[
https://issues.apache.org/jira/browse/OPENNLP-1505?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17746449#comment-17746449
]
ASF GitHub Bot commented on OPENNLP-1505:
-----------------------------------------
mawiesne commented on code in PR #543:
URL: https://github.com/apache/opennlp/pull/543#discussion_r1272235280
##########
opennlp-tools/src/main/java/opennlp/tools/util/StringUtil.java:
##########
@@ -79,6 +81,30 @@ public static String toLowerCase(CharSequence string) {
return new String(cp, 0, cp.length);
}
+ public static CharBuffer toLowerCaseCharBuffer(CharSequence sequence) {
+ CharBuffer result = CharBuffer.allocate(sequence.length());
+ for (int cp : sequence.codePoints().toArray()) {
+ for (char c : Character.toChars(Character.toLowerCase(cp))) {
+ result.append(c);
+ }
+ }
+ result.clear();
+ return result;
+ }
+
+ /*
+ public static CharBuffer toLowerCaseCharBuffer(CharSequence string) {
Review Comment:
Will do / beautify once PR passes thorough testing from the community or
build servers.
> Reduce object creation in NGramCharModel and StringUtil
> -------------------------------------------------------
>
> Key: OPENNLP-1505
> URL: https://issues.apache.org/jira/browse/OPENNLP-1505
> Project: OpenNLP
> Issue Type: Improvement
> Components: Language Detector
> Affects Versions: 2.2.0
> Reporter: Martin Wiesner
> Assignee: Martin Wiesner
> Priority: Major
> Fix For: 2.2.1
>
>
> During a profiling session, I noticed that many tests in
> opennlp.tools.langdetect take quite some time for execution. Digging deeper
> into those tests, it quickly became obvious that StringUtil#toLowerCase() was
> creating new Strings for every call of this method (see
> NGramCharModel#add(...) lines 99 to 108.
> Being called in NGramCharModel quite frequently, this resulted in creation of
> millions of String objects during building ngrams for given input.
> Aims:
> * Reduce objection creation and thus creation of millions of string objects
> * Improve runtime of the langdetect tests (and potentially others)
> Idea:
> * Use (Heap)CharBuffer instead of String so that underlying char arrays can
> be re-used, instead of copying the chars over to a new string for each
> "toLowerCase"...
> Note:
> * A corresponding patch / PR should be tested with/against the Evaluation
> suite.
> Comments welcome.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
