/*
**********************************************************************
-* Copyright (C) 2002-2010, International Business Machines
+* Copyright (C) 2002-2015, International Business Machines
* Corporation and others. All Rights Reserved.
**********************************************************************
* file name: regex.h
#include "unicode/uregex.h"
-U_NAMESPACE_BEGIN
+// Forward Declarations
+struct UHashtable;
-// Forward Declarations...
+U_NAMESPACE_BEGIN
-class RegexMatcher;
-class RegexPattern;
-class UVector;
-class UVector32;
-class UVector64;
-class UnicodeSet;
-struct REStackFrame;
struct Regex8BitSet;
-class RuleBasedBreakIterator;
class RegexCImpl;
-
-
-
-
-/**
- * RBBIPatternDump Debug function, displays the compiled form of a pattern.
- * @internal
- */
-#ifdef REGEX_DEBUG
-U_INTERNAL void U_EXPORT2
- RegexPatternDump(const RegexPattern *pat);
-#else
- #undef RegexPatternDump
- #define RegexPatternDump(pat)
-#endif
-
+class RegexMatcher;
+class RegexPattern;
+struct REStackFrame;
+class RuleBasedBreakIterator;
+class UnicodeSet;
+class UVector;
+class UVector32;
+class UVector64;
/**
*
* @stable ICU 2.4
*/
-class U_I18N_API RegexPattern: public UObject {
+class U_I18N_API RegexPattern U_FINAL : public UObject {
public:
/**
* @return TRUE if the objects are different.
* @stable ICU 2.4
*/
- inline UBool operator!=(const RegexPattern& that) const {return ! operator ==(that);};
+ inline UBool operator!=(const RegexPattern& that) const {return ! operator ==(that);}
/**
* Assignment operator. After assignment, this RegexPattern will behave identically
/**
* Create an exact copy of this RegexPattern object. Since RegexPattern is not
- * intended to be subclasses, <code>clone()</code> and the copy construction are
+ * intended to be subclassed, <code>clone()</code> and the copy construction are
* equivalent operations.
* @return the copy of this RegexPattern
* @stable ICU 2.4
UParseError &pe,
UErrorCode &status);
-
/**
* Compiles the regular expression in string form into a RegexPattern
* object. These compile methods, rather than the constructors, are the usual
* @param status A reference to a UErrorCode to receive any errors.
* @return A regexPattern object for the compiled pattern.
*
- * @draft ICU 4.6
+ * @stable ICU 4.6
*/
static RegexPattern * U_EXPORT2 compile( UText *regex,
UParseError &pe,
uint32_t flags,
UParseError &pe,
UErrorCode &status);
-
-
+
/**
* Compiles the regular expression in string form into a RegexPattern
* object using the specified match mode flags. These compile methods,
* @param status A reference to a UErrorCode to receive any errors.
* @return A regexPattern object for the compiled pattern.
*
- * @draft ICU 4.6
+ * @stable ICU 4.6
*/
static RegexPattern * U_EXPORT2 compile( UText *regex,
uint32_t flags,
UParseError &pe,
UErrorCode &status);
-
/**
* Compiles the regular expression in string form into a RegexPattern
uint32_t flags,
UErrorCode &status);
-
/**
* Compiles the regular expression in string form into a RegexPattern
* object using the specified match mode flags. These compile methods,
* @param status A reference to a UErrorCode to receive any errors.
* @return A regexPattern object for the compiled pattern.
*
- * @draft ICU 4.6
+ * @stable ICU 4.6
*/
static RegexPattern * U_EXPORT2 compile( UText *regex,
uint32_t flags,
UErrorCode &status);
-
/**
* Get the match mode flags that were used when compiling this pattern.
virtual RegexMatcher *matcher(const UnicodeString &input,
UErrorCode &status) const;
-
- /**
- * Flag to disambiguate RegexPattern::matcher signature
- * @draft ICU 4.6
- */
- enum PatternIsUTextFlag { PATTERN_IS_UTEXT };
-
- /**
- * Creates a RegexMatcher that will match the given input against this pattern. The
- * RegexMatcher can then be used to perform match, find or replace operations
- * on the input. Note that a RegexPattern object must not be deleted while
- * RegexMatchers created from it still exist and might possibly be used again.
