Closed GoogleCodeExporter closed 8 years ago
The Solr server will take anything that my _documentToXmlFragment method
encodes with htmlspecialchars
and return it to its original state. What is probably happening is that your
source content has HTML entities in
it before it even gets to my conversion, but you want to be able to search on
the character the entities
represent. For that, you should normalize your source content before giving it
to Solr (or perhaps there is a
filter that can do this for you on the Solr side?) - a possible method would be
to use html_entity_decode
http://us3.php.net/manual/en/function.html-entity-decode.php on any fields you
suspect to have pre-
existing entities. I however can't do this in the client because it would be a
transform on the data that
someone else may not want (there may actually be someone out there that would
want to be able to search on
the full token "©" or similar.
On my Solr 1.4 instance, I was able to verify that I could find a document
indexed with a field "text" with value
"& < >" by the searches "text:&", "text:<", and "text:>" - just as it seems you
want to do. Those characters in
the document's field would have been converted to entities and then decoded by
Solr when indexed. For
reference, the text field had the following type definition in my schema.xml:
<!-- Less flexible matching, but less false matches. Probably not ideal for product names,
but may be good for SKUs. Can insert dashes in the wrong place and still match. -->
<fieldType name="textTight" class="solr.TextField" positionIncrementGap="100" >
<analyzer>
<tokenizer class="solr.WhitespaceTokenizerFactory"/>
<filter class="solr.SynonymFilterFactory" synonyms="synonyms.txt" ignoreCase="true" expand="false"/>
<filter class="solr.StopFilterFactory" ignoreCase="true" words="stopwords.txt"/>
<filter class="solr.LowerCaseFilterFactory"/>
<filter class="solr.RemoveDuplicatesTokenFilterFactory"/>
</analyzer>
</fieldType>
So, if what I've explained represents and understanding of your issue, I think
your problem is best solved by
either normalizing your source content to not contain HTML entities before you
pass it to Solr or by using an
appropriate field type filter defined within your schema.xml on the Solr server.
Original comment by donovan....@gmail.com
on 10 Dec 2009 at 4:53
Taking a look at our code - thanks for the pointers
Original comment by pwola...@gmail.com
on 10 Dec 2009 at 1:42
Original comment by donovan....@gmail.com
on 8 Feb 2010 at 6:26
Original issue reported on code.google.com by
pwola...@gmail.com
on 10 Dec 2009 at 2:20Attachments: