Jaechang Nam created JOSHUA-329:
-----------------------------------
Summary: A suspicious use of incrementer in for statement
Key: JOSHUA-329
URL: https://issues.apache.org/jira/browse/JOSHUA-329
Project: Joshua
Issue Type: Bug
Reporter: Jaechang Nam
Priority: Trivial
In a recent snapshot of the github mirror, I've found a suspicious incrementer
in src/main/java/org/apache/joshua/decoder/ff/lm/LanguageModelFF.java.
{code:java}
269 for (int i = 0; i < tokens.length; i++) {
270 if (tokens[i] > 0) { // skip nonterminals
271 for (int j = 0; j < alignments.length; j += 2) {
272 if (alignments[j] == i) {
273 String annotation = sentence.getAnnotation((int)alignments[i]
+ begin, "class");
274 if (annotation != null) {
275 // System.err.println(String.format(" word
%d source %d abs %d annotation %d/%s" ,
276 // i, alignments[i], alignments[i] +
begin, annotation, Vocabulary.word(annotation)));
277 tokens[i] = Vocabulary.id(annotation);
278 break;
279 }
280 }
281 }
282 }
283 }
{code}
In Line 273, alignments[i] should be alignments[j] if tokens.length is not same
as alignments.length? Since I don't have domain knowledge, this may not be
correct but just wanted to report this in case.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
