Hi All,
What is the easiest way to
-add new fields to an already indexed documents
-adding new values to one field of an already indexed
document.
I realize that the preferred way to do this is to
delete the document and readd it with modified values.
But In our case, the sheer number of documents that
need this kind of massaging sometimes get too high to
do that, as the parsing cost for tokenizing the
documents gets too high.
I am thinking in the direction which involves adding
an additional ".fdx" file ( like ".fdx1" or something
) and similarly another ".fnm" file and so on . While
optimizing I merge the two files of similar types
together ( like ".fnm" to ".fnm1"). For Searches
before optimizing I will have to somehow take care of
merging values from two files in memory. I don't know
yet how ????
Another way to do this while avoiding reparsing the
document is to somehow get a list of all the fields
and associated tokens for the document . Delete the
original document and readd the document back with new
values ( old plus some new ) thus avoiding reparsing.
Any help pointer will be appreciated.
Thanks
Manish
__________________________________________________
Do You Yahoo!?
HotJobs - Search Thousands of New Jobs
http://www.hotjobs.com
--
To unsubscribe, e-mail: <mailto:lucene-dev-unsubscribe@jakarta.apache.org>
For additional commands, e-mail: <mailto:lucene-dev-help@jakarta.apache.org>
What is the easiest way to
-add new fields to an already indexed documents
-adding new values to one field of an already indexed
document.
I realize that the preferred way to do this is to
delete the document and readd it with modified values.
But In our case, the sheer number of documents that
need this kind of massaging sometimes get too high to
do that, as the parsing cost for tokenizing the
documents gets too high.
I am thinking in the direction which involves adding
an additional ".fdx" file ( like ".fdx1" or something
) and similarly another ".fnm" file and so on . While
optimizing I merge the two files of similar types
together ( like ".fnm" to ".fnm1"). For Searches
before optimizing I will have to somehow take care of
merging values from two files in memory. I don't know
yet how ????
Another way to do this while avoiding reparsing the
document is to somehow get a list of all the fields
and associated tokens for the document . Delete the
original document and readd the document back with new
values ( old plus some new ) thus avoiding reparsing.
Any help pointer will be appreciated.
Thanks
Manish
__________________________________________________
Do You Yahoo!?
HotJobs - Search Thousands of New Jobs
http://www.hotjobs.com
--
To unsubscribe, e-mail: <mailto:lucene-dev-unsubscribe@jakarta.apache.org>
For additional commands, e-mail: <mailto:lucene-dev-help@jakarta.apache.org>