Github user alessandrobenedetti commented on a diff in the pull request:
https://github.com/apache/lucene-solr/pull/482#discussion_r227471350
--- Diff:
lucene/core/src/test/org/apache/lucene/analysis/TestStopFilter.java ---
@@ -47,58 +48,65 @@ public void testStopFilt() throws IOException {
assertTokenStreamContents(stream, new String[] { "Now", "The" });
}
+
+ private void logStopwords(String name, List<String> stopwords){
+ // helper method: converts a list
+ log(String.format("stopword list \"%s:\"", name));
+ for (int i = 0; i < stopwords.size(); i++) {
+ log(String.format("stopword (%d): %s ", i, stopwords.get(i)));
+ }
+ log("----------");
+ }
/**
* Test Position increments applied by StopFilter with and without
enabling this option.
*/
- public void testStopPositons() throws IOException {
+ public void testStopPositions() throws IOException {
+ final int NUMBER_OF_TOKENS = 20;
StringBuilder sb = new StringBuilder();
- ArrayList<String> a = new ArrayList<>();
- for (int i=0; i<20; i++) {
- String w = English.intToEnglish(i).trim();
- sb.append(w).append(" ");
- if (i%3 != 0) a.add(w);
+ List<String> stopwords = new ArrayList<>(NUMBER_OF_TOKENS);
+ for (int i = 0; i < NUMBER_OF_TOKENS; i++) {
+ String token = English.intToEnglish(i).trim();
+ sb.append(token).append(' ');
+ if (i%3 != 0) stopwords.add(token);
}
log(sb.toString());
- String stopWords[] = a.toArray(new String[0]);
- for (int i=0; i<a.size(); i++) log("Stop: "+stopWords[i]);
- CharArraySet stopSet = StopFilter.makeStopSet(stopWords);
+ CharArraySet stopSet = StopFilter.makeStopSet(stopwords);
+ logStopwords("All stopwords", stopwords);
// with increments
StringReader reader = new StringReader(sb.toString());
final MockTokenizer in = new MockTokenizer(MockTokenizer.WHITESPACE,
false);
in.setReader(reader);
- StopFilter stpf = new StopFilter(in, stopSet);
- doTestStopPositons(stpf);
+ StopFilter stopfilter = new StopFilter(in, stopSet);
+ doTestStopwordsPositions(stopfilter);
// with increments, concatenating two stop filters
- ArrayList<String> a0 = new ArrayList<>();
- ArrayList<String> a1 = new ArrayList<>();
- for (int i=0; i<a.size(); i++) {
- if (i%2==0) {
- a0.add(a.get(i));
+ List<String> evenStopwords = new ArrayList<>(stopwords.size());
+ List<String> oddStopwords = new ArrayList<>(stopwords.size());
+ for (int i=0; i < stopwords.size(); i++) {
+ if (i%2 == 0) {
+ evenStopwords.add(stopwords.get(i));
} else {
- a1.add(a.get(i));
+ oddStopwords.add(stopwords.get(i));
}
}
- String stopWords0[] = a0.toArray(new String[0]);
- for (int i=0; i<a0.size(); i++) log("Stop0: "+stopWords0[i]);
- String stopWords1[] = a1.toArray(new String[0]);
- for (int i=0; i<a1.size(); i++) log("Stop1: "+stopWords1[i]);
- CharArraySet stopSet0 = StopFilter.makeStopSet(stopWords0);
- CharArraySet stopSet1 = StopFilter.makeStopSet(stopWords1);
+ CharArraySet evenStopSet = StopFilter.makeStopSet(evenStopwords);
--- End diff --
I guess the reason behind the naming of "evenStopWords" and "oddStopWords"
is because of the way they are extracted from the original stop set.
But being the original stopwords set all numbers except the ones divisible
by 3, what ends up in the even and odd stopword set is not exactly even or odd.
maybe something like :
firstStopwordsHalf
secondStopwordsHalf
but I lack of fantasy at the moment to identify good names :)
---
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
