+// © 2016 and later: Unicode, Inc. and others.
+// License & terms of use: http://www.unicode.org/copyright.html
//
-// Copyright (C) 2002-2007 International Business Machines Corporation
+// Copyright (C) 2002-2015 International Business Machines Corporation
// and others. All rights reserved.
//
// file: regeximp.h
#ifndef _REGEXIMP_H
#define _REGEXIMP_H
+#include "unicode/utypes.h"
+#include "unicode/uobject.h"
+#include "unicode/uniset.h"
+#include "unicode/utext.h"
+
#include "cmemory.h"
+#include "ucase.h"
U_NAMESPACE_BEGIN
-#ifdef REGEX_DEBUG /* For debugging, define REGEX_DEBUG in regex.h, not here in this file. */
+// For debugging, define REGEX_DEBUG
+// To define with configure,
+// CPPFLAGS="-DREGEX_DEBUG" ./runConfigureICU --enable-debug --disable-release Linux
+
+#ifdef REGEX_DEBUG
//
// debugging options. Enable one or more of the three #defines immediately following
//
#define REGEX_SCAN_DEBUG_PRINTF(a)
#endif
-#ifdef REGEX_DUMP_DEBUG
-#define REGEX_DUMP_DEBUG_PRINTF(a) printf a
-#else
-#define REGEX_DUMP_DEBUG_PRINTF(a)
-#endif
-
-#ifdef REGEX_RUN_DEBUG
-#define REGEX_RUN_DEBUG_PRINTF(a) printf a
-#define REGEX_DUMP_DEBUG_PRINTF(a) printf a
-#else
-#define REGEX_RUN_DEBUG_PRINTF(a)
-#endif
-
//
// Opcode types In the compiled form of the regexp, these are the type, or opcodes,
URX_BACKSLASH_BU = 53, // \b or \B in UREGEX_UWORD mode, using Unicode style
// word boundaries.
URX_DOLLAR_D = 54, // $ end of input test, in UNIX_LINES mode.
- URX_DOLLAR_MD = 55 // $ end of input test, in MULTI_LINE and UNIX_LINES mode.
+ URX_DOLLAR_MD = 55, // $ end of input test, in MULTI_LINE and UNIX_LINES mode.
+ URX_BACKSLASH_H = 56, // Value field: 0: \h 1: \H
+ URX_BACKSLASH_R = 57, // Any line break sequence.
+ URX_BACKSLASH_V = 58 // Value field: 0: \v 1: \V
};
"LOOP_DOT_I", \
"BACKSLASH_BU", \
"DOLLAR_D", \
- "DOLLAR_MD"
+ "DOLLAR_MD", \
+ "URX_BACKSLASH_H", \
+ "URX_BACKSLASH_R", \
+ "URX_BACKSLASH_V"
//
// Convenience macros for assembling and disassembling a compiled operation.
//
-#define URX_BUILD(type, val) (int32_t)((type << 24) | (val))
#define URX_TYPE(x) ((uint32_t)(x) >> 24)
#define URX_VAL(x) ((x) & 0xffffff)
// Match Engine State Stack Frame Layout.
//
struct REStackFrame {
- int32_t fInputIdx; // Position of next character in the input string
- int32_t fPatIdx; // Position of next Op in the compiled pattern
- int32_t fExtra[2]; // Extra state, for capture group start/ends
+ // Header
+ int64_t fInputIdx; // Position of next character in the input string
+ int64_t fPatIdx; // Position of next Op in the compiled pattern
+ // (int64_t for UVector64, values fit in an int32_t)
+ // Remainder
+ int64_t fExtra[1]; // Extra state, for capture group start/ends
// atomic parentheses, repeat counts, etc.
// Locations assigned at pattern compile time.
+ // Variable-length array.
};
+// number of UVector elements in the header
+#define RESTACKFRAME_HDRCOUNT 2
//
// Start-Of-Match type. Used by find() to quickly scan to positions where a
(v)==START_STRING? "START_STRING" : \
"ILLEGAL")
-
//
// 8 bit set, to fast-path latin-1 set membership tests.
//
}
+// Case folded UText Iterator helper class.
+// Wraps a UText, provides a case-folded enumeration over its contents.
+// Used in implementing case insensitive matching constructs.
+// Implementation in rematch.cpp
+
+class CaseFoldingUTextIterator: public UMemory {
+ public:
+ CaseFoldingUTextIterator(UText &text);
+ ~CaseFoldingUTextIterator();
+
+ UChar32 next(); // Next case folded character
+
+ UBool inExpansion(); // True if last char returned from next() and the
+ // next to be returned both originated from a string
+ // folding of the same code point from the orignal UText.
+ private:
+ UText &fUText;
+ const UChar *fFoldChars;
+ int32_t fFoldLength;
+ int32_t fFoldIndex;
+
+};
+
+
+// Case folded UChar * string iterator.
+// Wraps a UChar *, provides a case-folded enumeration over its contents.
+// Used in implementing case insensitive matching constructs.
+// Implementation in rematch.cpp
+
+class CaseFoldingUCharIterator: public UMemory {
+ public:
+ CaseFoldingUCharIterator(const UChar *chars, int64_t start, int64_t limit);
+ ~CaseFoldingUCharIterator();
+
+ UChar32 next(); // Next case folded character
+
+ UBool inExpansion(); // True if last char returned from next() and the
+ // next to be returned both originated from a string
+ // folding of the same code point from the orignal UText.
+
+ int64_t getIndex(); // Return the current input buffer index.
+
+ private:
+ const UChar *fChars;
+ int64_t fIndex;
+ int64_t fLimit;
+ const UChar *fFoldChars;
+ int32_t fFoldLength;
+ int32_t fFoldIndex;
+
+};
+
U_NAMESPACE_END
#endif
projects / apple / icu.git / blobdiff