Revision: 20327
          http://sourceforge.net/p/gate/code/20327
Author:   ian_roberts
Date:     2019-05-30 14:01:23 +0000 (Thu, 30 May 2019)
Log Message:
-----------
Back-porting 
https://github.com/GateNLP/mimir/commit/5214f32689bd8ff1ebef126dbd870bbf4ed98e50
 and 
https://github.com/GateNLP/mimir/commit/58a682eb7f870d05fb2286fa10c36edb9528a3b5
 from Mimir version 6.1-SNAPSHOT: in cases where the index has more than just 
the "head" batch, the total number of documents reported by the index was off 
by one, and the direct index lexical cluster was ignoring the last tail-N batch

Modified Paths:
--------------
    mimir/trunk/mimir-core/src/gate/mimir/index/AtomicIndex.java

Modified: mimir/trunk/mimir-core/src/gate/mimir/index/AtomicIndex.java
===================================================================
--- mimir/trunk/mimir-core/src/gate/mimir/index/AtomicIndex.java        
2019-05-30 09:31:31 UTC (rev 20326)
+++ mimir/trunk/mimir-core/src/gate/mimir/index/AtomicIndex.java        
2019-05-30 14:01:23 UTC (rev 20327)
@@ -489,10 +489,10 @@
           new ContiguousDocumentalStrategy(cutPoints),
           false, // flat = all component indexes have the same term list
           bloomFilters, // Bloom Filters
-          numberOfDocuments, 
-          numberOfTerms, 
-          numberOfPostings, 
-          numberOfOccurences, 
+          numberOfDocuments == -1 ? -1 : numberOfDocuments + 1, 
+          numberOfTerms == -1 ? -1 : numberOfTerms + 1, 
+          numberOfPostings == -1 ? -1 : numberOfPostings + 1, 
+          numberOfOccurences == -1 ? -1 : numberOfOccurences + 1, 
           maxCount, 
           null, // payload
           true, // hasCounts 
@@ -517,9 +517,9 @@
     
     // prepare the lexical cluster
     Index[] indexes = new Index[batches.size()];
-    int[] cutPoints = new int[indexes.length];
+    int[] cutPoints = new int[indexes.length + 1];
     cutPoints[0] = 0;
-    String[] cutPointTerms = new String[indexes.length];
+    String[] cutPointTerms = new String[indexes.length + 1];
     cutPointTerms[0] = longToTerm(0);
     int numberOfTerms = -1;
     int numberOfDocuments = -1;
@@ -554,10 +554,10 @@
     return new LexicalCluster(indexes,
           new ContiguousLexicalStrategy(cutPoints, cutPointTerms),
           bloomFilters, // Bloom Filters
-          numberOfDocuments, 
-          numberOfTerms, 
-          numberOfPostings, 
-          numberOfOccurences, 
+          numberOfDocuments == -1 ? -1 : numberOfDocuments + 1, 
+          numberOfTerms == -1 ? -1 : numberOfTerms + 1, 
+          numberOfPostings == -1 ? -1 : numberOfPostings + 1, 
+          numberOfOccurences == -1 ? -1 : numberOfOccurences + 1, 
           maxCount, 
           null, // payload
           true, // hasCounts 

This was sent by the SourceForge.net collaborative development platform, the 
world's largest Open Source development site.



_______________________________________________
GATE-cvs mailing list
GATE-cvs@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/gate-cvs

Reply via email to