+// © 2016 and later: Unicode, Inc. and others.
+// License & terms of use: http://www.unicode.org/copyright.html
/*
**********************************************************************
-* Copyright (C) 2002-2012, International Business Machines
+* Copyright (C) 2002-2016, International Business Machines
* Corporation and others. All Rights Reserved.
**********************************************************************
* file name: regex.h
-* encoding: US-ASCII
+* encoding: UTF-8
* indentation:4
*
* created on: 2002oct22
* \file
* \brief C++ API: Regular Expressions
*
- * <h2>Regular Expression API</h2>
- *
- * <p>The ICU API for processing regular expressions consists of two classes,
- * <code>RegexPattern</code> and <code>RegexMatcher</code>.
- * <code>RegexPattern</code> objects represent a pre-processed, or compiled
+ * The ICU API for processing regular expressions consists of two classes,
+ * `RegexPattern` and `RegexMatcher`.
+ * `RegexPattern` objects represent a pre-processed, or compiled
* regular expression. They are created from a regular expression pattern string,
- * and can be used to create <code>RegexMatcher</code> objects for the pattern.</p>
+ * and can be used to create `RegexMatcher` objects for the pattern.
*
- * <p>Class <code>RegexMatcher</code> bundles together a regular expression
+ * Class `RegexMatcher` bundles together a regular expression
* pattern and a target string to which the search pattern will be applied.
- * <code>RegexMatcher</code> includes API for doing plain find or search
+ * `RegexMatcher` includes API for doing plain find or search
* operations, for search and replace operations, and for obtaining detailed
- * information about bounds of a match. </p>
+ * information about bounds of a match.
*
- * <p>Note that by constructing <code>RegexMatcher</code> objects directly from regular
+ * Note that by constructing `RegexMatcher` objects directly from regular
* expression pattern strings application code can be simplified and the explicit
- * need for <code>RegexPattern</code> objects can usually be eliminated.
- * </p>
+ * need for `RegexPattern` objects can usually be eliminated.
+ *
*/
#include "unicode/utypes.h"
// Forward Declarations
+struct UHashtable;
+
+#if U_SHOW_CPLUSPLUS_API
U_NAMESPACE_BEGIN
struct Regex8BitSet;
class UVector32;
class UVector64;
-/**
- * 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 <code>RegexPattern</code> represents a compiled regular expression. It includes
+ * Class `RegexPattern` represents a compiled regular expression. It includes
* factory methods for creating a RegexPattern object from the source (string) form
* of a regular expression, methods for creating RegexMatchers that allow the pattern
* to be applied to input text, and a few convenience methods for simple common
* uses of regular expressions.
*
- * <p>Class RegexPattern is not intended to be subclassed.</p>
+ * Class RegexPattern is not intended to be subclassed.
*
* @stable ICU 2.4
*/
-class U_I18N_API RegexPattern: public UObject {
+class U_I18N_API RegexPattern U_FINAL : public UObject {
public:
/**
* default constructor. Create a RegexPattern object that refers to no actual
* pattern. Not normally needed; RegexPattern objects are usually
- * created using the factory method <code>compile()</code>.
+ * created using the factory method `compile()`.
*
* @stable ICU 2.4
*/
/**
* Comparison operator. Two RegexPattern objects are considered equal if they
- * were constructed from identical source patterns using the same match flag
+ * were constructed from identical source patterns using the same #URegexpFlag
* settings.
* @param that a RegexPattern object to compare with "this".
* @return TRUE if the objects are equivalent.
/**
* Comparison operator. Two RegexPattern objects are considered equal if they
- * were constructed from identical source patterns using the same match flag
+ * were constructed from identical source patterns using the same #URegexpFlag
* settings.
* @param that a RegexPattern object to compare with "this".
* @return TRUE if the objects are different.
/**
* 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
* object. These compile methods, rather than the constructors, are the usual
* way that RegexPattern objects are created.
*
- * <p>Note that RegexPattern objects must not be deleted while RegexMatcher
+ * Note that RegexPattern objects must not be deleted while RegexMatcher
* objects created from the pattern are active. RegexMatchers keep a pointer
* back to their pattern, so premature deletion of the pattern is a
- * catastrophic error.</p>
+ * catastrophic error.
*
- * <p>All pattern match mode flags are set to their default values.</p>
+ * All #URegexpFlag pattern match mode flags are set to their default values.
*
- * <p>Note that it is often more convenient to construct a RegexMatcher directly
+ * Note that it is often more convenient to construct a RegexMatcher directly
* from a pattern string rather than separately compiling the pattern and
- * then creating a RegexMatcher object from the pattern.</p>
+ * then creating a RegexMatcher object from the pattern.
*
* @param regex The regular expression to be compiled.
* @param pe Receives the position (line and column nubers) of any error
* object. These compile methods, rather than the constructors, are the usual
* way that RegexPattern objects are created.
*
- * <p>Note that RegexPattern objects must not be deleted while RegexMatcher
+ * Note that RegexPattern objects must not be deleted while RegexMatcher
* objects created from the pattern are active. RegexMatchers keep a pointer
* back to their pattern, so premature deletion of the pattern is a
- * catastrophic error.</p>
+ * catastrophic error.
*
- * <p>All pattern match mode flags are set to their default values.</p>
+ * All #URegexpFlag pattern match mode flags are set to their default values.
*
- * <p>Note that it is often more convenient to construct a RegexMatcher directly
+ * Note that it is often more convenient to construct a RegexMatcher directly
* from a pattern string rather than separately compiling the pattern and
- * then creating a RegexMatcher object from the pattern.</p>
+ * then creating a RegexMatcher object from the pattern.
*
* @param regex The regular expression to be compiled. Note, the text referred
* to by this UText must not be deleted during the lifetime of the
/**
* Compiles the regular expression in string form into a RegexPattern
- * object using the specified match mode flags. These compile methods,
+ * object using the specified #URegexpFlag match mode flags. These compile methods,
* rather than the constructors, are the usual way that RegexPattern objects
* are created.
*
- * <p>Note that RegexPattern objects must not be deleted while RegexMatcher
+ * Note that RegexPattern objects must not be deleted while RegexMatcher
* objects created from the pattern are active. RegexMatchers keep a pointer
* back to their pattern, so premature deletion of the pattern is a
- * catastrophic error.</p>
+ * catastrophic error.
*
- * <p>Note that it is often more convenient to construct a RegexMatcher directly
+ * Note that it is often more convenient to construct a RegexMatcher directly
* from a pattern string instead of than separately compiling the pattern and
- * then creating a RegexMatcher object from the pattern.</p>
+ * then creating a RegexMatcher object from the pattern.
*
* @param regex The regular expression to be compiled.
- * @param flags The match mode flags to be used.
+ * @param flags The #URegexpFlag match mode flags to be used, e.g. #UREGEX_CASE_INSENSITIVE.
* @param pe Receives the position (line and column numbers) of any error
* within the regular expression.)
* @param status A reference to a UErrorCode to receive any errors.
/**
* Compiles the regular expression in string form into a RegexPattern
- * object using the specified match mode flags. These compile methods,
+ * object using the specified #URegexpFlag match mode flags. These compile methods,
* rather than the constructors, are the usual way that RegexPattern objects
* are created.
*
- * <p>Note that RegexPattern objects must not be deleted while RegexMatcher
+ * Note that RegexPattern objects must not be deleted while RegexMatcher
* objects created from the pattern are active. RegexMatchers keep a pointer
* back to their pattern, so premature deletion of the pattern is a
- * catastrophic error.</p>
+ * catastrophic error.
*
- * <p>Note that it is often more convenient to construct a RegexMatcher directly
+ * Note that it is often more convenient to construct a RegexMatcher directly
* from a pattern string instead of than separately compiling the pattern and
- * then creating a RegexMatcher object from the pattern.</p>
+ * then creating a RegexMatcher object from the pattern.
*
* @param regex The regular expression to be compiled. Note, the text referred
* to by this UText must not be deleted during the lifetime of the
* RegexPattern object or any RegexMatcher object created from it.
- * @param flags The match mode flags to be used.
+ * @param flags The #URegexpFlag match mode flags to be used, e.g. #UREGEX_CASE_INSENSITIVE.
* @param pe Receives the position (line and column numbers) of any error
* within the regular expression.)
* @param status A reference to a UErrorCode to receive any errors.
/**
* Compiles the regular expression in string form into a RegexPattern
- * object using the specified match mode flags. These compile methods,
+ * object using the specified #URegexpFlag match mode flags. These compile methods,
* rather than the constructors, are the usual way that RegexPattern objects
* are created.
*
- * <p>Note that RegexPattern objects must not be deleted while RegexMatcher
+ * Note that RegexPattern objects must not be deleted while RegexMatcher
* objects created from the pattern are active. RegexMatchers keep a pointer
* back to their pattern, so premature deletion of the pattern is a
- * catastrophic error.</p>
+ * catastrophic error.
*
- * <p>Note that it is often more convenient to construct a RegexMatcher directly
+ * Note that it is often more convenient to construct a RegexMatcher directly
* from a pattern string instead of than separately compiling the pattern and
- * then creating a RegexMatcher object from the pattern.</p>
+ * then creating a RegexMatcher object from the pattern.
*
* @param regex The regular expression to be compiled.
- * @param flags The match mode flags to be used.
+ * @param flags The #URegexpFlag match mode flags to be used, e.g. #UREGEX_CASE_INSENSITIVE.
* @param status A reference to a UErrorCode to receive any errors.
* @return A regexPattern object for the compiled pattern.
*
/**
* Compiles the regular expression in string form into a RegexPattern
- * object using the specified match mode flags. These compile methods,
+ * object using the specified #URegexpFlag match mode flags. These compile methods,
* rather than the constructors, are the usual way that RegexPattern objects
* are created.
*
- * <p>Note that RegexPattern objects must not be deleted while RegexMatcher
+ * Note that RegexPattern objects must not be deleted while RegexMatcher
* objects created from the pattern are active. RegexMatchers keep a pointer
* back to their pattern, so premature deletion of the pattern is a
- * catastrophic error.</p>
+ * catastrophic error.
*
- * <p>Note that it is often more convenient to construct a RegexMatcher directly
+ * Note that it is often more convenient to construct a RegexMatcher directly
* from a pattern string instead of than separately compiling the pattern and
- * then creating a RegexMatcher object from the pattern.</p>
+ * then creating a RegexMatcher object from the pattern.
*
* @param regex The regular expression to be compiled. Note, the text referred
* to by this UText must not be deleted during the lifetime of the
* RegexPattern object or any RegexMatcher object created from it.
- * @param flags The match mode flags to be used.
+ * @param flags The #URegexpFlag match mode flags to be used, e.g. #UREGEX_CASE_INSENSITIVE.
* @param status A reference to a UErrorCode to receive any errors.
* @return A regexPattern object for the compiled pattern.
*
UErrorCode &status);
/**
- * Get the match mode flags that were used when compiling this pattern.
- * @return the match mode flags
+ * Get the #URegexpFlag match mode flags that were used when compiling this pattern.
+ * @return the #URegexpFlag match mode flags
* @stable ICU 2.4
*/
virtual uint32_t flags() const;
* 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 retain a reference to the supplied input string, and all regexp
* pattern matching operations happen directly on this original string. It is
* critical that the string not be altered or deleted before use by the regular
private:
/**
* Cause a compilation error if an application accidentally attempts to
- * create a matcher with a (UChar *) string as input rather than
+ * create a matcher with a (char16_t *) string as input rather than
* a UnicodeString. Avoids a dangling reference to a temporary string.
- * <p>
- * To efficiently work with UChar *strings, wrap the data in a UnicodeString
+ *
+ * To efficiently work with char16_t *strings, wrap the data in a UnicodeString
* using one of the aliasing constructors, such as
- * <code>UnicodeString(UBool isTerminated, const UChar *text, int32_t textLength);</code>
+ * `UnicodeString(UBool isTerminated, const char16_t *text, int32_t textLength);`
* or in a UText, using
- * <code>utext_openUChars(UText *ut, const UChar *text, int64_t textLength, UErrorCode *status);</code>
+ * `utext_openUChars(UText *ut, const char16_t *text, int64_t textLength, UErrorCode *status);`
*
- * @internal
*/
- RegexMatcher *matcher(const UChar *input,
+ RegexMatcher *matcher(const char16_t *input,
UErrorCode &status) const;
public:
virtual UText *patternText(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.
+ * @param status A UErrorCode to receive any errors.
+ *
+ * @stable 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.
+ *
+ * @stable 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
/**
- * Split a string into fields. Somewhat like split() from Perl or Java.
+ * 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.
* 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
+ * `RegexMatcher::split()` is preferable to this function
*
* @param input The string to be split into fields. The field delimiters
* match the pattern (in the "this" object)
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 */
};
*
* @stable ICU 2.4
*/
-class U_I18N_API RegexMatcher: public UObject {
+class U_I18N_API RegexMatcher U_FINAL : public UObject {
public:
/**
* its matcher() method to create the RegexMatcher objects.
*
* @param regexp The Regular Expression to be compiled.
- * @param flags Regular expression options, such as case insensitive matching.
- * @see UREGEX_CASE_INSENSITIVE
+ * @param flags #URegexpFlag options, such as #UREGEX_CASE_INSENSITIVE.
* @param status Any errors are reported by setting this UErrorCode variable.
* @stable ICU 2.6
*/
* its matcher() method to create the RegexMatcher objects.
*
* @param regexp The regular expression to be compiled.
- * @param flags Regular expression options, such as case insensitive matching.
- * @see UREGEX_CASE_INSENSITIVE
+ * @param flags #URegexpFlag options, such as #UREGEX_CASE_INSENSITIVE.
* @param status Any errors are reported by setting this UErrorCode variable.
*
* @stable ICU 4.6
* created for the same expression, it will be more efficient to
* separately create and cache a RegexPattern object, and use
* its matcher() method to create the RegexMatcher objects.
- * <p>
+ *
* The matcher will retain a reference to the supplied input string, and all regexp
* pattern matching operations happen directly on the original string. It is
* critical that the string not be altered or deleted before use by the regular
* @param regexp The Regular Expression to be compiled.
* @param input The string to match. The matcher retains a reference to the
* caller's string; mo copy is made.
- * @param flags Regular expression options, such as case insensitive matching.
- * @see UREGEX_CASE_INSENSITIVE
+ * @param flags #URegexpFlag options, such as #UREGEX_CASE_INSENSITIVE.
* @param status Any errors are reported by setting this UErrorCode variable.
* @stable ICU 2.6
*/
* created for the same expression, it will be more efficient to
* separately create and cache a RegexPattern object, and use
* its matcher() method to create the RegexMatcher objects.
- * <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
*
* @param regexp The Regular Expression to be compiled.
* @param input The string to match. The matcher retains a shallow clone of the text.
- * @param flags Regular expression options, such as case insensitive matching.
- * @see UREGEX_CASE_INSENSITIVE
+ * @param flags #URegexpFlag options, such as #UREGEX_CASE_INSENSITIVE.
* @param status Any errors are reported by setting this UErrorCode variable.
*
* @stable ICU 4.6
private:
/**
* Cause a compilation error if an application accidentally attempts to
- * create a matcher with a (UChar *) string as input rather than
+ * create a matcher with a (char16_t *) string as input rather than
* a UnicodeString. Avoids a dangling reference to a temporary string.
- * <p>
- * To efficiently work with UChar *strings, wrap the data in a UnicodeString
+ *
+ * To efficiently work with char16_t *strings, wrap the data in a UnicodeString
* using one of the aliasing constructors, such as
- * <code>UnicodeString(UBool isTerminated, const UChar *text, int32_t textLength);</code>
+ * `UnicodeString(UBool isTerminated, const char16_t *text, int32_t textLength);`
* or in a UText, using
- * <code>utext_openUChars(UText *ut, const UChar *text, int64_t textLength, UErrorCode *status);</code>
- *
- * @internal
+ * `utext_openUChars(UText *ut, const char16_t *text, int64_t textLength, UErrorCode *status);`
*/
- RegexMatcher(const UnicodeString ®exp, const UChar *input,
+ RegexMatcher(const UnicodeString ®exp, const char16_t *input,
uint32_t flags, UErrorCode &status);
public:
* always starts at the beginning of the input region;
* unlike that function, it does not require that the entire region be matched.
*
- * <p>If the match succeeds then more information can be obtained via the <code>start()</code>,
- * <code>end()</code>, and <code>group()</code> functions.</p>
+ * If the match succeeds then more information can be obtained via the start(),
+ * end(), and group() functions.
*
* @param status A reference to a UErrorCode to receive any errors.
* @return TRUE if there is a match at the start of the input string.
* The match may be of any length, and is not required to extend to the end
* of the input string. Contrast with match().
*
- * <p>If the match succeeds then more information can be obtained via the <code>start()</code>,
- * <code>end()</code>, and <code>group()</code> functions.</p>
+ * If the match succeeds then more information can be obtained via the start(),
+ * end(), and group() functions.
