[SOLR-11700] WordDelimiterGraphFilterFactory token positions

• Type: Bug
• Status: Open
• Priority: Major
• Resolution: Unresolved
• Affects Version/s: 7.1
• Component/s: Schema and Analysis

Environment

Mac OSX, JDK 8

• Assignee: Unassigned
• Reporter: Chad Siongco
• Created: 29/Nov/17 08:28
• Updated: 08/Jun/19 15:15

Description

Token position Generated after WordDelimiterGraphFilterFactory are incorrect.

This causes problems when doing phrase searches.

As stated in the following link,
https://lucene.apache.org/solr/guide/6_6/filter-descriptions.html

<analyzer type="query">
<tokenizer class="solr.WhitespaceTokenizerFactory"/>
<filter class="solr.WordDelimiterGraphFilterFactory" catenateAll="1"/>
</analyzer>

In: "XL-4000/ES"
Tokenizer to Filter: "XL-4000/ES"(1)
Out: "XL"(1), "4000"(2), "ES"(3), "XL4000ES"(3)

But in my Machine, notice that the concatenated word is at position 1, it should be position 3:
Out: XL4000ES"(1)", XL"(1), "4000"(2), "ES"(3), "