Java Code Examples for java.util.regex.Pattern#LITERAL

private void checkRegexp(MethodContext mc, Expression regexExpr, int flags) {
    if((flags & Pattern.LITERAL) == 0) {
        if(regexExpr.getCode() == AstCode.GetStatic) {
            FieldReference fr = (FieldReference) regexExpr.getOperand();
            if(fr.getName().equals("separator") && fr.getDeclaringType().getInternalName().equals("java/io/File")) {
                mc.report("RegexFileSeparator", 0, regexExpr);
            }
        } else if(regexExpr.getCode() == AstCode.InvokeVirtual) {
            MethodReference mr = (MethodReference)regexExpr.getOperand();
            if(mr.getName().equals("getSeparator") && mr.getDeclaringType().getInternalName().equals("java/nio/file/FileSystem")) {
                mc.report("RegexFileSeparator", 0, regexExpr);
            }
        }
    }
    Object regexObj = Nodes.getConstant(regexExpr);
    if(!(regexObj instanceof String)) {
        return;
    }
    String regex = (String)regexObj;
    try {
        Pattern.compile(regex, flags);
    } catch (PatternSyntaxException e) {
        mc.report("RegexBadSyntax", 0, regexExpr, Roles.REGEXP.create(regex), ERROR_MESSAGE.create(e.getMessage()));
    }
}
 

private int getRegexFlag() {
	int flag = 0;
	if (unixLinesCkb.isSelected())
		flag |= Pattern.UNIX_LINES;
	if (caseInsensitiveCkb.isSelected())
		flag |= Pattern.CASE_INSENSITIVE;
	if (commentsCkb.isSelected())
		flag |= Pattern.COMMENTS;
	if (multilineCkb.isSelected())
		flag |= Pattern.MULTILINE;
	if (literalCkb.isSelected())
		flag |= Pattern.LITERAL;
	if (dotallCkb.isSelected())
		flag |= Pattern.DOTALL;
	if (unicodeCaseCkb.isSelected())
		flag |= Pattern.UNICODE_CASE;
	if (canonEqCkb.isSelected())
		flag |= Pattern.CANON_EQ;
	return flag;
}
 

public void init(InputStream in) throws ConfigurationException {
  propertiesConfiguration = new PropertiesConfiguration();
  propertiesConfiguration.setDelimiterParsingDisabled(true);
  propertiesConfiguration.load(in, "UTF-8");
  configuration = new DataConfiguration(propertiesConfiguration);
  configuration.setDelimiterParsingDisabled(true);
  String pa = configuration.getString(PROP_PATTERN);
  int flags = 0;
  flags = flags | (configuration.getBoolean(PROP_PATTERN_CANON_EQ, false) ? Pattern.CANON_EQ : 0);
  flags = flags | (configuration.getBoolean(PROP_PATTERN_CASE_INSENSITIVE, false) ? Pattern.CASE_INSENSITIVE : 0);
  flags = flags | (configuration.getBoolean(PROP_PATTERN_COMMENTS, false) ? Pattern.COMMENTS : 0);
  flags = flags | (configuration.getBoolean(PROP_PATTERN_DOTALL, false) ? Pattern.DOTALL : 0);
  flags = flags | (configuration.getBoolean(PROP_PATTERN_LITERAL, false) ? Pattern.LITERAL : 0);
  flags = flags | (configuration.getBoolean(PROP_PATTERN_MULTILINE, false) ? Pattern.MULTILINE : 0);
  flags = flags | (configuration.getBoolean(PROP_PATTERN_UNICODE_CASE, false) ? Pattern.UNICODE_CASE : 0);
  flags = flags | (configuration.getBoolean(PROP_PATTERN_UNIX_LINES, false) ? Pattern.UNIX_LINES : 0);

  pattern = Pattern.compile(pa, flags);
  groupCount = countGroups(pattern);
  name = configuration.getString(PROP_NAME, "NAME NOT SET!");
  description = configuration.getString(PROP_DESCRIPTION, "DESCRIPTION NOT SET!");
  testMessage = configuration.getString(PROP_TEST_MESSAGE, "");