*
* @param startIndex The input string (native) index at which to begin matching.
* @param status A reference to a UErrorCode to receive any errors.
* 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>
+ * If a match is found, `start()`, `end()` and `group()`
* will provide more information regarding the match.
- * <p>Note that if the input string is changed by the application,
+ * 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>
+ * position may not be valid with the altered input string.
* @return TRUE if a match is found.
* @stable ICU 2.4
*/
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, `start()`, `end()` and `group()`
+ * will provide more information regarding the match.
+ *
+ * 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.
+ * @param status A reference to a UErrorCode to receive any errors.
+ * @return TRUE if a match is found.
+ * @stable 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.
* Returns a string containing the text captured by the given group
* during the previous match operation. Group(0) is the entire match.
*
+ * A zero length string is returned both for capture groups that did not
+ * participate in the match and for actual zero length matches.
+ * To distinguish between these two cases use the function start(),
+ * which returns -1 for non-participating groups.
+ *
* @param groupNum 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
*/
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.
*
+ * A group length of zero is returned both for capture groups that did not
+ * participate in the match and for actual zero length matches.
+ * To distinguish between these two cases use the function start(),
+ * which returns -1 for non-participating groups.
+ *
* @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
*/
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.
- *
- * @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 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.
- *
- * @internal ICU 4.4 technology preview
- */
- virtual UText *group(int32_t groupNum, UText *dest, UErrorCode &status) const;
-
-
/**
* Returns the index in the input string of the start of the text matched
* during the previous match operation.
*/
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.
*/
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
* The effect is to remove any memory of previous matches,
* and to cause subsequent find() operations to begin at
* the specified (native) position in the input string.
- * <p>
+ *
* The matcher's region is reset to its default, which is the entire
* input string.
- * <p>
+ *
* An alternative to this function is to set a match region
* beginning at the desired index.
*
private:
/**
* Cause a compilation error if an application accidentally attempts to
- * reset a matcher with a (UChar *) string as input rather than
+ * reset a matcher with a (char16_t *) string as input rather than
* a UnicodeString. Avoids a dangling reference to a temporary string.
- * <p>
- * To efficiently work with UChar *strings, wrap the data in a UnicodeString
+ *
+ * To efficiently work with char16_t *strings, wrap the data in a UnicodeString
* using one of the aliasing constructors, such as
- * <code>UnicodeString(UBool isTerminated, const UChar *text, int32_t textLength);</code>
+ * `UnicodeString(UBool isTerminated, const char16_t *text, int32_t textLength);`
* or in a UText, using
- * <code>utext_openUChars(UText *ut, const UChar *text, int64_t textLength, UErrorCode *status);</code>
+ * `utext_openUChars(UText *ut, const char16_t *text, int64_t textLength, UErrorCode *status);`
*
- * @internal
*/
- RegexMatcher &reset(const UChar *input);
+ RegexMatcher &reset(const char16_t *input);
public:
/**
* the pattern with the replacement string. This is a convenience
* function that provides a complete find-and-replace operation.
*
- * <p>This function first resets this RegexMatcher. It then scans the input string
+ * This function first resets this RegexMatcher. It then scans the input string
* looking for a match of the pattern. Input that is not part
* of the match is appended directly to the result string; the match is replaced
* in the result by the replacement string. The replacement string may contain
- * references to captured groups.</p>
+ * references to captured groups.
*
- * <p>The state of the matcher (the position at which a subsequent find()
+ * The state of the matcher (the position at which a subsequent find()
* would begin) after completing a replaceFirst() is not specified. The
- * RegexMatcher should be reset before doing additional find() operations.</p>
+ * RegexMatcher should be reset before doing additional find() operations.
*
* @param replacement a string containing the replacement text.
* @param status a reference to a UErrorCode to receive any errors.
* the pattern with the replacement string. This is a convenience
* function that provides a complete find-and-replace operation.
*
- * <p>This function first resets this RegexMatcher. It then scans the input string
+ * This function first resets this RegexMatcher. It then scans the input string
* looking for a match of the pattern. Input that is not part
* of the match is appended directly to the result string; the match is replaced
* in the result by the replacement string. The replacement string may contain
- * references to captured groups.</p>
+ * references to captured groups.
