| Index: trunk/lucene-search-2/src/org/wikimedia/lsearch/test/WikiQueryParserTest.java |
| — | — | @@ -366,6 +366,14 @@ |
| 367 | 367 | q = parser.parseFourPass("\"うろパン\"",NamespacePolicy.IGNORE,false); |
| 368 | 368 | assertEquals("contents:\"うろ ろハ ハン\" title:\"うろ ろハ ハン\"^2.0 (alttitle1:\"うろ ろハ ハン\"^6.0 alttitle2:\"うろ ろハ ハン\"^6.0 alttitle3:\"うろ ろハ ハン\"^6.0)",q.toString()); |
| 369 | 369 | |
| | 370 | + // Malayalam |
| | 371 | + analyzer = Analyzers.getSearcherAnalyzer("ml"); |
| | 372 | + bs = new FieldBuilder("ml").getBuilder(); |
| | 373 | + parser = new WikiQueryParser(bs.getFields().contents(),"0",analyzer,bs,NamespacePolicy.IGNORE); |
| | 374 | + q = parser.parseFourPass("കൊറിയ ",NamespacePolicy.IGNORE,false); |
| | 375 | + assertEquals("contents:കറയ title:കറയ^2.0 (alttitle1:കറയ^6.0 alttitle2:കറയ^6.0 alttitle3:കറയ^6.0)",q.toString()); |
| | 376 | + |
| | 377 | + |
| 370 | 378 | // Test field extraction |
| 371 | 379 | HashSet<NamespaceFilter> fs = parser.getFieldNamespaces("main:something [1]:else all:oh []:nja"); |
| 372 | 380 | assertEquals(3,fs.size()); |
| Index: trunk/lucene-search-2/src/org/wikimedia/lsearch/analyzers/WikiQueryParser.java |
| — | — | @@ -296,7 +296,7 @@ |
| 297 | 297 | continue; // ignore whitespaces |
| 298 | 298 | |
| 299 | 299 | // pluses and minuses, underscores can be within words, *,? are for wildcard queries |
| 300 | | - if(Character.isLetterOrDigit(ch) || ch=='-' || ch=='+' || ch=='_' || ch=='*'){ |
| | 300 | + if(!Character.isWhitespace(ch) && ch != ':' && ch != '(' && ch != ')' && ch !='[' && ch != ']' && ch != '.' && ch != ',' && ch != ';' && ch != '"'){ |
| 301 | 301 | if(length<buffer.length) |
| 302 | 302 | buffer[length++] = ch; |
| 303 | 303 | } else{ |