It is possible to have the knowledge base in a database instead of a xml file? (for example to use it as an output with the Identity Caputure and use it as a input with Identity Update) is any way to configure Oyster to do that? of if not, would it be too much complicated to modify the source code to do that?
Thanks a lot,
Juan Gordon
Juan Gordon
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Juan,
You would have to change the code to do that.
You might find it simpler to take the XML version of KB, parse it using
Python or Java, and load the flat-file version into the database.
As long as you don't make changes to references in the database, i.e.
read-only, you can keep the XML version to re-load for update operations.
Thanks,
-jrt-
It is possible to have the knowledge base in a database instead of a xml
file? (for example to use it as an output with the Identity Caputure and
use it as a input with Identity Update) is any way to configure Oyster to
do that? of if not, would it be too much complicated to modify the source
code to do that?
I was thinking about the solution that you said, but I'm a little afraid about it, because i dont know how big can be the knowledge base if is stored in a XML (i'm planning to use Oyster to process millions of entities) , i suppose that the XML file should have a size limit.
Maybe another solution it could be to omit the knowledge base and just write the output of Identity Caputure in a Database by modifying the class OysterDatabaseWriter, as far i know the OysterId of the entities is persistent so it works well for what im looking for.
What do you think?
Thanks again
Greetings,
Juan Gordon
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Juan,
Yes you can write and read from a database once you develop the schema and
modified the code.
We are currently using the XML for 26 million references to about 1 million
entities.
Thanks,
-jrt-
I was thinking about the solution that you said, but I'm a little afraid
about it, because i dont know how big can be the knowledge base if is
stored in a XML (i'm planning to use Oyster to process millions of
entities) , i suppose that the XML file should have a size limit.
Maybe another solution it could be to omit the knowledge base and just
write the output of Identity Caputure in a Database by modifying the class
OysterDatabaseWriter, as far i know the OysterId of the entities is
persistent so it works well for what im looking for.
Hi there,
It is possible to have the knowledge base in a database instead of a xml file? (for example to use it as an output with the Identity Caputure and use it as a input with Identity Update) is any way to configure Oyster to do that? of if not, would it be too much complicated to modify the source code to do that?
Thanks a lot,
Juan Gordon
Juan Gordon
Juan,
You would have to change the code to do that.
You might find it simpler to take the XML version of KB, parse it using
Python or Java, and load the flat-file version into the database.
As long as you don't make changes to references in the database, i.e.
read-only, you can keep the XML version to re-load for update operations.
Thanks,
-jrt-
On Wed, Apr 30, 2014 at 5:03 PM, Juan Gordon jegordon@users.sf.net wrote:
Hi John ,
Thank you for your answer.
I was thinking about the solution that you said, but I'm a little afraid about it, because i dont know how big can be the knowledge base if is stored in a XML (i'm planning to use Oyster to process millions of entities) , i suppose that the XML file should have a size limit.
Maybe another solution it could be to omit the knowledge base and just write the output of Identity Caputure in a Database by modifying the class OysterDatabaseWriter, as far i know the OysterId of the entities is persistent so it works well for what im looking for.
What do you think?
Thanks again
Greetings,
Juan Gordon
Juan,
Yes you can write and read from a database once you develop the schema and
modified the code.
We are currently using the XML for 26 million references to about 1 million
entities.
Thanks,
-jrt-
On Wed, Apr 30, 2014 at 6:26 PM, Juan Gordon jegordon@users.sf.net wrote: