[jira] [Updated] (SOLR-10218) The Schema API commands "add-field-type" and "replace-field-type" improperly specify SimilarityFactory params

["Steve Rowe (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/SOLR-10218?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Steve Rowe updated SOLR-10218:
------------------------------
    Summary: The Schema API commands "add-field-type" and "replace-field-type" improperly specify SimilarityFactory params  (was: The Schema API "replace-field-type" does not generate the SolrParameters for a SimilarityFactory correctly)

> The Schema API commands "add-field-type" and "replace-field-type" improperly specify SimilarityFactory params
> -------------------------------------------------------------------------------------------------------------
>
>                 Key: SOLR-10218
>                 URL: https://issues.apache.org/jira/browse/SOLR-10218
>             Project: Solr
>          Issue Type: Bug
>      Security Level: Public(Default Security Level. Issues are Public) 
>          Components: Schema and Analysis
>    Affects Versions: 6.4.1
>            Reporter: Benjamin Deininger
>            Assignee: Steve Rowe
>            Priority: Minor
>         Attachments: SOLR-10218.patch
>
>
> When sending a JSON POST to the Schema API to replace a field type, the following JSON does not pass the SolrParameters properly to the BM25SimilarityFactory.  
> {code:javascript}
> {"replace-field-type":{"name":"tint","class":"solr.TrieIntField","positionIncrementGap":"0","precisionStep":"8","similarity":{"class":"solr.BM25SimilarityFactory","k1":1.25,"b":0.75}}}
> {code}
> The `appendAttrs` function in the FieldTypeXmlAdapter parses k1 and b into attributes instead of children.  https://github.com/apache/lucene-solr/blob/master/solr/core/src/java/org/apache/solr/rest/schema/FieldTypeXmlAdapter.java#L155
> {code:xml}
> <similarity b="0.75" class="org.apache.lucene.search.similarities.BM25Similarity" k1="1.25"/>
> {code}
> Based on the XML examples for similarity, this should actually be the following :
> {code:xml}
> <similarity class="org.apache.lucene.search.similarities.BM25Similarity">
>      <float name=“k1”>0.1</float>
>      <float name=“b”>0.1</float>
> </similarity>
> {code}
> The similarities block in JSON should be handled differently so that the XML is generated appropriately.
> {code:java}
>     protected static Element appendSimilarityAttrs(Document doc, Element elm, Map<String,?> json) {
>         String clazz = (String) json.get("class");
>         elm.setAttribute("class", clazz);
>         json.remove("class");
>         for (Map.Entry<String,?> entry : json.entrySet()) {
>             Object val = entry.getValue();
>             if (val != null && !(val instanceof Map)) {
>                 Element element = doc.createElement(val.getClass().getSimpleName().toLowerCase());
>                 element.setAttribute("name", entry.getKey());
>                 element.setTextContent(entry.getValue().toString());
>                 elm.appendChild(element);
>             }
>         }
>         return elm;
>     }
> {code}



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org