Author: hossman
Date: Thu Mar 26 18:42:25 2009
New Revision: 758795
URL: http://svn.apache.org/viewvc?rev=758795&view=rev
Log:
SOLR-952: Cleanup duplicated code in deprecated HighlightingUtils
Modified:
lucene/solr/trunk/CHANGES.txt
lucene/solr/trunk/src/java/org/apache/solr/util/HighlightingUtils.java
Modified: lucene/solr/trunk/CHANGES.txt
URL:
http://svn.apache.org/viewvc/lucene/solr/trunk/CHANGES.txt?rev=758795&r1=758794&r2=758795&view=diff
==============================================================================
--- lucene/solr/trunk/CHANGES.txt (original)
+++ lucene/solr/trunk/CHANGES.txt Thu Mar 26 18:42:25 2009
@@ -369,6 +369,8 @@
22. SOLR-1068: Use fsync on replicated index and configuration files (yonik,
shalin)
+23. SOLR-952: Cleanup duplicated code in deprecated HighlightingUtils (hossman)
+
Build
----------------------
1. SOLR-776: Added in ability to sign artifacts via Ant for releases
(gsingers)
Modified: lucene/solr/trunk/src/java/org/apache/solr/util/HighlightingUtils.java
URL:
http://svn.apache.org/viewvc/lucene/solr/trunk/src/java/org/apache/solr/util/HighlightingUtils.java?rev=758795&r1=758794&r2=758795&view=diff
==============================================================================
--- lucene/solr/trunk/src/java/org/apache/solr/util/HighlightingUtils.java
(original)
+++ lucene/solr/trunk/src/java/org/apache/solr/util/HighlightingUtils.java Thu
Mar 26 18:42:25 2009
@@ -37,6 +37,8 @@
import org.apache.solr.search.DocList;
import org.apache.solr.search.SolrIndexSearcher;
import org.apache.solr.schema.SchemaField;
+import org.apache.solr.highlight.SolrHighlighter;
+import org.apache.solr.highlight.DefaultSolrHighlighter;
import org.apache.lucene.analysis.*;
import org.apache.lucene.document.Document;
@@ -44,13 +46,14 @@
import org.apache.lucene.search.highlight.*;
/**
- * Collection of Utility and Factory methods for Highlighting.
+ * DEPRECATED Collection of Utility and Factory methods for Highlighting.
*
- * @deprecated use highlit.SolrHighlighter
+ * @deprecated use DefaultSolrHighlighter
+ * @see DefaultSolrHighlighter
*/
public class HighlightingUtils implements HighlightParams {
-
- private static SolrParams DEFAULTS = null;
+
+ static SolrParams DEFAULTS = null;
static {
Map<String,String> map = new HashMap<String,String>();
map.put(SNIPPETS, "1");
@@ -61,372 +64,98 @@
DEFAULTS = new MapSolrParams(map);
}
-
+
+ private static SolrHighlighterX HIGHLIGHTER = new SolrHighlighterX();
+
/** Combine request parameters with highlighting defaults. */
- private static SolrParams getParams(SolrQueryRequest request) {
+ static SolrParams getParams(SolrQueryRequest request) {
return new DefaultSolrParams(request.getParams(), DEFAULTS);
}
/**
- * Check whether Highlighting is enabled for this request.
- * @param request The current SolrQueryRequest
- * @return <code>true</code> if highlighting enabled, <code>false</code> if
not.
+ * @deprecated use DefaultSolrHighlighter
+ * @see DefaultSolrHighlighter#isHighlightingEnabled
*/
public static boolean isHighlightingEnabled(SolrQueryRequest request) {
- return getParams(request).getBool(HIGHLIGHT, false);
+ return HIGHLIGHTER.isHighlightingEnabled(getParams(request));
}
/**
- * Return a Highlighter appropriate for this field.
- * @param query The current Query
- * @param fieldName The name of the field
- * @param request The current SolrQueryRequest
+ * @deprecated use DefaultSolrHighlighter
+ * @see DefaultSolrHighlighter
*/
public static Highlighter getHighlighter(Query query, String fieldName,
SolrQueryRequest request) {
- Highlighter highlighter = new Highlighter(
- getFormatter(fieldName, request),
- getQueryScorer(query, fieldName, request));
- highlighter.setTextFragmenter(getFragmenter(fieldName, request));
- return highlighter;
+ return HIGHLIGHTER.getHighlighterX(query, fieldName, request);
}
/**
- * Return a QueryScorer suitable for this Query and field.
