Changeset 12913

Timestamp:

2006-09-28T16:54:36+12:00 (18 years ago)

Author:

kjdon

Message:

we now check for existence of stem index for our stem method in FindWordNumbers - if it wasn't built, then we assume stem is 0

File:

• trunk/gsdl/src/mgpp/text/Terms.cpp (modified) (4 diffs)

trunk/gsdl/src/mgpp/text/Terms.cpp

@@ -211 +211 @@
   equivWords.erase (equivWords.begin(), equivWords.end());
 
+  // if the stem method specified is not a valid one (i.e. there was no appropriate stem index, then we set it to 0)
+  // unless we have partial matching, in which case we are not doing stem indexes anyway.
+  if (!(stemMethod & STEM_PARTIAL_MATCH) && indexData.stemFile[stemMethod-1] == NULL) {
+    cerr << "Stem index for method "<<stemMethod<< " was not built, so not doing stemming\n";
+    stemMethod = 0;
+  }
   /* [JFG - Mar 06: Accent folding patch] */
   /* use flag PARTIAL_MATCH */  
@@ -230 +236 @@
     } else {
       // partial matching,
-      PartialMatchSearchWordBlockDictEl (indexData.dictFile, indexData.biWords, indexData.bdh.entries_per_wblk, indexData.bdh.word_dict_size, numLevels, term, wordDictEl, equivWords, 
-            (stemMethod & STEM_CaseFolding)? true : false);
-            // TODO: Accent Folding is not handled here!!
+      PartialMatchSearchWordBlockDictEl (indexData.dictFile, indexData.biWords, indexData.bdh.entries_per_wblk, indexData.bdh.word_dict_size, numLevels, term, wordDictEl, equivWords, (stemMethod & STEM_CaseFolding)? true : false);
+      // TODO: Accent Folding is not handled here!!
       return;
     }
@@ -238 +243 @@
               
   // need to stem this word and find it in the blocked stem index
-  unsigned char  mgWord[MAXSTEMLEN + 1];
+  unsigned char mgWord[MAXSTEMLEN + 1];
   UCArray stemTerm;
   unsigned long stemmerNum = 0;
@@ -247 +252 @@
     //TODO: throw an error here
   }
+
   stemmerNum = indexData.sih[stemMethod-1].stemmer_num;