# Percolating documents containing PDF attachments

**URL:** <https://discuss.elastic.co/t/percolating-documents-containing-pdf-attachments/10509>\
**Category:** Elasticsearch\
**Created:** [January 25, 2013, 8:29pm UTC](https://discuss.elastic.co/t/percolating-documents-containing-pdf-attachments/10509 "2013-01-25T20:29:18Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![Adam\_Georgiou](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/adam_georgiou/32/1409_2.png) [@Adam\_Georgiou](https://discuss.elastic.co/u/Adam_Georgiou)\
**Post date:** [January 25, 2013, 8:29pm UTC](https://discuss.elastic.co/t/percolating-documents-containing-pdf-attachments/10509/1 "2013-01-25T20:29:18Z")

</div>

I have an index setup in which we store documents containing PDF  
attachments, via the TIKA plugin. The index works great, in terms of  
indexing and searching on documents. However, when trying to run documents  
of similar type through the percolator index, queries referencing text in  
the attachment don't match, even when they normally return documents if ran  
as searches.

For example, we store this query in the percolator index:

{"query": {"constant\_score": {"filter": {"and": [{"terms": {"author":  
["george"]}}, {"query": {"match\_phrase": {"ATTACHMENT": "company"}}}]}}}}

In summary, I want documents containing the word "george" in the author  
field and containing the word "company" within the text of the attachment.  
Using this query against the document index returns documents. Storing it  
in the percolator index, and running documents against it does not. In  
fact, none of the queries referencing the attachment match.

I'm thinking, perhaps, there's some sort of secondary mapping I need to set  
up for the \_percolator index, in order to tell it to decode the attachment  
and parse it, but can't find any documentation alluding to it. It's also  
just a wild guess, as the setup I keep describing is purely a sandbox my  
team is using to evaluate elasticsearch. In other words, we don't really  
know what we're doing... at least not yet!

Does anyone have any idea as to what might be happening?

-Adam

--

---

<div class="post-metadata">

**Author:** ![Adam\_Georgiou](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/adam_georgiou/32/1409_2.png) [@Adam\_Georgiou](https://discuss.elastic.co/u/Adam_Georgiou)\
**Post date:** [January 28, 2013, 2:48pm UTC](https://discuss.elastic.co/t/percolating-documents-containing-pdf-attachments/10509/2 "2013-01-28T14:48:04Z")

</div>

I'm guessing this might have gotten buried, having been posted late Friday  
afternoon.

Givin' it a one-time-only, beginning of the week bump and I won't buy you  
guys (for a while)...

-Adam

On Friday, January 25, 2013 3:29:18 PM UTC-5, Adam Georgiou wrote:

> I have an index setup in which we store documents containing PDF  
> attachments, via the TIKA plugin. The index works great, in terms of  
> indexing and searching on documents. However, when trying to run documents  
> of similar type through the percolator index, queries referencing text in  
> the attachment don't match, even when they normally return documents if ran  
> as searches.
> 
> For example, we store this query in the percolator index:
> 
> {"query": {"constant\_score": {"filter": {"and": [{"terms": {"author":  
> ["george"]}}, {"query": {"match\_phrase": {"ATTACHMENT": "company"}}}]}}}}
> 
> In summary, I want documents containing the word "george" in the author  
> field and containing the word "company" within the text of the attachment.  
> Using this query against the document index returns documents. Storing it  
> in the percolator index, and running documents against it does not. In  
> fact, none of the queries referencing the attachment match.
> 
> I'm thinking, perhaps, there's some sort of secondary mapping I need to  
> set up for the \_percolator index, in order to tell it to decode the  
> attachment and parse it, but can't find any documentation alluding to it.  
> It's also just a wild guess, as the setup I keep describing is purely a  
> sandbox my team is using to evaluate elasticsearch. In other words, we  
> don't really know what we're doing... at least not yet!
> 
> Does anyone have any idea as to what might be happening?
> 
> -Adam

---

<div class="post-metadata">

**Author:** ![Adam\_Georgiou](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/adam_georgiou/32/1409_2.png) [@Adam\_Georgiou](https://discuss.elastic.co/u/Adam_Georgiou)\
**Post date:** [January 28, 2013, 3:31pm UTC](https://discuss.elastic.co/t/percolating-documents-containing-pdf-attachments/10509/3 "2013-01-28T15:31:16Z")

</div>

EDIT:

and I won't bother\* you guys for a while...

sorry for the mistype

On Monday, January 28, 2013 9:48:04 AM UTC-5, Adam Georgiou wrote:

> I'm guessing this might have gotten buried, having been posted late Friday  
> afternoon.
> 
> Givin' it a one-time-only, beginning of the week bump and I won't buy you  
> guys (for a while)...
> 
> -Adam
> 
> On Friday, January 25, 2013 3:29:18 PM UTC-5, Adam Georgiou wrote:
> 
> > I have an index setup in which we store documents containing PDF  
> > attachments, via the TIKA plugin. The index works great, in terms of  
> > indexing and searching on documents. However, when trying to run documents  
> > of similar type through the percolator index, queries referencing text in  
> > the attachment don't match, even when they normally return documents if ran  
> > as searches.
> > 
> > For example, we store this query in the percolator index:
> > 
> > {"query": {"constant\_score": {"filter": {"and": [{"terms": {"author":  
> > ["george"]}}, {"query": {"match\_phrase": {"ATTACHMENT": "company"}}}]}}}}
> > 
> > In summary, I want documents containing the word "george" in the author  
> > field and containing the word "company" within the text of the attachment.  
> > Using this query against the document index returns documents. Storing it  
> > in the percolator index, and running documents against it does not. In  
> > fact, none of the queries referencing the attachment match.
> > 
> > I'm thinking, perhaps, there's some sort of secondary mapping I need to  
> > set up for the \_percolator index, in order to tell it to decode the  
> > attachment and parse it, but can't find any documentation alluding to it.  
> > It's also just a wild guess, as the setup I keep describing is purely a  
> > sandbox my team is using to evaluate elasticsearch. In other words, we  
> > don't really know what we're doing... at least not yet!
> > 
> > Does anyone have any idea as to what might be happening?
> > 
> > -Adam

---

<div class="post-metadata">

**Author:** ![karmi](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/karmi/32/44951_2.png) [@karmi](https://discuss.elastic.co/u/karmi)\
**Post date:** [January 29, 2013, 7:46am UTC](https://discuss.elastic.co/t/percolating-documents-containing-pdf-attachments/10509/4 "2013-01-29T07:46:00Z")

</div>

> In summary, I want documents containing the word "george" in the author  
> field and containing the word "company" within the text of the attachment.  
> Using this query against the document index returns documents. Storing it  
> in the percolator index, and running documents against it does not. In  
> fact, none of the queries referencing the attachment match.
> 
> I'm thinking, perhaps, there's some sort of secondary mapping I need to  
> set up for the \_percolator index, in order to tell it to decode the  
> attachment and parse it, but can't find any documentation alluding to it.

Never used percolator with attachments, but if you need to configure the  
settings/mappings for the `_percolator` index, try deleting it, and  
creating it with custom settings/mappings just as any other index. (I'm  
actually not sure the configuration "sticks", but it would be a good  
starting point).

Karel

--  
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 [elasticsearch+unsubscribe@googlegroups.com](mailto:elasticsearch+unsubscribe@googlegroups.com).  
For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out).

---

<div class="post-metadata">

**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\
**Post date:** [July 6, 2017, 2:54am UTC](https://discuss.elastic.co/t/percolating-documents-containing-pdf-attachments/10509/5 "2017-07-06T02:54:15Z")

</div>