mstoeck3/autopsy-flatpak
1
0
Fork 0
mirror of https://github.com/overcuriousity/autopsy-flatpak.git synced 2025-07-13 08:26:15 +00:00
Code Issues Packages Projects Releases Wiki Activity

sort list of docs based on ID before picking unique chunk

Browse Source
This commit is contained in:
Brian Carrier 2015-05-12 13:10:54 -04:00
parent c83f80c3f4
commit 671a2d1e68
1 changed files with 29 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

38
KeywordSearch/src/org/sleuthkit/autopsy/keywordsearch/LuceneQuery.java
View File

@ -20,6 +20,7 @@ package org.sleuthkit.autopsy.keywordsearch;
import java.util.ArrayList; import java.util.ArrayList;
import java.util.Collection; import java.util.Collection;
import java.util.Collections;
import java.util.Comparator; import java.util.Comparator;
import java.util.List; import java.util.List;
import java.util.Map; import java.util.Map;
@ -214,7 +215,7 @@ class LuceneQuery implements KeywordSearchQuery {
Map<String, Map<String, List<String>>> highlightResponse = response.getHighlighting(); Map<String, Map<String, List<String>>> highlightResponse = response.getHighlighting();
// get the unique set of files with hits // get the unique set of files with hits
Set<SolrDocument> uniqueSolrDocumentsWithHits = filterDuplicateSolrDocuments(resultList); Set<SolrDocument> uniqueSolrDocumentsWithHits = filterOneHitPerDocument(resultList);
allMatchesFetched = start + MAX_RESULTS >= resultList.getNumFound(); allMatchesFetched = start + MAX_RESULTS >= resultList.getNumFound();
@ -305,7 +306,24 @@ class LuceneQuery implements KeywordSearchQuery {
* @param resultList * @param resultList
* @return * @return
*/ */
private Set<SolrDocument> filterDuplicateSolrDocuments(SolrDocumentList resultList) { private Set<SolrDocument> filterOneHitPerDocument(SolrDocumentList resultList) {
// sort the list so that we consistently pick the same chunk each time.
// note this sort is doing a string comparison and not an integer comparison, so
// chunk 10 will be smaller than chunk 9.
Collections.sort(resultList, new Comparator<SolrDocument>() {
@Override
public int compare(SolrDocument left, SolrDocument right) {
// ID is in the form of ObjectId_Chunk
String leftID = left.getFieldValue(Server.Schema.ID.toString()).toString();
String rightID = right.getFieldValue(Server.Schema.ID.toString()).toString();
return leftID.compareTo(rightID);
}
});
// NOTE: We could probably just iterate through the list and compare each ID with the
// previous ID to get the unique documents faster than using this set now that the list
// is sorted.
Set<SolrDocument> solrDocumentsWithMatches = new TreeSet<>(new SolrDocumentComparatorIgnoresChunkId()); Set<SolrDocument> solrDocumentsWithMatches = new TreeSet<>(new SolrDocumentComparatorIgnoresChunkId());
solrDocumentsWithMatches.addAll(resultList); solrDocumentsWithMatches.addAll(resultList);
return solrDocumentsWithMatches; return solrDocumentsWithMatches;
@ -464,24 +482,26 @@ class LuceneQuery implements KeywordSearchQuery {
public int compare(SolrDocument left, SolrDocument right) { public int compare(SolrDocument left, SolrDocument right) {
// ID is in the form of ObjectId_Chunk // ID is in the form of ObjectId_Chunk
String idName = Server.Schema.ID.toString(); final String idName = Server.Schema.ID.toString();
// get object id of left doc
String leftID = left.getFieldValue(idName).toString(); String leftID = left.getFieldValue(idName).toString();
int index = leftID.indexOf(Server.ID_CHUNK_SEP); int index = leftID.indexOf(Server.ID_CHUNK_SEP);
if (index != -1) { if (index != -1) {
leftID = leftID.substring(0, index); leftID = leftID.substring(0, index);
} }
// get object id of right doc
String rightID = right.getFieldValue(idName).toString(); String rightID = right.getFieldValue(idName).toString();
index = rightID.indexOf(Server.ID_CHUNK_SEP); index = rightID.indexOf(Server.ID_CHUNK_SEP);
if (index != -1) { if (index != -1) {
rightID = rightID.substring(0, index); rightID = rightID.substring(0, index);
} }
if(Integer.parseInt(leftID) < Integer.parseInt(rightID)) Integer leftInt = new Integer(leftID);
return -1; Integer rightInt = new Integer(rightID);
if(Integer.parseInt(leftID) > Integer.parseInt(rightID)) return leftInt.compareTo(rightInt);
return 1;
return 0;
} }
} }
} }