-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I agree. If there are no objections, I'll revert and change initial_data.sql to 
use the hard-coded hashes this code generated.


Marlon


On 8/12/11 10:32 AM, Franklin, Matthew B. wrote:
>> -----Original Message-----
>> From: Ciancetta, Jesse E. [mailto:[email protected]]
>> Sent: Monday, August 08, 2011 9:11 AM
>> To: [email protected]
>> Subject: RE: [discuss] hashing, salting, and initial_data.sql
>>
>>> -----Original Message-----
>>> From: Marlon Pierce [mailto:[email protected]]
>>> Sent: Monday, August 08, 2011 8:59 AM
>>> To: [email protected]
>>> Subject: Re: [discuss] hashing, salting, and initial_data.sql
>>>
> Yes, but I was thinking about implementing a (hopefully) more elegant
> solution.
>>>
>>> For what use case?  The only thing I can think of where this might be useful
>>> would be for moving users over from some other container to Rave -- but I
>>> would think in that case you'd inevitably end up needing to write some kind 
>>> of
>>> custom migration utility anyway and I'd see dealing with the passwords as 
>>> part
>>> of that.
> 
> 
>> + 1
> 
>> I just noticed that hashing and salting of passwords was added to the 
>> SqlFileParser class.  This class (SqlFileParser) is very generic and can be 
>> used in any situation where SQL statements and child scripts are to be 
>> parsed from a file.  The hashing and salting definitely does not belong in 
>> this class as it is unique to a particular table in a particular context.  
> 
>>>
>>> Is there some other use case you have in mind?
>>>
> 
> Marlon
> 
> 
> On 8/8/11 8:39 AM, Ciancetta, Jesse E. wrote:
>>>>>> -----Original Message-----
>>>>>> From: Marlon Pierce [mailto:[email protected]]
>>>>>> Sent: Thursday, August 04, 2011 4:53 PM
>>>>>> To: [email protected]
>>>>>> Subject: [discuss] hashing, salting, and initial_data.sql
>>>>>>
>>>>> I'm looking at hashing and salting passwords stored in Rave's database. 
>>>>> This
>>>>> works fine for new user accounts, but the demo accounts (canonical,
>>>>> john.doe, etc) are a problem because they are inserted directly into the 
>>>>> DB
> by
>>>>> DataSourcePopulator.java by reading initial_data.sql. It would be possible
>>> to
>>>>> grok the "@user_id_" lines from initial_data.sql and hash the passwords
> there
>>>>> in SqlFileParser.java before inserting in the DB, but this would be an 
>>>>> ugly
>>> and
>>>>> fragile hack.
>>>>>
>>>>>
>>>>> Other suggestions?  Should we populate the database of demo users
> through
>>>>> JPA instead of inserting directly via SQL commands?
>>>>>
>>>>>> Is there some reason you can't salt and hash the passwords for the demo
> accounts manually and then insert the pre-salted/hashed values directly into
> the initial_data.sql file (with a comment block explaining what's being done
> and what the actual passwords are)?
>>>>>
>>>>>> Admittedly not the most elegant solution, but seems good enough for
> what we need to do.
>>>>>
>>>>>
>>>>> Marlon
> 
-----BEGIN PGP SIGNATURE-----
Version: GnuPG/MacGPG2 v2.0.16 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJORTq/AAoJEEfVXEODPFIDn5gH/iHsuE9xshj1KStBJCOg0+oz
KYr74mlN8BL7dWYAJZGf+OIN6SqDXeoH1O8IJYFjI38dZOD0+/RAsxmQa0KCzdb2
lPDCIxZZO5FBDoXYRxsBYr8TWWiOujcp42pAWazomNVp9KMyoFKqXa9X6BVB/krU
POKTMPqtR+pn1TESJDOgBxdFiLdSYPn3cKkjYgAQDSjihESAz10ryDGmnEoCRk8B
lZEwlZ4zzn1coQOr5fxNR1x2WQ54TrsxkYt9uq02ZsTXGeDIYMfqGKfgVEJ/0Ryw
+3GNktB26uMVD9eVP7iITO1U2wOKK+xB+iLtejYA771lFLPYxYcllSck4XXeBeE=
=aZSG
-----END PGP SIGNATURE-----

Reply via email to