}
 

/**
 * update the finder object with data from our UI
 */
private void updateFinder() {
    int flag = 0;
    if (!jChkRegex.isSelected()) {
        flag |= Pattern.LITERAL;
    }
    flag |= (jChkIgnoreCase.isSelected()) ? Pattern.CASE_INSENSITIVE : 0;
    if (jChkIgnoreCase.isSelected()) {
        flag |= Pattern.CASE_INSENSITIVE;
    }
    String regex = (String) jCmbFind.getSelectedItem();
    if (regex != null && regex.length() > 0) {
        Pattern pattern = Pattern.compile(regex, flag);
        finder.setWrap(jChkWrap.isSelected());
        finder.setPattern(pattern);
        ActionUtils.insertIntoCombo(jCmbFind, regex);
    } else {
        finder.setPattern(null);
    }
}
 

private static int parseOptionFlags(String opts) {
    int flags = 0;
    for(int i = 0; i < opts.length(); ++i) {
        switch(opts.charAt(i)) {
            // Standard 'special construct match flags'
            case 'i': flags |= Pattern.CASE_INSENSITIVE; break;
            case 'd': flags |= Pattern.UNIX_LINES; break;
            case 'm': flags |= Pattern.MULTILINE; break;
            case 's': flags |= Pattern.DOTALL; break;
            case 'u': flags |= Pattern.UNICODE_CASE; break;
            case 'x': flags |= Pattern.COMMENTS; break;
            // And pick a letters for remaining flags
            case 'l': flags |= Pattern.LITERAL; break;
            case 'c': flags |= Pattern.CANON_EQ; break;
            default:
                throw new InvalidParameterValueException("Invalid option: " + opts.charAt(i));
        }
    }
    return flags;
}
 

private void sawRegExPattern(int stackDepth, int flags) {
    if (stack.getStackDepth() < stackDepth) {
        return;
    }
    OpcodeStack.Item it = stack.getStackItem(stackDepth);
    if (it.getSpecialKind() == OpcodeStack.Item.FILE_SEPARATOR_STRING && (flags & Pattern.LITERAL) == 0) {
        bugReporter.reportBug(new BugInstance(this, "RE_CANT_USE_FILE_SEPARATOR_AS_REGULAR_EXPRESSION", HIGH_PRIORITY)
                .addClassAndMethod(this).addCalledMethod(this).addSourceLine(this));
        return;
    }
    Object value = it.getConstant();
    if (!(value instanceof String)) {
        return;
    }
    String regex = (String) value;
    try {
        Pattern.compile(regex, flags);
    } catch (IllegalArgumentException e) {
        String message = e.getMessage();
        int eol = message.indexOf('\n');
        if (eol > 0) {
            message = message.substring(0, eol);
        }
        BugInstance bug = new BugInstance(this, "RE_BAD_SYNTAX_FOR_REGULAR_EXPRESSION", HIGH_PRIORITY)
                .addClassAndMethod(this).addCalledMethod(this).addString(message).describe(StringAnnotation.ERROR_MSG_ROLE)
                .addString(regex).describe(StringAnnotation.REGEX_ROLE);
        String options = getOptions(flags);
        if (options.length() > 0) {
            bug.addString("Regex flags: " + options).describe(StringAnnotation.STRING_MESSAGE);
        }
        bug.addSourceLine(this);
        bugReporter.reportBug(bug);
    }
}
 

public static int flagsFromString(String flags) {
    int pFlags = 0;
    for (String s : Strings.delimitedListToStringArray(flags, "|")) {
        if (s.isEmpty()) {
            continue;
        }
        s = s.toUpperCase(Locale.ROOT);
        if ("CASE_INSENSITIVE".equals(s)) {
            pFlags |= Pattern.CASE_INSENSITIVE;
        } else if ("MULTILINE".equals(s)) {
            pFlags |= Pattern.MULTILINE;
        } else if ("DOTALL".equals(s)) {
            pFlags |= Pattern.DOTALL;
        } else if ("UNICODE_CASE".equals(s)) {
            pFlags |= Pattern.UNICODE_CASE;
        } else if ("CANON_EQ".equals(s)) {
            pFlags |= Pattern.CANON_EQ;
        } else if ("UNIX_LINES".equals(s)) {
            pFlags |= Pattern.UNIX_LINES;
        } else if ("LITERAL".equals(s)) {
            pFlags |= Pattern.LITERAL;
        } else if ("COMMENTS".equals(s)) {
            pFlags |= Pattern.COMMENTS;
        } else if (("UNICODE_CHAR_CLASS".equals(s)) || ("UNICODE_CHARACTER_CLASS".equals(s))) {
            pFlags |= UNICODE_CHARACTER_CLASS;
        } else {
            throw new IllegalArgumentException("Unknown regex flag [" + s + "]");
        }
    }
    return pFlags;
}
 

