Apache
/
lucene
mirror of https://github.com/apache/lucene.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Revert "remove nori/tools dependency on ICU" 

This reverts commit dd7b3d8d95.
This commit is contained in:
Namgyu Kim 2019-07-23 23:36:26 +09:00
parent d92159b896
commit c5741c201f
2 changed files with 20 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
lucene/analysis/nori/build.xml
View File

@ -57,8 +57,13 @@
<untar src="${build.dir}/${dict.version}.tar" dest="${build.dir}"/> <untar src="${build.dir}/${dict.version}.tar" dest="${build.dir}"/>
</target> </target>
<path id="tools.dependencies">
<fileset dir="../icu/lib"/>
</path>
<path id="tools.classpath"> <path id="tools.classpath">
<path refid="classpath"/> <path refid="classpath"/>
<path refid="tools.dependencies"/>
<pathelement location="${build.dir}/classes/java"/> <pathelement location="${build.dir}/classes/java"/>
<pathelement location="${build.dir}/classes/tools"/> <pathelement location="${build.dir}/classes/tools"/>
</path> </path>
@ -90,7 +95,14 @@
</sequential> </sequential>
</target> </target>
<target name="compile-tools" depends="compile-core, common.compile-tools"> <!-- we don't actually need to compile this thing, we just want its lib -->
<target name="resolve-icu">
<ant dir="../icu/" target="resolve" inheritAll="false">
<propertyset refid="uptodate.and.compiled.properties"/>
</ant>
</target>
<target name="compile-tools" depends="resolve-icu, compile-core, common.compile-tools">
<compile <compile
srcdir="src/tools/java" srcdir="src/tools/java"
destdir="${build.dir}/classes/tools"> destdir="${build.dir}/classes/tools">

12
lucene/analysis/nori/src/tools/java/org/apache/lucene/analysis/ko/util/TokenInfoDictionaryBuilder.java
View File

@ -25,7 +25,6 @@ import java.io.InputStreamReader;
import java.nio.charset.Charset; import java.nio.charset.Charset;
import java.nio.charset.CharsetDecoder; import java.nio.charset.CharsetDecoder;
import java.nio.charset.CodingErrorAction; import java.nio.charset.CodingErrorAction;
import java.text.Normalizer;
import java.util.ArrayList; import java.util.ArrayList;
import java.util.Arrays; import java.util.Arrays;
import java.util.Collections; import java.util.Collections;
@ -36,6 +35,7 @@ import org.apache.lucene.util.IntsRefBuilder;
import org.apache.lucene.util.fst.Builder; import org.apache.lucene.util.fst.Builder;
import org.apache.lucene.util.fst.FST; import org.apache.lucene.util.fst.FST;
import com.ibm.icu.text.Normalizer2;
import org.apache.lucene.util.fst.PositiveIntOutputs; import org.apache.lucene.util.fst.PositiveIntOutputs;
public class TokenInfoDictionaryBuilder { public class TokenInfoDictionaryBuilder {
@ -45,11 +45,13 @@ public class TokenInfoDictionaryBuilder {
private String encoding = "utf-8"; private String encoding = "utf-8";
private Normalizer.Form normalForm; private boolean normalizeEntries = false;
private Normalizer2 normalizer;
public TokenInfoDictionaryBuilder(String encoding, boolean normalizeEntries) { public TokenInfoDictionaryBuilder(String encoding, boolean normalizeEntries) {
this.encoding = encoding; this.encoding = encoding;
this.normalForm = normalizeEntries ? Normalizer.Form.NFKC : null; this.normalizeEntries = normalizeEntries;
this.normalizer = normalizeEntries ? Normalizer2.getInstance(null, "nfkc", Normalizer2.Mode.COMPOSE) : null;
} }
public TokenInfoDictionaryWriter build(String dirname) throws IOException { public TokenInfoDictionaryWriter build(String dirname) throws IOException {
@ -86,10 +88,10 @@ public class TokenInfoDictionaryBuilder {
} }
// NFKC normalize dictionary entry // NFKC normalize dictionary entry
if (normalForm != null) { if (normalizeEntries) {
String[] normalizedEntry = new String[entry.length]; String[] normalizedEntry = new String[entry.length];
for (int i = 0; i < entry.length; i++) { for (int i = 0; i < entry.length; i++) {
normalizedEntry[i] = Normalizer.normalize(entry[i], normalForm); normalizedEntry[i] = normalizer.normalize(entry[i]);
} }
lines.add(normalizedEntry); lines.add(normalizedEntry);
} else { } else {