*
- * <p>The state of the matcher (the position at which a subsequent find()
+ * The state of the matcher (the position at which a subsequent find()
* would begin) after completing a replaceFirst() is not specified. The
- * RegexMatcher should be reset before doing additional find() operations.</p>
+ * RegexMatcher should be reset before doing additional find() operations.
*
* @param replacement a string containing the replacement text.
* @param dest a mutable UText in which the results are placed.
* Implements a replace operation intended to be used as part of an
* incremental find-and-replace.
*
- * <p>The input string, starting from the end of the previous replacement and ending at
+ * The input string, starting from the end of the previous replacement and ending at
* the start of the current match, is appended to the destination string. Then the
* replacement string is appended to the output string,
- * including handling any substitutions of captured text.</p>
+ * including handling any substitutions of captured text.
*
- * <p>For simple, prepackaged, non-incremental find-and-replace
- * operations, see replaceFirst() or replaceAll().</p>
+ * For simple, prepackaged, non-incremental find-and-replace
+ * operations, see replaceFirst() or replaceAll().
*
* @param dest A UnicodeString to which the results of the find-and-replace are appended.
* @param replacement A UnicodeString that provides the text to be substituted for
* Implements a replace operation intended to be used as part of an
* incremental find-and-replace.
*
- * <p>The input string, starting from the end of the previous replacement and ending at
+ * The input string, starting from the end of the previous replacement and ending at
* the start of the current match, is appended to the destination string. Then the
* replacement string is appended to the output string,
- * including handling any substitutions of captured text.</p>
+ * including handling any substitutions of captured text.
*
- * <p>For simple, prepackaged, non-incremental find-and-replace
- * operations, see replaceFirst() or replaceAll().</p>
+ * For simple, prepackaged, non-incremental find-and-replace
+ * operations, see replaceFirst() or replaceAll().
*
* @param dest A mutable UText to which the results of the find-and-replace are appended.
* Must not be NULL.
/**
* As the final step in a find-and-replace operation, append the remainder
* of the input string, starting at the position following the last appendReplacement(),
- * to the destination string. <code>appendTail()</code> is intended to be invoked after one
- * or more invocations of the <code>RegexMatcher::appendReplacement()</code>.
+ * to the destination string. `appendTail()` is intended to be invoked after one
+ * or more invocations of the `RegexMatcher::appendReplacement()`.
*
* @param dest A UnicodeString to which the results of the find-and-replace are appended.
* @return the destination string.
/**
* As the final step in a find-and-replace operation, append the remainder
* of the input string, starting at the position following the last appendReplacement(),
- * to the destination string. <code>appendTail()</code> is intended to be invoked after one
- * or more invocations of the <code>RegexMatcher::appendReplacement()</code>.
+ * to the destination string. `appendTail()` is intended to be invoked after one
+ * or more invocations of the `RegexMatcher::appendReplacement()`.
*
* @param dest A mutable UText to which the results of the find-and-replace are appended.
* Must not be NULL.
/**
- * Split a string into fields. Somewhat like split() from Perl.
+ * 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.
/**
- * Split a string into fields. Somewhat like split() from Perl.
+ * 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.
* infinite loop.
* When a limit is set a match operation will fail with an error if the
* limit is exceeded.
- * <p>
+ *
* The units of the limit are steps of the match engine.
* Correspondence with actual processor time will depend on the speed
* of the processor and the details of the specific pattern, but will
* typically be on the order of milliseconds.
- * <p>
+ *
* By default, the matching time is not limited.
- * <p>
+ *
*
* @param limit The limit value, or 0 for no limit.
* @param status A reference to a UErrorCode to receive any errors.
/**
* 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
* maintained on the heap. This function sets the limit to the amount of memory
- * that can be used for this purpose. A backtracking stack overflow will
+ * that can be used for this purpose. A backtracking stack overflow will
* result in an error from the match operation that caused it.
- * <p>
+ *
* A limit is desirable because a malicious or poorly designed pattern can use
* excessive memory, potentially crashing the process. A limit is enabled
* by default.
- * <p>
+ *
* @param limit The maximum size, in bytes, of the matching backtrack stack.
* A value of zero means no limit.
* The limit must be greater or equal to zero.
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);
};
U_NAMESPACE_END
+#endif // U_SHOW_CPLUSPLUS_API
+
#endif // UCONFIG_NO_REGULAR_EXPRESSIONS
#endif
projects / apple / icu.git / blobdiff