 sec to think what it actually does
<snibril> techers are expecially clueless
<binrapt> He apparently wants you to find out that you cannot recreate the original string :)
<lck> darn, gtg, another cl*** starts
<binrapt> Write a supar cool key generator! <<= 1; &= rand() % 2
<snibril> that code is convoluted to start with, the whole casting business is useless
<_m_> I suggest you replace the >>=1 with >>=8. After that, the p***word checking code can get optimized.
<hac_> what's the most efficient way to compute a hash code for three small integers?
<binrapt> _m_: Hah
<snibril> not to mention storing temp results in an array
<_m_> hac_: 0
<snibril> binrapt: rand%2 is unrandom
<binrapt> snibril: Nothing is random!
<hac_> _m_: i don't follow :)
<matja> windows failures are random
<snibril> hao: define efficient and "hash code"
<mauke> hac_: 0 is an efficient hash code for anything
<binrapt> matja: Nope
<matja> hash_code=x*biggest_value_of_x*biggest_value_of_y+y*biggest_value_of_y+z
<matja> :p
<mauke> int hash(int, int, int) { return 0; } // fast!!
<hac_> i see :P
<noam_> inline it
<snibril> that is an encoding, not a hash
<noam_> xor them
<binrapt> I claim that a copy of this universe would have the same things happening
<dizzi> its hash with pretty much colisions :)
<binrapt> dizzi: Kinda :P
<hac_> well, sorry, a unique hash code
<matja> binrapt: a copy isnt possible :)
<matja> so how would you ever prove it
<hac_> for instance, two integers.. x << 16 | y
<snibril> hashes are seldom unique
<snibril> a hash, or a digest?
<binrapt> matja: Simply, never!
<matja> :D
Return to c++
or
Go to some related
logs:
wow.no
|
|