@VisibleForTesting
static Pattern buildPattern( boolean literalParsing, boolean caseSensitive, boolean wholeWord,
  String patternString, boolean isUnicode ) {
  int flags = 0;
  if ( literalParsing && !wholeWord ) {
    flags |= Pattern.LITERAL;
  }
  if ( !caseSensitive ) {
    flags |= Pattern.CASE_INSENSITIVE;
  }
  if ( isUnicode ) {
    flags |= Pattern.UNICODE_CHARACTER_CLASS;
  }

  /*
   * XXX: I don't like this parameter. I think it would almost always be better for the user to define either word
   * boundaries or ^/$ anchors explicitly in their pattern.
   */
  if ( wholeWord ) {
    if ( literalParsing ) {
      patternString = "\\Q" + patternString + "\\E";
    }
    patternString = "\\b" + patternString + "\\b";
  }

  return Pattern.compile( patternString, flags );
}
 

int getCompileFlags(ProcessContext context) {
    int flags = (context.getProperty(UNIX_LINES).asBoolean() ? Pattern.UNIX_LINES : 0)
            | (context.getProperty(CASE_INSENSITIVE).asBoolean() ? Pattern.CASE_INSENSITIVE : 0)
            | (context.getProperty(COMMENTS).asBoolean() ? Pattern.COMMENTS : 0)
            | (context.getProperty(MULTILINE).asBoolean() ? Pattern.MULTILINE : 0)
            | (context.getProperty(LITERAL).asBoolean() ? Pattern.LITERAL : 0)
            | (context.getProperty(DOTALL).asBoolean() ? Pattern.DOTALL : 0)
            | (context.getProperty(UNICODE_CASE).asBoolean() ? Pattern.UNICODE_CASE : 0)
            | (context.getProperty(CANON_EQ).asBoolean() ? Pattern.CANON_EQ : 0)
            | (context.getProperty(UNICODE_CHARACTER_CLASS).asBoolean() ? Pattern.UNICODE_CHARACTER_CLASS : 0);
    return flags;
}
 

/**
 * Sets the pattern from a string and flags
 * @param pat String of pattern
 * @param regex true if the pattern should be a regexp
 * @param ignoreCase true to ignore case
 * @throws java.util.regex.PatternSyntaxException
 */
public void setPattern(String pat, boolean regex, boolean ignoreCase)
	throws PatternSyntaxException {
	if (pat != null && pat.length() > 0) {
		int flag = (regex) ? 0 : Pattern.LITERAL;
		flag |= (ignoreCase) ? Pattern.CASE_INSENSITIVE : 0;
		setPattern(Pattern.compile(pat, flag));
	} else {
		setPattern(null);
	}
}
 

@VisibleForTesting
static Pattern buildPattern( boolean literalParsing, boolean caseSensitive, boolean wholeWord,
                             String patternString, boolean isUnicode ) {
  int flags = 0;
  if ( literalParsing && !wholeWord ) {
    flags |= Pattern.LITERAL;
  }
  if ( !caseSensitive ) {
    flags |= Pattern.CASE_INSENSITIVE;
  }
  if ( isUnicode ) {
    flags |= Pattern.UNICODE_CHARACTER_CLASS;
  }

