Precompile regular expressions in Lang.loadFromResource(String,

Languages)

- Precompile regular expressions in PhoneticEngine.encode(String,
LanguageSet)
- Precompile regular expressions in
org.apache.commons.codec.language.bm.Rule.parse*(*)

Author: garydgregory

src/changes/changes.xml

@@ -48,8 +48,11 @@ The <action> type attribute can be add,update,fix,remove.
       <action type="fix" dev="ggregory" due-to="Gary Gregory">Remove -nouses directive from maven-bundle-plugin. OSGi package imports now state 'uses' definitions for package imports, this doesn't affect JPMS (from org.apache.commons:commons-parent:80).</action>
       <action type="fix" dev="ggregory" due-to="Gary Gregory">Refactor DigestUtils.updateDigest(MessageDigest, File) to use NIO.</action>
       <action type="fix" dev="ggregory" due-to="Gary Gregory" issue="CODEC-328" >Clarify Javadoc for org.apache.commons.codec.digest.UnixCrypt.crypt(byte[],String).</action>
-      <action type="fix" dev="ggregory" due-to="Gary Gregory">Precompile and resuse x3 regular expression in DaitchMokotoffSoundex.Rule.</action>
-      <action type="fix" dev="ggregory" due-to="Gary Gregory">Precompile and resuse regular expressions in DaitchMokotoffSoundex.parseRules(Scanner, String, Map, Map).</action>
+      <action type="fix" dev="ggregory" due-to="Gary Gregory">Precompile regular expressions in DaitchMokotoffSoundex.Rule.</action>
+      <action type="fix" dev="ggregory" due-to="Gary Gregory">Precompile regular expressions in DaitchMokotoffSoundex.parseRules(Scanner, String, Map, Map).</action>
+      <action type="fix" dev="ggregory" due-to="Gary Gregory">Precompile regular expressions in Lang.loadFromResource(String, Languages).</action>
+      <action type="fix" dev="ggregory" due-to="Gary Gregory">Precompile regular expressions in PhoneticEngine.encode(String, LanguageSet).</action>
+      <action type="fix" dev="ggregory" due-to="Gary Gregory">Precompile regular expressions in org.apache.commons.codec.language.bm.Rule.parse*(*).</action>
       <!-- ADD -->
       <action type="add" dev="ggregory" due-to="Gary Gregory">Add HmacUtils.hmac(Path).</action>      
       <action type="add" dev="ggregory" due-to="Gary Gregory">Add HmacUtils.hmacHex(Path).</action>      

src/main/java/org/apache/commons/codec/language/bm/Lang.java

@@ -103,6 +103,8 @@ public boolean matches(final String txt) {
 
     private static final String LANGUAGE_RULES_RN = "/org/apache/commons/codec/language/bm/%s_lang.txt";
 
+    private static final Pattern PLUS = Pattern.compile("\\+");
+
     static {
         for (final NameType s : NameType.values()) {
             LANGS.put(s, loadFromResource(String.format(LANGUAGE_RULES_RN, s.getName()), Languages.getInstance(s)));
@@ -163,15 +165,15 @@ public static Lang loadFromResource(final String languageRulesResourceName, fina
                     }
 
                     // split it up
-                    final String[] parts = line.split("\\s+");
+                    final String[] parts = ResourceConstants.SPACES.split(line);
 
                     if (parts.length != 3) {
                         throw new IllegalArgumentException("Malformed line '" + rawLine +
                                 "' in language resource '" + languageRulesResourceName + "'");
                     }
 
                     final Pattern pattern = Pattern.compile(parts[0]);
-                    final String[] langs = parts[1].split("\\+");
+                    final String[] langs = PLUS.split(parts[1]);
                     final boolean accept = parts[2].equals("true");
 
                     rules.add(new LangRule(pattern, new HashSet<>(Arrays.asList(langs)), accept));

src/main/java/org/apache/commons/codec/language/bm/PhoneticEngine.java

@@ -29,6 +29,7 @@
 import java.util.Objects;
 import java.util.Set;
 import java.util.TreeMap;
+import java.util.regex.Pattern;
 import java.util.stream.Collectors;
 
 import org.apache.commons.codec.language.bm.Languages.LanguageSet;
@@ -227,6 +228,8 @@ public boolean isFound() {
 
     private static final Map<NameType, Set<String>> NAME_PREFIXES = new EnumMap<>(NameType.class);
 