- * <p>
- * The matcher will make a shallow clone of the supplied input text, and all regexp
- * pattern matching operations happen on this clone. While read-only operations on
- * the supplied text are permitted, it is critical that the underlying string not be
- * altered or deleted before use by the regular expression operations is complete.
- *
- * @param input The input text to which the regular expression will be applied.
- * @param flag Must be RegexPattern::PATTERN_IS_UTEXT; used to disambiguate
- * method signature.
- * @param status A reference to a UErrorCode to receive any errors.
- * @return A RegexMatcher object for this pattern and input.
- *
- * @draft ICU 4.6
- */
- virtual RegexMatcher *matcher(UText *input,
- PatternIsUTextFlag flag,
- UErrorCode &status) const;
-
private:
/**
- * Cause a compilation error if an application accidently attempts to
+ * Cause a compilation error if an application accidentally attempts to
* create a matcher with a (UChar *) string as input rather than
* a UnicodeString. Avoids a dangling reference to a temporary string.
* <p>
* or in a UText, using
* <code>utext_openUChars(UText *ut, const UChar *text, int64_t textLength, UErrorCode *status);</code>
*
- * @internal
*/
RegexMatcher *matcher(const UChar *input,
UErrorCode &status) const;
/**
* Test whether a string matches a regular expression. This convenience function
- * both compiles the reguluar expression and applies it in a single operation.
+ * both compiles the regular expression and applies it in a single operation.
* Note that if the same pattern needs to be applied repeatedly, this method will be
* less efficient than creating and reusing a RegexMatcher object.
*
UParseError &pe,
UErrorCode &status);
-
/**
* Test whether a string matches a regular expression. This convenience function
- * both compiles the reguluar expression and applies it in a single operation.
+ * both compiles the regular expression and applies it in a single operation.
* Note that if the same pattern needs to be applied repeatedly, this method will be
* less efficient than creating and reusing a RegexMatcher object.
*
* @param status A reference to a UErrorCode to receive any errors.
* @return True if the regular expression exactly matches the full input string.
*
- * @draft ICU 4.6
+ * @stable ICU 4.6
*/
static UBool U_EXPORT2 matches(UText *regex,
UText *input,
UParseError &pe,
UErrorCode &status);
-
/**
* Returns the regular expression from which this pattern was compiled. This method will work
* even if the pattern was compiled from a UText.
* UText, and that UText was modified, the returned UText may no longer reflect the RegexPattern
* object.
*
- * @draft ICU 4.6
+ * @stable ICU 4.6
*/
virtual UText *patternText(UErrorCode &status) const;
/**
- * Split a string into fields. Somewhat like split() from Perl.
- * The pattern matches identify delimiters that separate the input
- * into fields. The input data between the matches becomes the
- * fields themselves.
- * <p>
- * For the best performance on split() operations,
- * <code>RegexMatcher::split</code> is perferable to this function
+ * Get the group number corresponding to a named capture group.
+ * The returned number can be used with any function that access
+ * capture groups by number.
+ *
+ * The function returns an error status if the specified name does not
+ * appear in the pattern.
+ *
+ * @param groupName The capture group name.
+ * @param status A UErrorCode to receive any errors.
+ *
+ * @draft ICU 55
+ */
+ virtual int32_t groupNumberFromName(const UnicodeString &groupName, UErrorCode &status) const;
+
+
+ /**
+ * Get the group number corresponding to a named capture group.
+ * The returned number can be used with any function that access
+ * capture groups by number.
+ *
+ * The function returns an error status if the specified name does not
+ * appear in the pattern.
+ *
+ * @param groupName The capture group name,
+ * platform invariant characters only.
+ * @param nameLength The length of the name, or -1 if the name is
+ * nul-terminated.
+ * @param status A UErrorCode to receive any errors.
+ *
+ * @draft ICU 55
+ */
+ virtual int32_t groupNumberFromName(const char *groupName, int32_t nameLength, UErrorCode &status) const;
+
+
+ /**
+ * Split a string into fields. Somewhat like split() from Perl or Java.
+ * Pattern matches identify delimiters that separate the input
+ * into fields. The input data between the delimiters becomes the
+ * fields themselves.
+ *
+ * If the delimiter pattern includes capture groups, the captured text will
+ * also appear in the destination array of output strings, interspersed
+ * with the fields. This is similar to Perl, but differs from Java,
+ * which ignores the presence of capture groups in the pattern.
+ *
+ * Trailing empty fields will always be returned, assuming sufficient
+ * destination capacity. This differs from the default behavior for Java
+ * and Perl where trailing empty fields are not returned.
+ *
+ * The number of strings produced by the split operation is returned.
+ * This count includes the strings from capture groups in the delimiter pattern.
+ * This behavior differs from Java, which ignores capture groups.
+ *
+ * For the best performance on split() operations,
+ * <code>RegexMatcher::split</code> is preferable to this function
*
* @param input The string to be split into fields. The field delimiters
* match the pattern (in the "this" object)
/**
- * Split a string into fields. Somewhat like split() from Perl.
- * The pattern matches identify delimiters that separate the input
- * into fields. The input data between the matches becomes the
- * fields themselves.
- * <p>
+ * Split a string into fields. Somewhat like split() from Perl or Java.
+ * Pattern matches identify delimiters that separate the input
+ * into fields. The input data between the delimiters becomes the
+ * fields themselves.
+ *
+ * If the delimiter pattern includes capture groups, the captured text will
+ * also appear in the destination array of output strings, interspersed
+ * with the fields. This is similar to Perl, but differs from Java,
+ * which ignores the presence of capture groups in the pattern.
+ *
+ * Trailing empty fields will always be returned, assuming sufficient
+ * destination capacity. This differs from the default behavior for Java
+ * and Perl where trailing empty fields are not returned.
+ *
+ * The number of strings produced by the split operation is returned.
+ * This count includes the strings from capture groups in the delimiter pattern.
+ * This behavior differs from Java, which ignores capture groups.
+ *
* For the best performance on split() operations,
- * <code>RegexMatcher::split</code> is perferable to this function
+ * <code>RegexMatcher::split</code> is preferable to this function
*
* @param input The string to be split into fields. The field delimiters
* match the pattern (in the "this" object)
* of fields, the trailing part of the input string, including any
* field delimiters, is placed in the last destination string.
* @param status A reference to a UErrorCode to receive any errors.
- * @return The number of fields into which the input string was split.
+ * @return The number of destination strings used.
*
- * @draft ICU 4.6
+ * @stable ICU 4.6
*/
virtual int32_t split(UText *input,
UText *dest[],
UVector32 *fGroupMap; // Map from capture group number to position of
// the group's variables in the matcher stack frame.
- int32_t fMaxCaptureDigits;
-
UnicodeSet **fStaticSets; // Ptr to static (shared) sets for predefined
// regex character classes, e.g. Word.
Regex8BitSet *fInitialChars8;
UBool fNeedsAltInput;
+ UHashtable *fNamedCaptureMap; // Map from capture group names to numbers.
+
friend class RegexCompile;
friend class RegexMatcher;
friend class RegexCImpl;
//
void init(); // Common initialization, for use by constructors.
void zap(); // Common cleanup
-#ifdef REGEX_DEBUG
+
void dumpOp(int32_t index) const;
- friend void U_EXPORT2 RegexPatternDump(const RegexPattern *);
-#endif
+ public:
+#ifndef U_HIDE_INTERNAL_API
+ /**
+ * Dump a compiled pattern. Internal debug function.
+ * @internal
+ */
+ void dumpPattern() const;
+#endif /* U_HIDE_INTERNAL_API */
};
/**
- * class RegexMatcher bundles together a reular expression pattern and
+ * class RegexMatcher bundles together a regular expression pattern and
* input text to which the expression can be applied. It includes methods
* for testing for matches, and for find and replace operations.
*
*
* @stable ICU 2.4
*/
-class U_I18N_API RegexMatcher: public UObject {
+class U_I18N_API RegexMatcher U_FINAL : public UObject {
public:
/**
* @see UREGEX_CASE_INSENSITIVE
* @param status Any errors are reported by setting this UErrorCode variable.
*
- * @draft ICU 4.6
+ * @stable ICU 4.6
*/
RegexMatcher(UText *regexp, uint32_t flags, UErrorCode &status);
-
+
/**
* Construct a RegexMatcher for a regular expression.
