Update a new Solr created field when a particular crawled field contains a certain string

[mitchelljj]

I have a field called "DC-ED.audience" that contains multiple strings that are separated by commas (see below example): "DC-ED.audience":["Institutions of Higher Education", "Administrators; Counselors"]

If I create new fields within Solr before doing the initial crawl like a "Institutions of Higher Education" field then when doing the crawl I would like to key on the "DC-ED.audience" field and when a string like "Institutions of Higher Education" is found update the new "Institutions of Higher Education" field with the value of "TRUE".

[essiembre]

There probably a few ways you could do this, but I suggest you look at the ReplaceTagger part of the Importer module. You could use it in a way similar to this:

<tagger class="com.norconex.importer.handler.tagger.impl.ReplaceTagger">
  <replace fromField="YourFieldHavingTheValue" toField="Institutions_of_Higher_Education"
          regex="true">
      <fromValue>.*DC-ED\.audience.*Institutions of Higher Education.*</fromValue>
      <toValue>true</toValue>
  </replace>
  <replace fromField="Institutions_of_Higher_Education" regex="true">
      <fromValue>^(?!true)*$</fromValue>
      <toValue></toValue>
  </replace>
</tagger>

If your DC-ED.audience field is not a metadata field already extracted but is part of the body, you can do something similar with TextPatternTagger.

[mitchelljj]

Thanks for the information! So if you don't add the second section within ReplaceTagger which I believe replaces any case of not true with no value what will the field of "Institutions_of_Higher_Education" contain or will this field not display within those records?

[essiembre]

I have not tested your particular case, but I believe it will copy the content over as is if it could not perform a replace. So you are correct, the second one is to make sure it is blank when not "true". In fact, it maybe best to store "false" instead of blank in your case.