- * @param query The current query
- * @param fieldName The name of the field
- * @param request The SolrQueryRequest
+ * @deprecated use DefaultSolrHighlighter
+ * @see DefaultSolrHighlighter
*/
public static QueryScorer getQueryScorer(Query query, String fieldName,
SolrQueryRequest request) {
- boolean reqFieldMatch = getParams(request).getFieldBool(fieldName,
FIELD_MATCH, false);
- if (reqFieldMatch) {
- return new QueryScorer(query, request.getSearcher().getReader(),
fieldName);
- }
- else {
- return new QueryScorer(query);
- }
+ return HIGHLIGHTER.getQueryScorerX(query, fieldName, request);
}
/**
- * Return a String array of the fields to be highlighted.
- * Falls back to the programatic defaults, or the default search field if
the list of fields
- * is not specified in either the handler configuration or the request.
- * @param query The current Query
- * @param request The current SolrQueryRequest
- * @param defaultFields Programmatic default highlight fields, used if
nothing is specified in the handler config or the request.
+ * @deprecated use DefaultSolrHighlighter
+ * @see DefaultSolrHighlighter#getHighlightFields
*/
public static String[] getHighlightFields(Query query, SolrQueryRequest
request, String[] defaultFields) {
- String fields[] = getParams(request).getParams(FIELDS);
-
- // if no fields specified in the request, or the handler, fall back to
programmatic default, or default search field.
- if(emptyArray(fields)) {
- // use default search field if highlight fieldlist not specified.
- if (emptyArray(defaultFields)) {
- fields = new
String[]{request.getSchema().getDefaultSearchFieldName()};
- }
- else {
- fields = defaultFields;
- }
- }
- else if (fields.length == 1) {
- // if there's a single request/handler value, it may be a space/comma
separated list
- fields = SolrPluginUtils.split(fields[0]);
- }
-
- return fields;
- }
-
- private static boolean emptyArray(String[] arr) {
- return (arr == null || arr.length == 0 ||
- (arr.length == 1 && (arr[0] == null || arr[0].trim().length() == 0)));
+ return HIGHLIGHTER.getHighlightFields(query, request, defaultFields);
}
/**
- * Return the max number of snippets for this field. If this has not
- * been configured for this field, fall back to the configured default
- * or the solr default.
- * @param fieldName The name of the field
- * @param request The current SolrQueryRequest
+ * @deprecated use DefaultSolrHighlighter
+ * @see DefaultSolrHighlighter
*/
public static int getMaxSnippets(String fieldName, SolrQueryRequest
request) {
- return Integer.parseInt(getParams(request).getFieldParam(fieldName,
SNIPPETS));
+ return HIGHLIGHTER.getMaxSnippetsX(fieldName, request);
}
/**
- * Return a formatter appropriate for this field. If a formatter
- * has not been configured for this field, fall back to the configured
- * default or the solr default (SimpleHTMLFormatter).
- *
- * @param fieldName The name of the field
- * @param request The current SolrQueryRequest
- * @return An appropriate Formatter.
+ * @deprecated use DefaultSolrHighlighter
+ * @see DefaultSolrHighlighter
*/
public static Formatter getFormatter(String fieldName, SolrQueryRequest
request) {
- SolrParams p = getParams(request);
-
- // SimpleHTMLFormatter is the only supported Formatter at the moment
- return new SimpleHTMLFormatter(p.getFieldParam(fieldName, SIMPLE_PRE),
p.getFieldParam(fieldName, SIMPLE_POST));
+ return HIGHLIGHTER.getFormatterX(fieldName, request);
}
/**
- * Return a fragmenter appropriate for this field. If a fragmenter
- * has not been configured for this field, fall back to the configured
- * default or the solr default (GapFragmenter).
- *
- * @param fieldName The name of the field
- * @param request The current SolrQueryRequest
- * @return An appropriate Fragmenter.
+ * @deprecated use DefaultSolrHighlighter
+ * @see DefaultSolrHighlighter
*/
public static Fragmenter getFragmenter(String fieldName, SolrQueryRequest
request) {
- int fragsize =
Integer.parseInt(getParams(request).getFieldParam(fieldName, FRAGSIZE));
- return (fragsize <= 0) ? new NullFragmenter() : new
GapFragmenter(fragsize);
+ return HIGHLIGHTER.getFragmenterX(fieldName, request);
}
/**
- * Generates a list of Highlighted query fragments for each item in a list
- * of documents, or returns null if highlighting is disabled.
- *
- * @param docs query results
- * @param query the query
- * @param req the current request
- * @param defaultFields default list of fields to summarize
- *
- * @return NamedList containing a NamedList for each document, which in
- * turns contains sets (field, summary) pairs.
