This is an automated email from the ASF dual-hosted git repository.
ggregory pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/commons-codec.git
The following commit(s) were added to refs/heads/master by this push:
new 3bf874e2 Optimize memory allocation
3bf874e2 is described below
commit 3bf874e2141dc08550c0b330c7a7006f358bb0f0
Author: Gary Gregory <garydgreg...@gmail.com>
AuthorDate: Mon Mar 18 09:27:34 2024 -0400
Optimize memory allocation
Add org.apache.commons.codec.language.bm.Rule.PhonemeExpr.size()
---
pom.xml | 6 +--
src/changes/changes.xml | 3 +-
.../commons/codec/language/bm/PhoneticEngine.java | 55 +++++++++++-----------
.../org/apache/commons/codec/language/bm/Rule.java | 27 +++++++++--
.../codec/language/bm/PhoneticEngineTest.java | 6 ++-
5 files changed, 60 insertions(+), 37 deletions(-)
diff --git a/pom.xml b/pom.xml
index 69f9392a..4a42ae67 100644
--- a/pom.xml
+++ b/pom.xml
@@ -28,7 +28,7 @@ limitations under the License.
</parent>
<groupId>commons-codec</groupId>
<artifactId>commons-codec</artifactId>
- <version>1.16.2-SNAPSHOT</version>
+ <version>1.17.0-SNAPSHOT</version>
<name>Apache Commons Codec</name>
<inceptionYear>2002</inceptionYear>
<description>
@@ -273,9 +273,9 @@ limitations under the License.
<checkstyle.config.file>${basedir}/src/conf/checkstyle.xml</checkstyle.config.file>
<jacoco.skip>false</jacoco.skip>
<!-- Commons Release Plugin -->
- <commons.release.version>1.16.1</commons.release.version>
+ <commons.release.version>1.17.0</commons.release.version>
<commons.bc.version>1.16.0</commons.bc.version>
- <commons.bc.next>1.16.2</commons.bc.next>
+ <commons.bc.next>1.17.1</commons.bc.next>
<commons.rc.version>RC1</commons.rc.version>
<commons.release.isDistModule>true</commons.release.isDistModule>
<commons.distSvnStagingUrl>scm:svn:https://dist.apache.org/repos/dist/dev/commons/${commons.componentid}</commons.distSvnStagingUrl>
diff --git a/src/changes/changes.xml b/src/changes/changes.xml
index 7180ce7d..bdb6e434 100644
--- a/src/changes/changes.xml
+++ b/src/changes/changes.xml
@@ -43,8 +43,9 @@ The <action> type attribute can be add,update,fix,remove.
<author>Apache Commons Developers</author>
</properties>
<body>
- <release version="1.16.2" date="2024-MM-DD" description="Feature and fix
release. Requires a minimum of Java 8.">
+ <release version="1.17.0" date="YYYY-MM-DD" description="Feature and fix
release. Requires a minimum of Java 8.">
<!-- ADD -->
+ <action type="add" dev="ggregory" due-to="Gary Gregory">Add
org.apache.commons.codec.language.bm.Rule.PhonemeExpr.size().</action>
<!-- FIX -->
<!-- UPDATE -->
<action dev="ggregory" type="update"
due-to="Dependabot, Gary Gregory">Bump org.apache.commons:commons-parent from
66 to 67 #250.</action>
diff --git
a/src/main/java/org/apache/commons/codec/language/bm/PhoneticEngine.java
b/src/main/java/org/apache/commons/codec/language/bm/PhoneticEngine.java
index 3621487d..84fbe351 100644
--- a/src/main/java/org/apache/commons/codec/language/bm/PhoneticEngine.java
+++ b/src/main/java/org/apache/commons/codec/language/bm/PhoneticEngine.java
@@ -106,9 +106,8 @@ public class PhoneticEngine {
* @param maxPhonemes the maximum number of phonemes to build up
*/
public void apply(final Rule.PhonemeExpr phonemeExpr, final int
maxPhonemes) {
- final Set<Rule.Phoneme> newPhonemes = new
LinkedHashSet<>(maxPhonemes);
-
- EXPR: for (final Rule.Phoneme left : this.phonemes) {
+ final Set<Rule.Phoneme> newPhonemes = new
LinkedHashSet<>(Math.min(phonemes.size() * phonemeExpr.size(), maxPhonemes));
+ EXPR: for (final Rule.Phoneme left : phonemes) {
for (final Rule.Phoneme right : phonemeExpr.getPhonemes()) {
final LanguageSet languages =
left.getLanguages().restrictTo(right.getLanguages());
if (!languages.isEmpty()) {
@@ -122,9 +121,8 @@ public class PhoneticEngine {
}
}
}
-
- this.phonemes.clear();
- this.phonemes.addAll(newPhonemes);
+ phonemes.clear();
+ phonemes.addAll(newPhonemes);
}
/**
@@ -133,7 +131,7 @@ public class PhoneticEngine {
* @return the phoneme set
*/
public Set<Rule.Phoneme> getPhonemes() {
- return this.phonemes;
+ return phonemes;
}
/**
@@ -155,22 +153,24 @@ public class PhoneticEngine {
* processed already), and {@code found} indicates if a matching rule was
found or not. In the case where a
* matching rule was found, {@code phonemeBuilder} is replaced with a new
builder containing the phonemes
* updated by the matching rule.
- *
+ * <p>
* Although this class is not thread-safe (it has mutable unprotected
fields), it is not shared between threads
* as it is constructed as needed by the calling methods.
+ * </p>
+ *
* @since 1.6
*/
private static final class RulesApplication {
+
private final Map<String, List<Rule>> finalRules;
private final CharSequence input;
-
private final PhonemeBuilder phonemeBuilder;
private int i;
private final int maxPhonemes;
private boolean found;
- public RulesApplication(final Map<String, List<Rule>> finalRules,
final CharSequence input,
- final PhonemeBuilder phonemeBuilder, final int
i, final int maxPhonemes) {
+ public RulesApplication(final Map<String, List<Rule>> finalRules,
final CharSequence input, final PhonemeBuilder phonemeBuilder, final int i,
+ final int maxPhonemes) {
Objects.requireNonNull(finalRules, "finalRules");
this.finalRules = finalRules;
this.phonemeBuilder = phonemeBuilder;
@@ -180,11 +180,11 @@ public class PhoneticEngine {
}
public int getI() {
- return this.i;
+ return i;
}
public PhonemeBuilder getPhonemeBuilder() {
- return this.phonemeBuilder;
+ return phonemeBuilder;
}
/**
@@ -195,31 +195,31 @@ public class PhoneticEngine {
* @return {@code this}
*/
public RulesApplication invoke() {
- this.found = false;
+ found = false;
int patternLength = 1;
- final List<Rule> rules = this.finalRules.get(input.subSequence(i,
i + patternLength));
+ final List<Rule> rules = finalRules.get(input.subSequence(i, i +
patternLength));
if (rules != null) {
for (final Rule rule : rules) {
final String pattern = rule.getPattern();
patternLength = pattern.length();
- if (rule.patternAndContextMatches(this.input, this.i)) {
- this.phonemeBuilder.apply(rule.getPhoneme(),
maxPhonemes);
- this.found = true;
+ if (rule.patternAndContextMatches(input, i)) {
+ phonemeBuilder.apply(rule.getPhoneme(), maxPhonemes);
+ found = true;
break;
}
}
}
- if (!this.found) {
+ if (!found) {
patternLength = 1;
}
- this.i += patternLength;
+ i += patternLength;
return this;
}
public boolean isFound() {
- return this.found;
+ return found;
}
}
@@ -269,11 +269,11 @@ public class PhoneticEngine {
* the type of names it will use
* @param ruleType
* the type of rules it will apply
- * @param concat
+ * @param concatenate
* if it will concatenate multiple encodings
*/
- public PhoneticEngine(final NameType nameType, final RuleType ruleType,
final boolean concat) {
- this(nameType, ruleType, concat, DEFAULT_MAX_PHONEMES);
+ public PhoneticEngine(final NameType nameType, final RuleType ruleType,
final boolean concatenate) {
+ this(nameType, ruleType, concatenate, DEFAULT_MAX_PHONEMES);
}
/**
@@ -283,20 +283,19 @@ public class PhoneticEngine {
* the type of names it will use
* @param ruleType
* the type of rules it will apply
- * @param concat
+ * @param concatenate
* if it will concatenate multiple encodings
* @param maxPhonemes
* the maximum number of phonemes that will be handled
* @since 1.7
*/
- public PhoneticEngine(final NameType nameType, final RuleType ruleType,
final boolean concat,
- final int maxPhonemes) {
+ public PhoneticEngine(final NameType nameType, final RuleType ruleType,
final boolean concatenate, final int maxPhonemes) {
if (ruleType == RuleType.RULES) {
throw new IllegalArgumentException("ruleType must not be " +
RuleType.RULES);
}
this.nameType = nameType;
this.ruleType = ruleType;
- this.concat = concat;
+ this.concat = concatenate;
this.lang = Lang.instance(nameType);
this.maxPhonemes = maxPhonemes;
}
diff --git a/src/main/java/org/apache/commons/codec/language/bm/Rule.java
b/src/main/java/org/apache/commons/codec/language/bm/Rule.java
index 641719e7..728e4bd0 100644
--- a/src/main/java/org/apache/commons/codec/language/bm/Rule.java
+++ b/src/main/java/org/apache/commons/codec/language/bm/Rule.java
@@ -166,6 +166,11 @@ public class Rule {
return new Phoneme(this.phonemeText.toString(),
this.languages.merge(lang));
}
+ @Override
+ public int size() {
+ return 1;
+ }
+
@Override
public String toString() {
return phonemeText.toString() + "[" + languages + "]";
@@ -174,19 +179,35 @@ public class Rule {
public interface PhonemeExpr {
Iterable<Phoneme> getPhonemes();
+
+ /**
+ * Gets the expression size in phonemes.
+ *
+ * @return the expression size in phonemes.
+ * @since 1.17.0
+ */
+ default int size() {
+ // All implementations are int-bound.
+ return (int)
Math.min(getPhonemes().spliterator().getExactSizeIfKnown(), Integer.MAX_VALUE);
+ }
}
public static final class PhonemeList implements PhonemeExpr {
- private final List<Phoneme> phonemes;
+ private final List<Phoneme> phonemeList;
public PhonemeList(final List<Phoneme> phonemes) {
- this.phonemes = phonemes;
+ this.phonemeList = phonemes;
}
@Override
public List<Phoneme> getPhonemes() {
- return this.phonemes;
+ return phonemeList;
+ }
+
+ @Override
+ public int size() {
+ return phonemeList.size();
}
}
diff --git
a/src/test/java/org/apache/commons/codec/language/bm/PhoneticEngineTest.java
b/src/test/java/org/apache/commons/codec/language/bm/PhoneticEngineTest.java
index b3eddb94..7c949f17 100644
--- a/src/test/java/org/apache/commons/codec/language/bm/PhoneticEngineTest.java
+++ b/src/test/java/org/apache/commons/codec/language/bm/PhoneticEngineTest.java
@@ -43,7 +43,8 @@ public class PhoneticEngineTest {
Arguments.of("SntJohn-Smith", "sntjonsmit", NameType.GENERIC,
RuleType.EXACT, Boolean.TRUE, TEN),
Arguments.of("d'ortley", "(ortlaj|ortlej)-(dortlaj|dortlej)",
NameType.GENERIC, RuleType.EXACT, Boolean.TRUE, TEN),
Arguments.of("van helsing",
"(elSink|elsink|helSink|helsink|helzink|xelsink)-(banhelsink|fanhelsink|fanhelzink|vanhelsink|vanhelzink|vanjelsink)",
NameType.GENERIC, RuleType.EXACT, Boolean.FALSE, TEN),
- Arguments.of("Judenburg",
"iudnbYrk|iudnbirk|iudnburk|xudnbirk|xudnburk|zudnbirk|zudnburk",
NameType.GENERIC, RuleType.APPROX, Boolean.TRUE, TEN)
+ Arguments.of("Judenburg",
"iudnbYrk|iudnbirk|iudnburk|xudnbirk|xudnburk|zudnbirk|zudnburk",
NameType.GENERIC, RuleType.APPROX, Boolean.TRUE, TEN),
+ Arguments.of("Judenburg",
"iudnbYrk|iudnbirk|iudnburk|xudnbirk|xudnburk|zudnbirk|zudnburk",
NameType.GENERIC, RuleType.APPROX, Boolean.TRUE, Integer.MAX_VALUE)
);
// @formatter:on
}
@@ -54,7 +55,8 @@ public class PhoneticEngineTest {
Arguments.of("bar", "bar|bor|var|vor", NameType.ASHKENAZI,
RuleType.APPROX, Boolean.FALSE, TEN),
Arguments.of("al", "|al", NameType.SEPHARDIC, RuleType.APPROX,
Boolean.FALSE, TEN),
Arguments.of("da", "da|di", NameType.GENERIC, RuleType.EXACT,
Boolean.FALSE, TEN),
- Arguments.of("'''", "", NameType.SEPHARDIC, RuleType.APPROX,
Boolean.FALSE, TEN)
+ Arguments.of("'''", "", NameType.SEPHARDIC, RuleType.APPROX,
Boolean.FALSE, TEN),
+ Arguments.of("'''", "", NameType.SEPHARDIC, RuleType.APPROX,
Boolean.FALSE, Integer.MAX_VALUE)
);
// @formatter:on
}