* This is a convenience method that avoids the need to explicitly create
* @see UREGEX_CASE_INSENSITIVE
* @param status Any errors are reported by setting this UErrorCode variable.
*
- * @draft ICU 4.6
+ * @stable ICU 4.6
*/
RegexMatcher(UText *regexp, UText *input,
uint32_t flags, UErrorCode &status);
private:
/**
- * Cause a compilation error if an application accidently attempts to
+ * Cause a compilation error if an application accidentally attempts to
* create a matcher with a (UChar *) string as input rather than
* a UnicodeString. Avoids a dangling reference to a temporary string.
* <p>
* or in a UText, using
* <code>utext_openUChars(UText *ut, const UChar *text, int64_t textLength, UErrorCode *status);</code>
*
- * @internal
*/
RegexMatcher(const UnicodeString ®exp, const UChar *input,
uint32_t flags, UErrorCode &status);
virtual UBool find();
+ /**
+ * Find the next pattern match in the input string.
+ * The find begins searching the input at the location following the end of
+ * the previous match, or at the start of the string if there is no previous match.
+ * If a match is found, <code>start(), end()</code> and <code>group()</code>
+ * will provide more information regarding the match.
+ * <p>Note that if the input string is changed by the application,
+ * use find(startPos, status) instead of find(), because the saved starting
+ * position may not be valid with the altered input string.</p>
+ * @param status A reference to a UErrorCode to receive any errors.
+ * @return TRUE if a match is found.
+ * @draft ICU 55
+ */
+ virtual UBool find(UErrorCode &status);
+
/**
* Resets this RegexMatcher and then attempts to find the next substring of the
* input string that matches the pattern, starting at the specified index.
*/
virtual UnicodeString group(int32_t groupNum, UErrorCode &status) const;
-
/**
* Returns the number of capturing groups in this matcher's pattern.
* @return the number of capture groups
/**
* Returns a shallow clone of the entire live input string with the UText current native index
* set to the beginning of the requested group.
- * Note that copying the entire input string may cause significant performance and memory issues.
- * @param dest The UText into which the input should be copied, or NULL to create a new UText
+ *
+ * @param dest The UText into which the input should be cloned, or NULL to create a new UText
* @param group_len A reference to receive the length of the desired capture group
* @param status A reference to a UErrorCode to receive any errors.
* Possible errors are U_REGEX_INVALID_STATE if no match
* U_INDEX_OUTOFBOUNDS_ERROR for a bad capture group number.
* @return dest if non-NULL, a shallow copy of the input text otherwise
*
- * @draft ICU 4.6
+ * @stable ICU 4.6
*/
virtual UText *group(UText *dest, int64_t &group_len, UErrorCode &status) const;
/**
- * @draft ICU 4.6
- */
- virtual UText *group(int32_t groupNum, UText *dest, int64_t &group_len, UErrorCode &status) const;
-
- /**
- * Returns a string containing the text captured by the given group
- * during the previous match operation. Group(0) is the entire match.
+ * Returns a shallow clone of the entire live input string with the UText current native index
+ * set to the beginning of the requested group.
*
- * @param groupNum the capture group number
- * @param dest A mutable UText in which the matching text is placed.
- * If NULL, a new UText will be created (which may not be mutable).
+ * @param groupNum The capture group number.
+ * @param dest The UText into which the input should be cloned, or NULL to create a new UText.
+ * @param group_len A reference to receive the length of the desired capture group
* @param status A reference to a UErrorCode to receive any errors.
* Possible errors are U_REGEX_INVALID_STATE if no match
- * has been attempted or the last match failed.
- * @return A string containing the matched input text. If a pre-allocated UText
- * was provided, it will always be used and returned.
+ * has been attempted or the last match failed and
+ * U_INDEX_OUTOFBOUNDS_ERROR for a bad capture group number.
+ * @return dest if non-NULL, a shallow copy of the input text otherwise
*
- * @internal ICU 4.4 technology preview
+ * @stable ICU 4.6
*/
- virtual UText *group(int32_t groupNum, UText *dest, UErrorCode &status) const;
-
+ virtual UText *group(int32_t groupNum, UText *dest, int64_t &group_len, UErrorCode &status) const;
/**
* Returns the index in the input string of the start of the text matched
virtual int32_t start(UErrorCode &status) const;
/**
- * @draft ICU 4.6
+ * Returns the index in the input string of the start of the text matched
+ * during the previous match operation.