+ * @deprecated use DefaultSolrHighlighter
+ * @see DefaultSolrHighlighter#doHighlighting
*/
@SuppressWarnings("unchecked")
public static NamedList doHighlighting(DocList docs, Query query,
SolrQueryRequest req, String[] defaultFields) throws IOException {
- if (!isHighlightingEnabled(req))
- return null;
-
- SolrIndexSearcher searcher = req.getSearcher();
- NamedList fragments = new SimpleOrderedMap();
- String[] fieldNames = getHighlightFields(query, req, defaultFields);
- Document[] readDocs = new Document[docs.size()];
- {
- // pre-fetch documents using the Searcher's doc cache
- Set<String> fset = new HashSet<String>();
- for(String f : fieldNames) { fset.add(f); }
- // fetch unique key if one exists.
- SchemaField keyField =
req.getSearcher().getSchema().getUniqueKeyField();
- if(null != keyField)
- fset.add(keyField.getName());
- searcher.readDocs(readDocs, docs, fset);
- }
-
- // Highlight each document
- DocIterator iterator = docs.iterator();
- for (int i = 0; i < docs.size(); i++) {
- int docId = iterator.nextDoc();
- Document doc = readDocs[i];
- NamedList docSummaries = new SimpleOrderedMap();
- for (String fieldName : fieldNames) {
- fieldName = fieldName.trim();
- String[] docTexts = doc.getValues(fieldName);
- if (docTexts == null) continue;
-
- // get highlighter, and number of fragments for this field
- Highlighter highlighter = getHighlighter(query, fieldName, req);
- int numFragments = getMaxSnippets(fieldName, req);
-
- String[] summaries;
- TextFragment[] frag;
- if (docTexts.length == 1) {
- // single-valued field
- TokenStream tstream;
- try {
- // attempt term vectors
- tstream = TokenSources.getTokenStream(searcher.getReader(),
docId, fieldName);
- }
- catch (IllegalArgumentException e) {
- // fall back to analyzer
- tstream = new
TokenOrderingFilter(searcher.getSchema().getAnalyzer().tokenStream(fieldName,
new StringReader(docTexts[0])), 10);
- }
- frag = highlighter.getBestTextFragments(tstream, docTexts[0],
false, numFragments);
- }
- else {
- // multi-valued field
- MultiValueTokenStream tstream;
- tstream = new MultiValueTokenStream(fieldName, docTexts,
searcher.getSchema().getAnalyzer(), true);
- frag = highlighter.getBestTextFragments(tstream,
tstream.asSingleValue(), false, numFragments);
- }
- // convert fragments back into text
- // TODO: we can include score and position information in output
as snippet attributes
- if (frag.length > 0) {
- ArrayList<String> fragTexts = new ArrayList<String>();
- for (int j = 0; j < frag.length; j++) {
- if ((frag[j] != null) && (frag[j].getScore() > 0)) {
- fragTexts.add(frag[j].toString());
- }
- }
- summaries = fragTexts.toArray(new String[0]);
- if (summaries.length > 0)
- docSummaries.add(fieldName, summaries);
- }
- }
- String printId = searcher.getSchema().printableUniqueKey(doc);
- fragments.add(printId == null ? null : printId, docSummaries);
- }
- return fragments;
+ return HIGHLIGHTER.doHighlighting(docs, query, req, defaultFields);
}
}
-/**
- * Helper class which creates a single TokenStream out of values from a
- * multi-valued field.
- */
-class MultiValueTokenStream extends TokenStream {
- private String fieldName;
- private String[] values;
- private Analyzer analyzer;
- private int curIndex; // next index into the values array
- private int curOffset; // offset into concatenated string
- private TokenStream currentStream; // tokenStream currently being
iterated
- private boolean orderTokenOffsets;
-
- /** Constructs a TokenStream for consecutively-analyzed field values
- *
- * @param fieldName name of the field
- * @param values array of field data
- * @param analyzer analyzer instance
- */
- public MultiValueTokenStream(String fieldName, String[] values,
- Analyzer analyzer, boolean orderTokenOffsets) {
- this.fieldName = fieldName;
- this.values = values;
- this.analyzer = analyzer;
- curIndex = -1;
- curOffset = 0;
- currentStream = null;
- this.orderTokenOffsets=orderTokenOffsets;
- }
-
- /** Returns the next token in the stream, or null at EOS. */
- @Override
- public Token next() throws IOException {
- int extra = 0;
- if(currentStream == null) {
- curIndex++;
- if(curIndex < values.length) {
- currentStream = analyzer.tokenStream(fieldName,
- new
StringReader(values[curIndex]));
- if (orderTokenOffsets) currentStream = new
TokenOrderingFilter(currentStream,10);
- // add extra space between multiple values
- if(curIndex > 0)
- extra = analyzer.getPositionIncrementGap(fieldName);
- } else {
- return null;
- }
- }
- Token nextToken = currentStream.next();
- if(nextToken == null) {
- curOffset += values[curIndex].length();
- currentStream = null;
- return next();
- }
- // create an modified token which is the offset into the concatenated
- // string of all values
- Token offsetToken = new Token(new String(nextToken.termBuffer(), 0,
nextToken.termLength()),
- nextToken.startOffset() + curOffset,
- nextToken.endOffset() + curOffset);
- offsetToken.setPositionIncrement(nextToken.getPositionIncrement() +
extra*10);
- return offsetToken;
- }
-
- /**
- * Returns all values as a single String into which the Tokens index with
- * their offsets.
