gistfile1.java

import org.marc4j.marc.Record;
import org.marc4j.marc.DataField;
import org.marc4j.marc.Subfield;
import org.solrmarc.tools.Utils;

// define the base level indexer so that its methods can be called from the script.
// note that the SolrIndexer code will set this value before the script methods are called.
org.solrmarc.index.SolrIndexer indexer = null;

/** Transformational hashes **/
static HashMap leader_6_7 = null;
public static leader_6_7_map() {
  if (leader_6_7 == null) {
    leader_6_7 = new HashMap();
    leader_6_7.put("aa", "Book"); // May be chapter level. 
    leader_6_7.put("ab", "Serial"); // serial 'component' may be article
    leader_6_7.put("am", "Book");
    leader_6_7.put("as", "Serial");
    leader_6_7.put("ta", "Book");
    leader_6_7.put("tm", "Book"); 
  }
  return leader_6_7;
}
static HashMap leader_6 = null;
public static leader_6_map() {
  if ( leader_6 == null) {
   leader_6 = new HashMap();
   leader_6.put("c", "Musical Score");
   leader_6.put("d", "Musical Score");
   leader_6.put("e", "Map/Globe");
   leader_6.put("f", "Map/Globe");
   leader_6.put("i", "Non-musical Recording");
   leader_6.put("j", "Musical Recording");
   leader_6.put("k", "Image");
   leader_6.put("m", "Software/Data");
   leader_6.put("g", "Video/Film");
  }
  return leader_6;
}
static HashMap field_007_0 = null;
// Map keys are Chars. 
public static field_007_0_map() {
  if ( field_007_0 == null) {
    field_007_0 = new HashMap();
    field_007_0.put('a', "Map/Globe"); // map
    field_007_0.put('d', "Map/Globe");// globe
    field_007_0.put('k', "Image"); // non projected graphic
    field_007_0.put('q', "Musical Score"); 
    field_007_0.put('r', "Image"); // remote sensing image
    field_007_0.put('v', "Video/Film"); // video
    field_007_0.put('m', "Video/Film"); // motion picture
  }
  return field_007_0;
}


/**
 * Looks up marc values for ordinary indexer spec passed in as argument, removing trailing spaces. Also adds in 490a values, each as a seperate
 value, for each 490 with indicator 1 == 0 
 * @param record auto passed in by solrmarc
 * @param spec an ordinary solrmarc field spec, or empty string.
 * @return Set of Strings suitable for Series title facet, including your spec and appropriate 490$a values. 
 */
Set getFormatFacet(Record record)
{
    LinkedHashSet resultSet = new LinkedHashSet();
    
    String leader =   record.getLeader().toString();
    //Set field006 = indexer.getFieldList(record, "006");
    //String field008 = indexer.getFieldList(record, "008");
    String gmd = indexer.getFirstFieldVal(record, "245h");


    Set field007Set = indexer.getFieldList(record, "007");    
    Set fields007Byte0 = new LinkedHashSet();
    Set fields007Byte1 = new LinkedHashSet();
    for ( String field007 : field007Set) {
      if (field007.length() > 0) { fields007Byte0.add( field007.charAt(0)); }
      if (field007.length() >= 2) { fields007Byte1.add( field007.charAt(1)); }
    }
    
    
    
    // First try bytes 6 and 7 of leader against our two byte
    // leader hash. 
    String mapped_leader_67 = leader_6_7_map().get(  leader.substring(6, 8)  );
    
    if (mapped_leader_67 != null) { resultSet.add( mapped_leader_67 ); }
    // If we didn't get one there, we can try just byte 6. 
    if (resultSet.size() == 0) {    
      String mapped_leader_6 = leader_6_map().get( leader.substring(6,7) );
      if (mapped_leader_6 != null) { resultSet.add( mapped_leader_6); }
    }
    // If we still didn't get one, give field(s) 007 byte 0 
    if ( resultSet.size() == 0) {
      for ( aByte : fields007Byte0) {
        mapped = field_007_0_map().get( aByte );
        if (mapped != null) { resultSet.add( mapped ); }
      }           
    }

    
    /* Okay, now add on our carrier/physical type things. Manuscript, Microform, Online, etc. */
    // Manuscript
    char leader_06 = (leader!= null && leader.length() > 5) ? leader.charAt(6) : 0;
    char leader_08 = (leader != null && leader.length() > 7) ? leader.charAt(8) : 0;
        // leader 6 t=Manuscript Language Material, d=Manuscript Music,
        // f=Manuscript Cartograhpic 
    if  ( leader_06 == 't' || leader_06 == 'd' || leader_06 == 'f' ||
            // leader 06 = 'b' is obsolete, but if it exists it means archival countrl
          leader_06 == 'b' ||
            // leader 8 a=archival control
          leader_08 == 'a'
          ) {
      resultSet.add("Manuscript/Archive"); 
    }
    
    // Microform
    // field 007 byte 0 means microfilm. But many of our microform items
    // don't have an 007. Theoretically we could look at the very confusing
    // 006 and 008, but those don't seem to be filled out either, need to
    // talk to a cataloger about that. Meanwhile, we'll resort to GMD.
    if (  fields007Byte0.contains('h') ||
          leader_6 == 'h' || // leader06 h is obsolete, but might still be in data. 
          (gmd != null && gmd.startsWith("[microform]")))
    {
      resultSet.add("Microform"); 
    }
    
    // Legacy "Available Online" just checked for the presence of an 856
    // with an http, as far as I can tell. This isn't sufficient to indicate
    // available online, the 856 http might not be full text. We'll try 007
    // and GMD for now. 006 and 008 might have helped if they were present,
    // but I think they generally aren't. 
    // Electronic resource => "Online".
    //field 007[0]=='c' means 'electronic resource', for field007[0]=='c', 
    //field007[1]=='r' means 'remote access'. 
    if ( (fields007Byte0.contains('c') && fields007Byte1.contains('r') ) ||
          // if the GMD is electronic resource, we count is as online only
          // if NO 007[0] is 'c', cause otherwise we already know it's
          //electronic but not online, or the above clause would have caught
          //it. 
         ( (! fields007Byte0.contains('c')) &&  gmd != null &&   
           gmd.startsWith("[electronic resource]")
         )
       )
    {
      resultSet.add("Online"); 
    }
    
    // THESES.  If it has a 502, it's a thesis. 
    if ( record.getVariableFields("502").size() > 0) {
       resultSet.add("Dissertation/Thesis");
       // Let's say if it's a thesis, it's not a 'Book. 
       resultSet.remove("Book");
       // Thought about removing Theses from "Manuscripts" bucket too,
       // but some of em really are ancient manuscript rare books type
       // stuff, even though some of them aren't. oh well. 
    }
    
        
    // If no other format at all, we'll still call it "Other".  
    if (resultSet.size() == 0) {
      resultSet.add("Other"); 
    }        
    
    return resultSet;
}