+ * @param status a reference to a UErrorCode to receive any errors.
+ * @return The (native) position in the input string of the start of the last match.
+ * @stable ICU 4.6
*/
virtual int64_t start64(UErrorCode &status) const;
virtual int32_t start(int32_t group, UErrorCode &status) const;
/**
- * @draft ICU 4.6
+ * Returns the index in the input string of the start of the text matched by the
+ * specified capture group during the previous match operation. Return -1 if
+ * the capture group exists in the pattern, but was not part of the last match.
+ *
+ * @param group the capture group number.
+ * @param status A reference to a UErrorCode to receive any errors. Possible
+ * errors are U_REGEX_INVALID_STATE if no match has been
+ * attempted or the last match failed, and
+ * U_INDEX_OUTOFBOUNDS_ERROR for a bad capture group number.
+ * @return the (native) start position of substring matched by the specified group.
+ * @stable ICU 4.6
*/
virtual int64_t start64(int32_t group, UErrorCode &status) const;
-
/**
* Returns the index in the input string of the first character following the
* text matched during the previous match operation.
+ *
* @param status A reference to a UErrorCode to receive any errors. Possible
* errors are U_REGEX_INVALID_STATE if no match has been
* attempted or the last match failed.
* @return the index of the last character matched, plus one.
* The index value returned is a native index, corresponding to
* code units for the underlying encoding type, for example,
- * a byte index for UTF8.
+ * a byte index for UTF-8.
* @stable ICU 2.4
*/
virtual int32_t end(UErrorCode &status) const;
/**
- * @draft ICU 4.6
+ * Returns the index in the input string of the first character following the
+ * text matched during the previous match operation.
+ *
+ * @param status A reference to a UErrorCode to receive any errors. Possible
+ * errors are U_REGEX_INVALID_STATE if no match has been
+ * attempted or the last match failed.
+ * @return the index of the last character matched, plus one.
+ * The index value returned is a native index, corresponding to
+ * code units for the underlying encoding type, for example,
+ * a byte index for UTF-8.
+ * @stable ICU 4.6
*/
virtual int64_t end64(UErrorCode &status) const;
/**
* Returns the index in the input string of the character following the
* text matched by the specified capture group during the previous match operation.
+ *
* @param group the capture group number
* @param status A reference to a UErrorCode to receive any errors. Possible
* errors are U_REGEX_INVALID_STATE if no match has been
* attempted or the last match failed and
* U_INDEX_OUTOFBOUNDS_ERROR for a bad capture group number
* @return the index of the first character following the text
- * captured by the specifed group during the previous match operation.
+ * captured by the specified group during the previous match operation.
* Return -1 if the capture group exists in the pattern but was not part of the match.
* The index value returned is a native index, corresponding to
* code units for the underlying encoding type, for example,
virtual int32_t end(int32_t group, UErrorCode &status) const;
/**
- * @draft ICU 4.6
+ * Returns the index in the input string of the character following the
+ * text matched by the specified capture group during the previous match operation.
+ *
+ * @param group the capture group number
+ * @param status A reference to a UErrorCode to receive any errors. Possible
+ * errors are U_REGEX_INVALID_STATE if no match has been
+ * attempted or the last match failed and
+ * U_INDEX_OUTOFBOUNDS_ERROR for a bad capture group number
+ * @return the index of the first character following the text
+ * captured by the specified group during the previous match operation.
+ * Return -1 if the capture group exists in the pattern but was not part of the match.
+ * The index value returned is a native index, corresponding to
+ * code units for the underlying encoding type, for example,
+ * a byte index for UTF8.
+ * @stable ICU 4.6
*/
virtual int64_t end64(int32_t group, UErrorCode &status) const;
-
/**
* Resets this matcher. The effect is to remove any memory of previous matches,
* and to cause subsequent find() operations to begin at the beginning of
* until after regexp operations on it are done.
* @return this RegexMatcher.
*
- * @draft ICU 4.6
+ * @stable ICU 4.6
*/
virtual RegexMatcher &reset(UText *input);
+
+ /**
+ * Set the subject text string upon which the regular expression is looking for matches
+ * without changing any other aspect of the matching state.