+    private static final Pattern QUOTE = Pattern.compile("'");
+
     static {
         NAME_PREFIXES.put(NameType.ASHKENAZI,
                 Collections.unmodifiableSet(
@@ -401,14 +404,14 @@ public String encode(String input, final Languages.LanguageSet languageSet) {
             }
         }
 
-        final List<String> words = Arrays.asList(input.split("\\s+"));
+        final List<String> words = Arrays.asList(ResourceConstants.SPACES.split(input));
         final List<String> words2 = new ArrayList<>();
 
         // special-case handling of word prefixes based upon the name type
         switch (this.nameType) {
         case SEPHARDIC:
             words.forEach(aWord -> {
-                final String[] parts = aWord.split("'", -1);
+                final String[] parts = QUOTE.split(aWord, -1);
                 words2.add(parts[parts.length - 1]);
             });
             words2.removeAll(NAME_PREFIXES.get(this.nameType));

src/main/java/org/apache/commons/codec/language/bm/ResourceConstants.java

@@ -17,6 +17,8 @@
 
 package org.apache.commons.codec.language.bm;
 
+import java.util.regex.Pattern;
+
 import org.apache.commons.codec.CharEncoding;
 
 /**
@@ -32,5 +34,7 @@ final class ResourceConstants {
     static final String ENCODING = CharEncoding.UTF_8;
     static final String EXT_CMT_END = "*/";
     static final String EXT_CMT_START = "/*";
+    static final Pattern SPACES = Pattern.compile("\\s+");
+
 
 }

src/main/java/org/apache/commons/codec/language/bm/Rule.java

@@ -300,6 +300,10 @@ public interface RPattern {
 
     private static final int HASH_INCLUDE_LENGTH = HASH_INCLUDE.length();
 
+    private static final Pattern AROUND_PLUS = Pattern.compile("[+]");
+
+    private static final Pattern AROUND_PIPE = Pattern.compile("[|]");
+
     private static final Map<NameType, Map<RuleType, Map<String, Map<String, List<Rule>>>>> RULES =
             new EnumMap<>(NameType.class);
 
@@ -452,7 +456,7 @@ private static Phoneme parsePhoneme(final String ph) {
             }
             final String before = ph.substring(0, open);
             final String in = ph.substring(open + 1, ph.length() - 1);
-            final Set<String> langs = new HashSet<>(Arrays.asList(in.split("[+]")));
+            final Set<String> langs = new HashSet<>(Arrays.asList(AROUND_PLUS.split(in)));
 
             return new Phoneme(before, Languages.LanguageSet.from(langs));
         }
@@ -467,7 +471,7 @@ private static PhonemeExpr parsePhonemeExpr(final String ph) {
 
             final List<Phoneme> phs = new ArrayList<>();
             final String body = ph.substring(1, ph.length() - 1);
-            for (final String part : body.split("[|]")) {
+            for (final String part : AROUND_PIPE.split(body)) {
                 phs.add(parsePhoneme(part));
             }
             if (body.startsWith("|") || body.endsWith("|")) {
@@ -521,7 +525,7 @@ private static Map<String, List<Rule>> parseRules(final Scanner scanner, final S
                     }
                 } else {
                     // rule
-                    final String[] parts = line.split("\\s+");
+                    final String[] parts = ResourceConstants.SPACES.split(line);
                     if (parts.length != 4) {
                         throw new IllegalArgumentException("Malformed rule statement split into " + parts.length +
                                                            " parts: " + rawLine + " in " + location);