  /*
   * XXX: I don't like this parameter. I think it would almost always be better for the user to define either word
   * boundaries or ^/$ anchors explicitly in their pattern.
   */
  if ( wholeWord ) {
    if ( literalParsing ) {
      patternString = "\\Q" + patternString + "\\E";
    }
    patternString = "\\b" + patternString + "\\b";
  }

  return Pattern.compile( patternString, flags );
}
 

@Test
public void testRegexpFlagParsing() {
    ModifiableSolrParams params = new ModifiableSolrParams();
    params.add(TermsParams.TERMS_REGEXP_FLAG, "case_insensitive", "literal", "comments", "multiline", "unix_lines",
            "unicode_case", "dotall", "canon_eq");
    try (TermsComponent termsComponent = new TermsComponent()) {
      int flags = termsComponent.resolveRegexpFlags(params);
      int expected = Pattern.CASE_INSENSITIVE | Pattern.LITERAL | Pattern.COMMENTS | Pattern.MULTILINE | Pattern.UNIX_LINES
          | Pattern.UNICODE_CASE | Pattern.DOTALL | Pattern.CANON_EQ;
      assertEquals(expected, flags);
    } catch (IOException e) {
      fail("Error closing TermsComponent");
    }
}
 

public static String flagsToString(int flags) {
    StringBuilder sb = new StringBuilder();
    if ((flags & Pattern.CASE_INSENSITIVE) != 0) {
        sb.append("CASE_INSENSITIVE|");
    }
    if ((flags & Pattern.MULTILINE) != 0) {
        sb.append("MULTILINE|");
    }
    if ((flags & Pattern.DOTALL) != 0) {
        sb.append("DOTALL|");
    }
    if ((flags & Pattern.UNICODE_CASE) != 0) {
        sb.append("UNICODE_CASE|");
    }
    if ((flags & Pattern.CANON_EQ) != 0) {
        sb.append("CANON_EQ|");
    }
    if ((flags & Pattern.UNIX_LINES) != 0) {
        sb.append("UNIX_LINES|");
    }
    if ((flags & Pattern.LITERAL) != 0) {
        sb.append("LITERAL|");
    }
    if ((flags & Pattern.COMMENTS) != 0) {
        sb.append("COMMENTS|");
    }
    if ((flags & UNICODE_CHARACTER_CLASS) != 0) {
        sb.append("UNICODE_CHAR_CLASS|");
    }
    return sb.toString();
}
 

public static int flagsFromString(String flags) {
    int pFlags = 0;
    for (String s : Strings.delimitedListToStringArray(flags, "|")) {
        if (s.isEmpty()) {
            continue;
        }
        s = s.toUpperCase(Locale.ROOT);
        if ("CASE_INSENSITIVE".equals(s)) {
            pFlags |= Pattern.CASE_INSENSITIVE;
        } else if ("MULTILINE".equals(s)) {
            pFlags |= Pattern.MULTILINE;
        } else if ("DOTALL".equals(s)) {
            pFlags |= Pattern.DOTALL;
        } else if ("UNICODE_CASE".equals(s)) {
            pFlags |= Pattern.UNICODE_CASE;
        } else if ("CANON_EQ".equals(s)) {
            pFlags |= Pattern.CANON_EQ;
        } else if ("UNIX_LINES".equals(s)) {
            pFlags |= Pattern.UNIX_LINES;
        } else if ("LITERAL".equals(s)) {
            pFlags |= Pattern.LITERAL;
        } else if ("COMMENTS".equals(s)) {
            pFlags |= Pattern.COMMENTS;
        } else if ("UNICODE_CHAR_CLASS".equals(s)) {
            pFlags |= UNICODE_CHARACTER_CLASS;
        } else {
            throw new IllegalArgumentException("Unknown regex flag [" + s + "]");
        }
    }
    return pFlags;
}
 

int getCompileFlags(ProcessContext context) {
    int flags = (context.getProperty(UNIX_LINES).asBoolean() ? Pattern.UNIX_LINES : 0)
            | (context.getProperty(CASE_INSENSITIVE).asBoolean() ? Pattern.CASE_INSENSITIVE : 0)
            | (context.getProperty(COMMENTS).asBoolean() ? Pattern.COMMENTS : 0)
            | (context.getProperty(MULTILINE).asBoolean() ? Pattern.MULTILINE : 0)
            | (context.getProperty(LITERAL).asBoolean() ? Pattern.LITERAL : 0)
            | (context.getProperty(DOTALL).asBoolean() ? Pattern.DOTALL : 0)
            | (context.getProperty(UNICODE_CASE).asBoolean() ? Pattern.UNICODE_CASE : 0)
            | (context.getProperty(CANON_EQ).asBoolean() ? Pattern.CANON_EQ : 0)
            | (context.getProperty(UNICODE_CHARACTER_CLASS).asBoolean() ? Pattern.UNICODE_CHARACTER_CLASS : 0);
    return flags;
}
 

private static Pattern getPattern(final Value parg, final Value farg)
            throws IllegalArgumentException {
        
        if (debug) {
            log.debug("regex pattern: " + parg);
            log.debug("regex flags: " + farg);
        }
        