+ * The new and previous text strings must have the same content.
+ *
+ * This function is intended for use in environments where ICU is operating on
+ * strings that may move around in memory. It provides a mechanism for notifying
+ * ICU that the string has been relocated, and providing a new UText to access the
+ * string in its new position.
+ *
+ * Note that the regular expression implementation never copies the underlying text
+ * of a string being matched, but always operates directly on the original text
+ * provided by the user. Refreshing simply drops the references to the old text
+ * and replaces them with references to the new.
+ *
+ * Caution: this function is normally used only by very specialized,
+ * system-level code. One example use case is with garbage collection that moves
+ * the text in memory.
+ *
+ * @param input The new (moved) text string.
+ * @param status Receives errors detected by this function.
+ *
+ * @stable ICU 4.8
+ */
+ virtual RegexMatcher &refreshInputText(UText *input, UErrorCode &status);
+
private:
/**
- * Cause a compilation error if an application accidently attempts to
+ * Cause a compilation error if an application accidentally attempts to
* reset a matcher with a (UChar *) string as input rather than
* a UnicodeString. Avoids a dangling reference to a temporary string.
* <p>
* or in a UText, using
* <code>utext_openUChars(UText *ut, const UChar *text, int64_t textLength, UErrorCode *status);</code>
*
- * @internal
*/
RegexMatcher &reset(const UChar *input);
public:
* a UnicodeString.
* @return the input text
*
- * @draft ICU 4.6
+ * @stable ICU 4.6
*/
virtual UText *inputText() const;
* UText parameter or by returning a shallow clone of the live input. Note that copying
* the entire input may cause significant performance and memory issues.
* @param dest The UText into which the input should be copied, or NULL to create a new UText
+ * @param status error code
* @return dest if non-NULL, a shallow copy of the input text otherwise
*
- * @draft ICU 4.6
+ * @stable ICU 4.6
*/
virtual UText *getInput(UText *dest, UErrorCode &status) const;
/**
* Identical to region(start, limit, status) but also allows a start position without
* resetting the region state.
+ * @param regionStart The region start
+ * @param regionLimit the limit of the region
* @param startIndex The (native) index within the region bounds at which to begin searches.
* @param status A reference to a UErrorCode to receive any errors.
* If startIndex is not within the specified region bounds,
* U_INDEX_OUTOFBOUNDS_ERROR is returned.
- * @draft ICU 4.6
+ * @stable ICU 4.6
*/
virtual RegexMatcher ®ion(int64_t regionStart, int64_t regionLimit, int64_t startIndex, UErrorCode &status);
virtual int32_t regionStart() const;
/**
- * @draft ICU 4.6
- */
+ * Reports the start index of this matcher's region. The searches this matcher
+ * conducts are limited to finding matches within regionStart (inclusive) and
+ * regionEnd (exclusive).
+ *
+ * @return The starting (native) index of this matcher's region.
+ * @stable ICU 4.6
+ */
virtual int64_t regionStart64() const;
virtual int32_t regionEnd() const;
/**
- * @draft ICU 4.6
- */
+ * Reports the end (limit) index (exclusive) of this matcher's region. The searches
+ * this matcher conducts are limited to finding matches within regionStart
+ * (inclusive) and regionEnd (exclusive).
+ *
+ * @return The ending point (native) of this matcher's region.
+ * @stable ICU 4.6
+ */
virtual int64_t regionEnd64() const;
/**
/**
* Return true if this matcher is using anchoring bounds.
- * By default, matchers use anchoring region boounds.
+ * By default, matchers use anchoring region bounds.
*
* @return TRUE if this matcher is using anchoring bounds.
* @stable ICU 4.0
/**
- * Return TRUE if the most recent matching operation touched the
- * end of the text being processed. In this case, additional input text could
- * change the results of that match.
+ * Return TRUE if the most recent matching operation attempted to access
+ * additional input beyond the available input text.
+ * In this case, additional input text could change the results of the match.
*
* hitEnd() is defined for both successful and unsuccessful matches.
* In either case hitEnd() will return TRUE if if the end of the text was
* @return a string containing the results of the find and replace.
* If a pre-allocated UText was provided, it will always be used and returned.
