|
Is this to encode or decode, and what function would I use to reverse it?
Also, to clarify, microsoft activation number which is 51 digits long, when decoded into a byte array, is said to be simply a byte representation in litle endian, or a multi presision representation.
You see, if I simply encrypt a string of digits I end up with unreadable data, I could use built in functions to encode it into a hex string or Base32 etc, but I want to represent it as digits.
This is very hard to explain for me, sorry.
Steve
|
|
|
|
|
We still don't get what data you are after. Your last example had version and ID in the array. But we don't know if serial number is supposed to go in the array with those pieces of data or is serial number is supposed to be derived from the data in the array. We also don't know if any of the other potential pieces of data in the array are larger than one byte.
|
|
|
|
|
stephen.darling wrote: This is very hard to explain for me,
Suggestion: forget all technical terms, explain what the *probably non-technical) user sees and is supposed to do (and why). i.e. provide a functional specification. That would be a good start.
|
|
|
|
|
OK, let me try one more time, seen as your all being patient with me, which is greatly appreciated.
I will have a license generator that will create a fixed length serial number like:
0000 0000 0000 0000
This will be built from a range of bytes, though the exact number of bytes is not set in stone yet.
I thought I would use some routine to set some bytes like:
byte 1 = 0x00 (ID of program)
byte 2 = 0x00 (version)
etc etc
Then, I want to encrypt the byte array, then present it as a string of digits (a number that is an encoding of the byte array)
0000 0000 0000 0000
Then when the user types this into the application, I decode it back into bytes so I can decrypt it and get at each byte.
If I simply created a serial number as a string of numbers:
string serial = ("1234 1234 1234 1234")
and encrypted it, it is no longer numbers.
To add, all of this comes from the following paper, with actual code in c++
http://www.licenturion.com/xp/fully-licensed-wpa.txt[^]
but even if I managed to convert the code into c#, it only has decoder functions and not encoding functions.
Thank you again
Steve
|
|
|
|
|
stephen.darling wrote: encrypted, it is no longer numbers.
That is where you go wrong IMO. Encryption manipulates bits, it does not enforce a representation or format.
Here is my suggested approach:
A. you come up with a byte array holding the information you want;
B. you encrypt that into a new byte array;
C. so now the only remaining problem is to represent the encrypted byte array as a series of numbers (for human comfort I guess), here are a few possibilities:
C1. use base-8 (i.e. octal digits) for each byte, yielding values in the range [000,377] = 3 digits
C2. combine two bytes into a ushort, then use decimal, yielding values [00000,65535] = 5 digits
C3. combine 4 bytes into a uint [[FIXED]], then use decimal, yielding [0, some 4 billion] = 10 digits (no gain!)
I recommend C2.
combining bytes (unsigned by default) into longer unsigned numbers can be done with simple formulas, such as:
byte hi=...;
byte lo=...;
ushort hilo = (hi<<8)|(lo);
BTW: one can discuss which byte is hi, which is lo; it does not really matter as long as you do it consistently one way or the other.
|
|
|
|
|
Thank you for the responce, would your option (c2) do the following, as I found the part of the paper that concerns me, so here it is qouted from the paper itself...
"a 41-digit decimal number. A
decimal number of this length roughly corresponds to a 136-bit binary
number. In fact, the 41-digit number is just the decimal encoding of
such a 136-bit multi-precision integer, which is stored in little
endian byte order as a byte array. Hence, the above Installation ID
can also be represented as a sequence of 17 bytes as in
0xXX 0xXX 0xXX 0xXX 0xXX 0xXX 0xXX 0xXX
0x94 0xAA 0x46 0xD6 0x0F 0xBD 0x2C 0xC8
0x00"
Now, when I looked at the code, there was alot of << operations etc, so that maybe closer.
Also, how do I combine it all together using your option?
If a ushor gives me 5 digits, how do I combine the bytes to form one 20 or 25 digit number?
Thank you again
Steve
|
|
|
|
|
You lost me again. Do you have to implement what is in the text, or are you just looking for a working approach?
Anyway, they may actually have a 136-bit number in there, although I suspect they invented that to explain things more easily (i.e. I think they give a mathematical expose, not a code explanation).
What I proposed is, for 2N bytes, you have N 5-digit numbers; you never hold an overall number. You do have a string containing those N 5-digit numbers and some spaces, if that is the format you choose. It takes some string operations to aggregate that (and later split it again).
|
|
|
|
|
I was just looking for a working approach, although, there is code in c++ that decodes a string of 41 digits like the following "34234234459890798734987298374 etc" into that byte array. When I use my graphic calculator, the long string of digits does end up being the byte array, it just seems so hard to implement.
It is stange, if I have a byte array, I can very quickly output it as a string, represented as a hex or base32 etc. But representing a byte array as numbers seems so difficult.
Thank you for your help anyway, I do appreceate it alot.
