davidludwig / smallsha1 Goto Github PK

if hash is only 20 bytes in size, as your documentation and example code 
suggest, your code writes into unallocated memory:

for (int hashByte = 20; --hashByte >= 0;)
{
  hash[hashByte] = (result[hashByte >> 2] >> (((3 - hashByte) & 0x3) << 3)) & 0xff;
}

hash needs to be at least 21 bytes in size.

What steps will reproduce the problem?
1. Downloaded sha1.cpp and sha1.h
2. Created a Project, files in the same dir as main.cpp
3. Compiled and ran each time 

What is the expected output? What do you see instead?
Expected output: a9993e364706816aba3e25717850c26c9cd0d89d
Random hash changes each time the program is run.

What version of the product are you using? On what operating system?
Dev-C++ 4.9.9.2 on Windows 7 64bit

Please provide any additional information below.
The only conclusion I can think of is the code was made for a 32bit system 
instead of 64bit?

I'm attempting to re-write the excellent smallsha1 for Arduino. The big issue 
is that the Arduino environment has 16-bit ints, instead of the 32 bits 
expected here. My naive strategy was to replace ints with longs.

Weirdly, calc(..) now works for arrays of three characters, but not for other 
lengths:

plain text: "abc"
result: a9993e364706816aba3e25717850c26c9cd0d89d
expected: a9993e364706816aba3e25717850c26c9cd0d89d

plain text: ""
result: 92b404e556588ced6c1acd4ebf053f6809f73a93
expected: da39a3ee5e6b4b0d3255bfef95601890afd80709

plain text: "hello"
result: 0c2a6006740560e244bea4096941fe2efb556fc0
expected: aaf4c61ddcc5e8a2dabede0f3b482cd9aea9434d

Any thoughts? I'm attaching the hacked version of the source.

Bruno

I download the source code and compile it,but the result seemed to be wrong.

using this main function:

int main(int argc, char* argv[])
{
    unsigned char hash[20];
    char hexstring[41];
    sha1::calc("Teststring",10,hash); // 10 is the length of the string
    sha1::toHexString(hash, hexstring);
    printf("%s\n",hexstring);
}

and i got these two different result.i dont know why.

mayli@matrix:~/code/sha1test$ ./test 
5cfe99f354c9eac90b08c47d67026d2257612f6d
mayli@matrix:~/code/sha1test$ sha1sum 1000sha1.txt 
2f20685189c918d40c6bda8b25429b158328e5cd  1000sha1.txt

by the way i calculated the sha1.zip file with sha1sum, where i got the same 
result as sha1 on the downloads page, so please show me a sample to how to make 
these two sha1 the same.

under Linux 2.6.31-14 g++ 4.4.1

Attached.

The integration of the SHA1 code worked flawlessly, calculation is working 
nicely.

But files need to be loaded into memory completely and this can be a problem 
for huge files.

Would it be possible to have some functions that do incremental encoding?
The idea would be:
Call INIT function to do basic stuff
Call CALC function with pointer to memory block and the length of the block 
again and again until the file is hashed completely
Call RESULT function to get the hash

Maybe this is some easy task for you?