ICU-491.11.1.tar.gz

[apple/icu.git] / icuSources / i18n / regeximp.h

diff --git a/icuSources/i18n/regeximp.h b/icuSources/i18n/regeximp.h
index 07206e6448d4c60aeabc6c7466c6030d553f3c95..31d333caf7733eb9b63501e52f755994505ff821 100644 (file)
--- a/icuSources/i18n/regeximp.h
+++ b/icuSources/i18n/regeximp.h
@@ -1,5 +1,5 @@
 //
 //

-//   Copyright (C) 2002-2007 International Business Machines Corporation
+//   Copyright (C) 2002-2012 International Business Machines Corporation

 //   and others. All rights reserved.
 //
 //   file:  regeximp.h
 //   and others. All rights reserved.
 //
 //   file:  regeximp.h


@@ -12,11 +12,21 @@
 #ifndef _REGEXIMP_H
 #define _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 "cmemory.h"

+#include "ucase.h"

 
 U_NAMESPACE_BEGIN
 
 
 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,
+//   ./runConfigureICU --enable-debug --disable-release Linux CPPFLAGS="-DREGEX_DEBUG"
+
+#ifdef REGEX_DEBUG   

 //
 //  debugging options.  Enable one or more of the three #defines immediately following
 //
 //
 //  debugging options.  Enable one or more of the three #defines immediately following
 //


@@ -279,12 +289,18 @@ enum {
 //  Match Engine State Stack Frame Layout.
 //
 struct REStackFrame {
 //  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.
                                          //   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
 
 //
 //  Start-Of-Match type.  Used by find() to quickly scan to positions where a


@@ -307,7 +323,6 @@ enum StartOfMatch {
                                (v)==START_STRING?  "START_STRING"  : \
                                                    "ILLEGAL")
 
                                (v)==START_STRING?  "START_STRING"  : \
                                                    "ILLEGAL")
 

-

 //
 //  8 bit set, to fast-path latin-1 set membership tests.
 //
 //
 //  8 bit set, to fast-path latin-1 set membership tests.
 //


@@ -348,6 +363,60 @@ inline void Regex8BitSet::operator = (const Regex8BitSet &s) {
 }
 
 
 }
 
 

+//  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  UCaseProps *fcsp;
+        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  UCaseProps *fcsp;
+        const  UChar      *fFoldChars;
+        int32_t            fFoldLength;
+        int32_t            fFoldIndex;
+
+};
+

 U_NAMESPACE_END
 #endif
 
 U_NAMESPACE_END
 #endif