        //BLZG-1200 Literals with language types are not included in REGEX
        if (QueryEvaluationUtil.isPlainLiteral(parg)
                && (farg == null || QueryEvaluationUtil.isPlainLiteral(farg))) {

            final String ptn = ((Literal) parg).getLabel();
            String flags = "";
            if (farg != null) {
                flags = ((Literal)farg).getLabel();
            }
            int f = 0;
            for (char c : flags.toCharArray()) {
                // See https://www.w3.org/TR/xpath-functions/#flags
                switch (c) {
                    case 's':
                        f |= Pattern.DOTALL;
                        break;
                    case 'm':
                        f |= Pattern.MULTILINE;
                        break;
                    case 'i': {
                    /*
                     * The SPARQL REGEX operator is based on the XQuery REGEX
                     * operator. That operator should be Unicode clean by
                     * default. Therefore, when case-folding is specified, we
                     * also need to include the UNICODE_CASE option.
                     * 
                     * @see <a
                     * href="https://sourceforge.net/apps/trac/bigdata/ticket/655"
                     * > SPARQL REGEX operator does not perform case-folding
                     * correctly for Unicode data </a>
                     */
                        f |= Pattern.CASE_INSENSITIVE;
                        f |= Pattern.UNICODE_CASE;
                        break;
                    }
                    case 'x':
                        f |= Pattern.COMMENTS;
                        break;
                    case 'd':
                        f |= Pattern.UNIX_LINES;
                        break;
                    case 'u': // Implicit with 'i' flag.
//                      f |= Pattern.UNICODE_CASE;
                        break;
                    case 'q':
                        f |= Pattern.LITERAL;
                        break;
                    default:
                        throw new IllegalArgumentException();
                }
            }
            final Pattern pattern = Pattern.compile(ptn, f);
            return pattern;
        }
        
        throw new IllegalArgumentException();
        
    }
 

private static Pattern getPattern(final Value pattern, final Value flags) 
		throws IllegalArgumentException {
	
	if (!QueryEvaluationUtil.isSimpleLiteral(pattern)) {
		throw new IllegalArgumentException(
				"incompatible operand for REPLACE: " + pattern);
	}

	String flagString = null;
	if (flags != null) {
		if (!QueryEvaluationUtil.isSimpleLiteral(flags)) {
			throw new IllegalArgumentException(
					"incompatible operand for REPLACE: " + flags);
		}
		flagString = ((Literal) flags).getLabel();
	}

	String patternString = ((Literal) pattern).getLabel();

	int f = 0;
	if (flagString != null) {
		for (char c : flagString.toCharArray()) {
               // See https://www.w3.org/TR/xpath-functions/#flags
			switch (c) {
			case 's':
				f |= Pattern.DOTALL;
				break;
			case 'm':
				f |= Pattern.MULTILINE;
				break;
			case 'i':
				f |= Pattern.CASE_INSENSITIVE;
				break;
			case 'x':
				f |= Pattern.COMMENTS;
				break;
			case 'd':
				f |= Pattern.UNIX_LINES;
				break;
			case 'u':
				f |= Pattern.UNICODE_CASE;
				break;
			case 'q':
				f |= Pattern.LITERAL;
				break;
			default:
				throw new IllegalArgumentException(flagString);
			}
		}
	}

	Pattern p = Pattern.compile(patternString, f);
	
	return p;
	
