2 rounds error?

[caligatio]

2 rounds error?

[NobleUplift]

Can you show your code?

[boinggg]

I'm just copy and pasting in the demo.
On Mar 18, 2014 6:59 AM, "NobleUplift" [email protected] wrote:

Can you show your code?

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/17#issuecomment-37935153
.

[NobleUplift]

If you compute a hash and copy it back into the input field, it will result in a different result than rounds of hashing. This is because the data is finalized to give you that hash value, so you aren't taking a hash and doing another round of iteration on it; you are taking what jsSHA believes is just a string and hashing it. Actually, if you did hash, finalize, rehash, finalize, etc., it puts a huge load on the JavaScript engine and won't make it past 1000 rounds.

[boinggg]

Oh is this normal? I'm pretty sure other places do it my way? Which is
slower but "universal"?
On Mar 18, 2014 7:31 AM, "NobleUplift" [email protected] wrote:

If you compute a hash and copy it back into the input field, it will
result in a different result than rounds of hashing. This is because the
data is finalized to give you that hash value, so you aren't taking a hash
and doing another round of iteration on it; you are taking what jsSHA
believes is just a string and hashing it. Actually, if you did hash,
finalize, rehash, finalize, etc., it puts a huge load on the JavaScript
engine and won't make it past 1000 rounds.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/17#issuecomment-37938917
.

[Caligatio]

Noble got the explanation 99.9% correct. The issue is I don't believe you're dealing with your first round output correctly.

Round 1
Input -> ASCII string of test (74657374 in hex)
Output -> A hex encoded string of ee26b0dd4af7e749aa1a8ee3c10ae9923f618980772e473f8819a5d4940e0db27ac185f8a0e1d5f84f88bc887fd67b143732c304cc5fa9ad8e6f57f50028a8ff

Your Round 2
Input -> ASCII string of ee26b0dd4af7e749aa1a8ee3c10ae9923f618980772e473f8819a5d4940e0db27ac185f8a0e1d5f84f88bc887fd67b143732c304cc5fa9ad8e6f57f50028a8ff (6565323662306464346166376537343961613161386565336331306165393932336636313839383037373265343733663838313961356434393430653064623237616331383566386130653164356638346638386263383837666436376231343337333263333034636335666139616438653666353766353030323861386666 in hex)
Output -> 5797448d5c775085bc82b808b9d01bb4ef795dfd05ea377d9b537d5e3495179f0b0c03f372518209db3dfbdae47c8529412ce9154dd1c5b99932cdf3b079c125

My Round 2
Input -> hex string of ee26b0dd4af7e749aa1a8ee3c10ae9923f618980772e473f8819a5d4940e0db27ac185f8a0e1d5f84f88bc887fd67b143732c304cc5fa9ad8e6f57f50028a8ff
Output -> faadcaf60afd35dfcdb5e9ea0d0a0531f6338c62187cff37a1efe11f1d41a34879731bc9db49df75aecf5d582ad09b5c6ded2d86bd1f07c11bd755d1fccc81fe

So I don't think you're calculating Round 2 right. If you want to emulate what the library is doing, set the initial input to TEXT, output to hex and then set round 2 input as hex.

[boinggg]

Oh so the iterations stay in binary. My concern is the ability to test the
iterations manually without a binary input or output. I guess that's a
personal concern though. I just want to be sure I can duplicate the hashing
if this library disappears.
On Mar 18, 2014 11:25 AM, "Brian Turek" [email protected] wrote:

Noble got the explanation 99.9% correct. The issue is I don't believe
you're dealing with your first round output correctly.

Round 1
Input -> ASCII string of test (74657374 in hex)
Output -> A hex encoded string of
ee26b0dd4af7e749aa1a8ee3c10ae9923f618980772e473f8819a5d4940e0db27ac185f8a0e1d5f84f88bc887fd67b143732c304cc5fa9ad8e6f57f50028a8ff

Your Round 2
Input -> ASCII string of
ee26b0dd4af7e749aa1a8ee3c10ae9923f618980772e473f8819a5d4940e0db27ac185f8a0e1d5f84f88bc887fd67b143732c304cc5fa9ad8e6f57f50028a8ff
(6565323662306464346166376537343961613161386565336331306165393932336636313839383037373265343733663838313961356434393430653064623237616331383566386130653164356638346638386263383837666436376231343337333263333034636335666139616438653666353766353030323861386666
in hex)
Output ->
5797448d5c775085bc82b808b9d01bb4ef795dfd05ea377d9b537d5e3495179f0b0c03f372518209db3dfbdae47c8529412ce9154dd1c5b99932cdf3b079c125

My Round 2
Input -> hex string of
ee26b0dd4af7e749aa1a8ee3c10ae9923f618980772e473f8819a5d4940e0db27ac185f8a0e1d5f84f88bc887fd67b143732c304cc5fa9ad8e6f57f50028a8ff

Output ->
faadcaf60afd35dfcdb5e9ea0d0a0531f6338c62187cff37a1efe11f1d41a34879731bc9db49df75aecf5d582ad09b5c6ded2d86bd1f07c11bd755d1fccc81fe

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/17#issuecomment-37969801
.

[Caligatio]

Closing this as it seems to be solved.