Chris -

Good idea regarding not putting the constant in Metadata.  I just committed
a change as per your suggestion.

Chris & Bertrand, I hear you regarding limiting code changes to the issue at
hand. I'll use separate commits for changes nor related to the issue.

- Keith


Chris Mattmann wrote:
> 
> Hi Keith,
> 
>> I just committed my first code change.  I figured that since the issue
>> had
>> already been discussed, there was no need for further review.  On this
>> and
>> other issues I'm making my best guesses about how to do things the team's
>> way.  If I screw up, feel free to correct me. ;)
> 
> Not correcting, just a minor objection (not your fault, it's not like it's
> documented anywhere). I think that it makes a bit more sense to put a key
> for resourceName into an interface class with String constants, than
> putting
> the key name in the Metadata class. To me, any key name that you need to
> reference outside of the Metadata class, should not be part of Metadata
> itself. The Metadata class implements interfaces so that it can inherit
> such
> externally used keys. In the case of resourceName, I'm not sure exactly
> what
> that has to do with a generic Metadata container. Could we think of a
> better
> place to put it, e.g., if it has to do with the parsers, wouldn't it make
> more sense perhaps to create a TikaParseKeys interface within
> org.apache.tika.metadata, and then have the Metadata class implement that
> interface? 
> 
> 
>> 
>> By the way, how do you feel about changing little things in the files
>> along
>> the way?  For example, my IDE, Intellij Idea, enables me to switch to
>> Java 5
>> for loop syntax by the click of a single button.  Would it be ok to do
>> that
>> kind of thing, if I'm already committing that file anyway?
> 
> I favor, small incremental patches/commit that fix the problem as defined,
> and nothing else really. So if issue TIKA-XXX reports a problem A, and you
> commit a fix that fixes A, but that additionally reformats some files (say
> you did it correctly and those files weren't using spaces for indentation,
> they were using tabs), I would object to such a commit because you're
> mixing
> 2 things, and the actual changes are harder to track. If your IntelliJ is
> telling you things about Tika to refactor loops to make the code more
> readable, I would prefer you create an issue for that, and then we can
> track
> your progress/commits on that specific issue, as opposed to attaching a
> rider to the bill ;)
> 
> Cheers,
>   Chris
> 
> 
>> 
>> Thanks,
>> Keith
> 
> ______________________________________________
> Chris Mattmann, Ph.D.
> [EMAIL PROTECTED]
> Cognizant Development Engineer
> Early Detection Research Network Project
> 
> _________________________________________________
> Jet Propulsion Laboratory            Pasadena, CA
> Office: 171-266B                     Mailstop:  171-246
> _______________________________________________________
> 
> Disclaimer:  The opinions presented within are my own and do not reflect
> those of either NASA, JPL, or the California Institute of Technology.
> 
> 
> 
> 

-- 
View this message in context: 
http://www.nabble.com/TIKA-72-Commit-tf4641591.html#a13260530
Sent from the Apache Tika - Development mailing list archive at Nabble.com.

Reply via email to