}
 

/**
 * Determines whether the two operands match according to the <code>regex</code> operator.
 *
 * @return <tt>true</tt> if the operands match according to the <tt>regex</tt> operator, <tt>false</tt> otherwise.
 */
public Value evaluate(Regex node, BindingSet bindings)
		throws QueryEvaluationException {
	Value arg = evaluate(node.getArg(), bindings);
	Value parg = evaluate(node.getPatternArg(), bindings);
	Value farg = null;
	ValueExpr flagsArg = node.getFlagsArg();
	if (flagsArg != null) {
		farg = evaluate(flagsArg, bindings);
	}

	if (QueryEvaluationUtil.isStringLiteral(arg) && QueryEvaluationUtil.isSimpleLiteral(parg)
			&& (farg == null || QueryEvaluationUtil.isSimpleLiteral(farg))) {
		String text = ((Literal) arg).getLabel();
		String ptn = ((Literal) parg).getLabel();
		String flags = "";
		if (farg != null) {
			flags = ((Literal) farg).getLabel();
		}
		// TODO should this Pattern be cached?
		int f = 0;
		for (char c : flags.toCharArray()) {
			switch (c) {
			case 's':
				f |= Pattern.DOTALL;
				break;
			case 'm':
				f |= Pattern.MULTILINE;
				break;
			case 'i':
				f |= Pattern.CASE_INSENSITIVE;
				f |= Pattern.UNICODE_CASE;
				break;
			case 'x':
				f |= Pattern.COMMENTS;
				break;
			case 'd':
				f |= Pattern.UNIX_LINES;
				break;
			case 'u':
				f |= Pattern.UNICODE_CASE;
				break;
			case 'q':
				f |= Pattern.LITERAL;
				break;
			default:
				throw new ValueExprEvaluationException(flags);
			}
		}
		Pattern pattern = Pattern.compile(ptn, f);
		boolean result = pattern.matcher(text).find();
		return BooleanLiteral.valueOf(result);
	}

	throw new ValueExprEvaluationException();
}
 

/**
 * Setup all patterns for text and column text filters
 */
public void update() {
   parentMatches.clear();
   // Update text filter pattern
   if (!Strings.isValid(xViewer.getCustomizeMgr().getFilterText())) {
      textPattern = null;
   } else {
      int flags = Pattern.CASE_INSENSITIVE;
      if (!xViewer.getCustomizeMgr().isFilterTextRegularExpression()) {
         flags = Pattern.LITERAL | flags;
      }
      textPattern = Pattern.compile(xViewer.getCustomizeMgr().getFilterText(), flags);
   }
   // Update column filter patterns
   colIdToPattern.clear();
   colIdToDateFilter.clear();
   for (String colId : xViewer.getCustomizeMgr().getColumnFilterData().getColIds()) {
      String colFilterText = xViewer.getCustomizeMgr().getColumnFilterText(colId);
      if (colFilterText != null) {
         boolean isWrapped = (colFilterText.matches("^\\(.*\\)$"));
         boolean isNot;
         if (isWrapped) {
            colFilterText = colFilterText.substring(1, colFilterText.length() - 1);
         }
         isNot = colFilterText.startsWith("!");
         if (isNot) {
            colFilterText = colFilterText.replaceFirst("^!", "");
         }
         colFilterText = Pattern.quote(colFilterText);
         // Handle != case  ^(.(?<!big))*$
         if (isNot) {
            if (colFilterText.equals("")) {
               colIdToPattern.put(colId, NOT_EMPTY_STR_PATTERN);
            } else {
               colIdToPattern.put(colId,
                  Pattern.compile("^(.(?<!" + colFilterText + "))*$", Pattern.CASE_INSENSITIVE));
            }
         }
         // Handle normal case
         else {
            if (colFilterText.equals("")) {
               colIdToPattern.put(colId, EMPTY_STR_PATTERN);
            } else {
               colIdToPattern.put(colId, Pattern.compile(colFilterText, Pattern.CASE_INSENSITIVE));
            }
         }
      }
      ColumnDateFilter dateFilter = xViewer.getCustomizeMgr().getColumnDateFilter(colId);
      if (dateFilter != null) {
         colIdToDateFilter.put(colId, dateFilter);
      }
   }
}
 

/**
 * Find the location of the given String in the document.  returns -1
 * if the search string is not found starting at position <code>start</code>
 * @param search The String to search for
 * @param start The beginning index of search
 * @return
 * @deprecated use {@link getMatcher} instead
 */
@Deprecated
public int getIndexOf(String search, int start) {
    int flag = Pattern.LITERAL;
    Pattern pattern = Pattern.compile(search, flag);
    return getIndexOf(pattern, start);
}