oodt-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Rishi Verma (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (OODT-652) New TikaCmdLineMetExtractor
Date Wed, 28 Aug 2013 18:58:56 GMT

    [ https://issues.apache.org/jira/browse/OODT-652?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13752731#comment-13752731

Rishi Verma commented on OODT-652:

Hey Chris,

The configuration file is expected to be a plain old Java properties file. So, just key=val
type of entries. An example is attached. The extractor is basically just using the org.apache.oodt.cas.metadata.extractors.MetReaderConfigReader
class to load a properties file. 

I was thinking we want a configuration file for two reasons:
1. When using this extractor in conjunction with OODT Crawler, a 'ProductType' must be specified
somehow in order for the product to be ingested. Placing the product type in a config file
is one way. Do you have any other possible methods to do this via OODT Crawler? I'd be very
curious to hear!
2. Sometimes we want to place configurable metadata that is ingested with every product. The
config file supports this need. What do you think?

> New TikaCmdLineMetExtractor
> ---------------------------
>                 Key: OODT-652
>                 URL: https://issues.apache.org/jira/browse/OODT-652
>             Project: OODT
>          Issue Type: New Feature
>          Components: metadata container
>    Affects Versions: 0.6
>            Reporter: Rishi Verma
>            Assignee: Rishi Verma
>             Fix For: 0.7
>         Attachments: OODT-652.rverma.08-27-2013.patch.txt
> Often times, we want to ingest a product and have some basic metadata automatically extracted
from it without much effort. The Apache Tika project has great features supporting the detection
of and extraction of metadata associated with a product to this effect. The purpose of this
issue is to integrate these metadata extraction capabilities of Tika, so that OODT can easily
leverage and make use of them. 
> At a minimum, this issue seeks to:
> * Incorporate and use Tika's 'parse' method to extract metadata automatically
> * Include the text content (if any) of a document inside a new metadata element dubbed
'content'. This will be useful for lucene and solr based free-text searches

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

View raw message