- */
- public String asSingleValue() {
- StringBuilder sb = new StringBuilder();
- for(String str : values)
- sb.append(str);
- return sb.toString();
- }
-
-}
-
/**
- * A simple modification of SimpleFragmenter which additionally creates new
- * fragments when an unusually-large position increment is encountered
- * (this behaves much better in the presence of multi-valued fields).
+ * subclass containing package protected versions of some protected methods,
used for proxying calls to deprecated methods that have been moved and made
protected.
*/
-class GapFragmenter extends SimpleFragmenter {
- /**
- * When a gap in term positions is observed that is at least this big, treat
- * the gap as a fragment delimiter.
- */
- public static final int INCREMENT_THRESHOLD = 50;
- protected int fragOffsetAccum = 0;
-
- public GapFragmenter() {
- }
-
- public GapFragmenter(int fragsize) {
- super(fragsize);
+class SolrHighlighterX extends DefaultSolrHighlighter {
+ Highlighter getHighlighterX(Query query, String fieldName, SolrQueryRequest
request) {
+ return getHighlighter(query, fieldName, request);
}
-
- /* (non-Javadoc)
- * @see
org.apache.lucene.search.highlight.TextFragmenter#start(java.lang.String)
- */
- @Override
- public void start(String originalText) {
- fragOffsetAccum = 0;
+ QueryScorer getQueryScorerX(Query query, String fieldName, SolrQueryRequest
request) {
+ return getQueryScorer(query, fieldName, request);
}
-
- /* (non-Javadoc)
- * @see
org.apache.lucene.search.highlight.TextFragmenter#isNewFragment(org.apache.lucene.analysis.Token)
- */
- @Override
- public boolean isNewFragment(Token token) {
- boolean isNewFrag =
- token.endOffset() >= fragOffsetAccum + getFragmentSize() ||
- token.getPositionIncrement() > INCREMENT_THRESHOLD;
- if(isNewFrag) {
- fragOffsetAccum += token.endOffset() - fragOffsetAccum;
- }
- return isNewFrag;
+ int getMaxSnippetsX(String fieldName, SolrQueryRequest request) {
+ return getMaxSnippets(fieldName, HighlightingUtils.getParams(request));
}
-}
-
-/** Orders Tokens in a window first by their startOffset ascending.
- * endOffset is currently ignored.
- * This is meant to work around fickleness in the highlighter only. It
- * can mess up token positions and should not be used for indexing or querying.
- */
-class TokenOrderingFilter extends TokenFilter {
- private final int windowSize;
- private final LinkedList<Token> queue = new LinkedList<Token>();
- private boolean done=false;
-
- protected TokenOrderingFilter(TokenStream input, int windowSize) {
- super(input);
- this.windowSize = windowSize;
+ Formatter getFormatterX(String fieldName, SolrQueryRequest request) {
+ return getFormatter(fieldName, HighlightingUtils.getParams(request));
}
-
- @Override
- public Token next() throws IOException {
- while (!done && queue.size() < windowSize) {
- Token newTok = input.next();
- if (newTok==null) {
- done=true;
- break;
- }
-
- // reverse iterating for better efficiency since we know the
- // list is already sorted, and most token start offsets will be too.
- ListIterator<Token> iter = queue.listIterator(queue.size());
- while(iter.hasPrevious()) {
- if (newTok.startOffset() >= iter.previous().startOffset()) {
- // insertion will be before what next() would return (what
- // we just compared against), so move back one so the insertion
- // will be after.
- iter.next();
- break;
- }
- }
- iter.add(newTok);
- }
-
- return queue.isEmpty() ? null : queue.removeFirst();
+ Fragmenter getFragmenterX(String fieldName, SolrQueryRequest request) {
+ return getFragmenter(fieldName, HighlightingUtils.getParams(request));
}
}
+
