I came up with idea of indexing only document-wide unique tokens in a
field. So for example having this document:
{
'category': 'wine',
'description': 'bottle of wine from Marlow brewery'
}
i'd like to remove all tokens in category field from description - making
it better suited for search.
Which means i need to find a place that has tookens for both fields and
remove some of them.
Is there there an easy way to achieve that during indexing on ES side?
You can add both fields to the _all field [1] and then apply an analyzer
with a unique token filter [2] to the _all field. Make sure to exclude
other fields from _all. You can achieve something similar to the _all field
with multi field merging [3].
On Thu, Aug 22, 2013 at 8:47 AM, maciej@ly.st wrote:
Hi
I came up with idea of indexing only document-wide unique tokens in a
field. So for example having this document:
{
'category': 'wine',
'description': 'bottle of wine from Marlow brewery'
}
i'd like to remove all tokens in category field from description - making
it better suited for search.
Which means i need to find a place that has tookens for both fields and
remove some of them.
Is there there an easy way to achieve that during indexing on ES side?
Thanks - this may be useful, but i need more flexibility. multifield can
add fields, but i want to subtract them. Could plugin do that?
On Thursday, August 22, 2013 5:09:16 PM UTC+1, Ivan Brusic wrote:
You can add both fields to the _all field [1] and then apply an analyzer
with a unique token filter [2] to the _all field. Make sure to exclude
other fields from _all. You can achieve something similar to the _all field
with multi field merging [3].
On Thu, Aug 22, 2013 at 8:47 AM, <mac...@ly.st <javascript:>> wrote:
On Thu, Aug 22, 2013 at 8:47 AM, <mac...@ly.st <javascript:>> wrote:
Hi
I came up with idea of indexing only document-wide unique tokens in a
field. So for example having this document:
{
'category': 'wine',
'description': 'bottle of wine from Marlow brewery'
}
i'd like to remove all tokens in category field from description - making
it better suited for search.
Which means i need to find a place that has tookens for both fields and
remove some of them.
Is there there an easy way to achieve that during indexing on ES side?
--
You received this message because you are subscribed to the Google Groups
"elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an
email to elasticsearc...@googlegroups.com <javascript:>.
For more options, visit https://groups.google.com/groups/opt_out.
On Friday, August 23, 2013 12:40:04 PM UTC+1, mac...@ly.st wrote:
Thanks - this may be useful, but i need more flexibility. multifield can
add fields, but i want to subtract them. Could plugin do that?
On Thursday, August 22, 2013 5:09:16 PM UTC+1, Ivan Brusic wrote:
You can add both fields to the _all field [1] and then apply an analyzer
with a unique token filter [2] to the _all field. Make sure to exclude
other fields from _all. You can achieve something similar to the _all field
with multi field merging [3].
On Thu, Aug 22, 2013 at 8:47 AM, mac...@ly.st wrote:
On Thu, Aug 22, 2013 at 8:47 AM, mac...@ly.st wrote:
Hi
I came up with idea of indexing only document-wide unique tokens in a
field. So for example having this document:
{
'category': 'wine',
'description': 'bottle of wine from Marlow brewery'
}
i'd like to remove all tokens in category field from description -
making it better suited for search.
Which means i need to find a place that has tookens for both fields and
remove some of them.
Is there there an easy way to achieve that during indexing on ES side?
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.