I think for now Ill have to generate the serial number by hand and leave it unencrypted.
Thank you again,
Steve
|
|
|
|
|
stephen.darling wrote: But representing a byte array as numbers seems so difficult.
Do you want to assume the byte array is a binary representation of a large number, then convert that number to it's base10 form in a string? For example:
byte[] bigNumber = {1, 22};
Would be converted to:
string strBigNumber = "278";
You can use the BigInteger data type in .Net Framework 4 to help with that. There are also articles on how to do this if you are using an earlier version of the .Net Framework.
|
|
|
|
|
Aw bignumber? Ill have a look into that. thank you
Steve
|
|
|
|
|
Hi,
I don't know if I may help, but if I understand it well
your problem goes like this:
1. you have a fixed sequence of bytes, say you have
an array of 16 bytes. This may represents a number (more
precisily an unsigned or signed integer) in extended precision
in binary format (i.e. in base 2).
2. you want to handle this number and to convert it in
decimal base (base 10). Hence you get a nice base 10
representation outputted as a formatted string.
3. you want to be able to go back from the decimal
representation to the binary (base 2) one.
Trivial Example: do suppose your array is just 4 bytes long
(i.e. the standard size of an int)
namely you have in hex the value 0x0abc
stored as a four byte array in binary (base2) as:
(leading zeroes omitted) ...0000101010111100
you then want to convert it to the decimal representation
(base 10) with result: the string "2748".
and then be able to convert it back to its binary format
in base 2 in order to access it byte wise.
Well if this is right I think then you shall use
multiple precision arithmetique techniques: i.e.
you define a size (say 128 bytes) for your extended
precision integer and then you want a couple of functions
encodeBase10, decodeBase10 in order to be able to represent
such a number as a decimal string and convert back from it.
If such is your intention, you may take a look at the
GNU library "gmp", i.e. a multi precision library.
-----------
Instead of the solution above, you may want to do a translation byte
by byte, then the other method proposed in this thread
is just to output a converted value for each byte
in decimal taking care to pad with leading zeros where
needed, i.e. you transform from this:
(base 2) 00011010 00001111 00000111 10000011 - four bytes
to this
(base 10) 026 015 007 131 - string with 12 digits
and then convert back by grouping the string digits
3 by 3.
or if using ushort's instead of bytes,
you play with groups of 5 decimal digits
(always padding leading zeroes).
In this second case, for example, you have that a 25 long
string base10 digit represents 6 unsigned short numbers
and hence 12 bytes.
Hope this points you to the desired solution,
Cheers
|
|
|
|
|
Don't you mean an uint at C3?
|
|
|
|
|
I sure do. Thanks.
|
|
|
|
|
Is it possible to write a code that adds today's date to a jpg photo?(the date is changed everyday)
|
|
|
|
|
To what? The filename? The image itself? Data inside the image (aka steganography)?
|
|
|
|
|
if you want to create a new image file containing the original image plus a legend or a watermark, this pseudo-code should do:
Bitmap bm=Bitmap.FromFile(filepath);
Graphics g=Graphics.FromImage(bm);
g.DrawString(...);
bm.Save(someOtherFilepath, ImageFormat.Jpeg);
bm.Dispose();
FYI: this simple scheme will fail when someOtherFilepath==filepath
|
|
|
|
|
Visible or not visible? If you want it visible then what you should be searching for is jpg watermark c# (which gives you an article here as the first google result). If you want it invisible then what you want is called steganography for which there is a whole series of articles here as well.
|
|
|
|
|
The quantity resulting from the summation of today's date and a JPG photo will result in two terms that connot be simplified. You can, however, store the result symbolically rather than numerically.
|
|
|
|
|
huh? what do you mean by that? can you send the codez plz. It's kind of urgent ya know.
|
|
|
|
|
I have a list of phone numbers, some of them are valid and some of them aren't. I have to write a simple program, which will check which phone number is valid and which isn't.
I thought to use SerialPort for this, but how can I disconnect directly after the first beep (before somebody answers the phone) or after I got message from the carrier that the phone number isn't valid?
Or may be there is some other way to solve it?
Thanks in advance
|
|
|
|
|
|
Hi fellas,
I wanna know what is the best pattern to my web application to use LINQ to SQL in project.
|
|
|
|
|
|
Is that a design pattern now? And I've been using it for ages! 
Real men don't use instructions. They are only the manufacturers opinion on how to put the thing together.
|
|
|
|
|
You are going at things completely backwards. First you figure out what problem you are trying to solve. Then you decide what patterns provide the best solution for that problem. Then you decide what tools to use to implement those patterns.
The way you are doing things is like asking "What is the best vehicle to buy if I want it to be blue?" It may work out fine, but you may end up trying to drive a blue boat through desert sand or a blue motorcycle across the ocean floor.
|
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