*
- * @draft ICU 4.6
+ * @stable ICU 4.6
*/
virtual UText *replaceAll(UText *replacement, UText *dest, UErrorCode &status);
* @return a string containing the results of the find and replace.
* If a pre-allocated UText was provided, it will always be used and returned.
*
- * @draft ICU 4.6
+ * @stable ICU 4.6
*/
virtual UText *replaceFirst(UText *replacement, UText *dest, UErrorCode &status);
*
* @return this RegexMatcher
*
- * @draft ICU 4.6
+ * @stable ICU 4.6
*/
virtual RegexMatcher &appendReplacement(UText *dest,
UText *replacement, UErrorCode &status);
*
* @param dest A mutable UText to which the results of the find-and-replace are appended.
* Must not be NULL.
+ * @param status error cod
* @return the destination string.
*
- * @draft ICU 4.6
+ * @stable ICU 4.6
*/
virtual UText *appendTail(UText *dest, UErrorCode &status);
* @param status A reference to a UErrorCode to receive any errors.
* @return The number of fields into which the input string was split.
*
- * @draft ICU 4.6
+ * @stable ICU 4.6
*/
virtual int32_t split(UText *input,
UText *dest[],
virtual int32_t getTimeLimit() const;
/**
- * Set the amount of heap storage avaliable for use by the match backtracking stack.
+ * Set the amount of heap storage available for use by the match backtracking stack.
* The matcher is also reset, discarding any results from previous matches.
* <p>
* ICU uses a backtracking regular expression engine, with the backtrack stack
/**
* Get the callback function for this URegularExpression.
*
- * @param callback Out paramater, receives a pointer to the user-supplied
+ * @param callback Out parameter, receives a pointer to the user-supplied
* callback function.
* @param context Out parameter, receives the user context pointer that
* was set when uregex_setMatchCallback() was called.
* time the callback function is set will be saved
* and passed to the callback each time that it is called.
* @param status A reference to a UErrorCode to receive any errors.
- * @draft ICU 4.6
+ * @stable ICU 4.6
*/
virtual void setFindProgressCallback(URegexFindProgressCallback *callback,
const void *context,
/**
* Get the find progress callback function for this URegularExpression.
*
- * @param callback Out paramater, receives a pointer to the user-supplied
+ * @param callback Out parameter, receives a pointer to the user-supplied
* callback function.
* @param context Out parameter, receives the user context pointer that
* was set when uregex_setFindProgressCallback() was called.
* @param status A reference to a UErrorCode to receive any errors.
- * @draft ICU 4.6
+ * @stable ICU 4.6
*/
virtual void getFindProgressCallback(URegexFindProgressCallback *&callback,
const void *&context,
UErrorCode &status);
-
+#ifndef U_HIDE_INTERNAL_API
/**
* setTrace Debug function, enable/disable tracing of the matching engine.
* For internal ICU development use only. DO NO USE!!!!
* @internal
*/
void setTrace(UBool state);
-
+#endif /* U_HIDE_INTERNAL_API */
/**
* ICU "poor man's RTTI", returns a UClassID for this class.
friend class RegexPattern;
friend class RegexCImpl;
public:
+#ifndef U_HIDE_INTERNAL_API
/** @internal */
void resetPreserveRegion(); // Reset matcher state, but preserve any region.
+#endif /* U_HIDE_INTERNAL_API */
private:
//
REStackFrame *resetStack();
inline REStackFrame *StateSave(REStackFrame *fp, int64_t savePatIdx, UErrorCode &status);
void IncrementTime(UErrorCode &status);
- UBool ReportFindProgress(int64_t matchIndex, UErrorCode &status);
+
+ // Call user find callback function, if set. Return TRUE if operation should be interrupted.
+ inline UBool findProgressInterrupt(int64_t matchIndex, UErrorCode &status);
int64_t appendGroup(int32_t groupNum, UText *dest, UErrorCode &status) const;
- UBool findUsingChunk();
+ UBool findUsingChunk(UErrorCode &status);
void MatchChunkAt(int32_t startIdx, UBool toEnd, UErrorCode &status);
UBool isChunkWordBoundary(int32_t pos);
// reported, or that permanently disables this matcher.
RuleBasedBreakIterator *fWordBreakItr;
-
-
};
U_NAMESPACE_END
projects / apple / icu.git / blobdiff