pcre32: Add 32-bit library

Create libpcre32 that operates on 32-bit characters (UTF-32).

This turned out to be surprisingly simple after the UTF-16 support
was introduced; mostly just extra ifdefs and adjusting and adding
some tests.

git-svn-id: svn://vcs.exim.org/pcre/code/trunk@1055 2f5784b3-3f2a-0410-8824-cb99058d5e15

1 files changed, 124 insertions, 9 deletions

diff --git a/pcre.h.in b/pcre.h.in
index 8c0bdda..959bc38 100644
--- a/pcre.h.in
+++ b/pcre.h.in
@@ -117,12 +117,14 @@ compiling). */
 #define PCRE_UNGREEDY           0x00000200  /* Compile */
 #define PCRE_NOTEMPTY           0x00000400  /* Exec, DFA exec */
 /* The next two are also used in exec and DFA exec */
-#define PCRE_UTF8               0x00000800  /* Compile (same as PCRE_UTF16) */
-#define PCRE_UTF16              0x00000800  /* Compile (same as PCRE_UTF8) */
+#define PCRE_UTF8               0x00000800  /* Compile (same as PCRE_UTF16 and PCRE_UTF32) */
+#define PCRE_UTF16              0x00000800  /* Compile (same as PCRE_UTF8 and PCRE_UTF32) */
+#define PCRE_UTF32              0x00000800  /* Compile (same as PCRE_UTF8 and PCRE_UTF16) */
 #define PCRE_NO_AUTO_CAPTURE    0x00001000  /* Compile */
 /* The next two are also used in exec and DFA exec */
-#define PCRE_NO_UTF8_CHECK      0x00002000  /* Compile (same as PCRE_NO_UTF16_CHECK) */
-#define PCRE_NO_UTF16_CHECK     0x00002000  /* Compile (same as PCRE_NO_UTF8_CHECK) */
+#define PCRE_NO_UTF8_CHECK      0x00002000  /* Compile (same as PCRE_NO_UTF16_CHECK and PCRE_NO_UTF32_CHECK) */
+#define PCRE_NO_UTF16_CHECK     0x00002000  /* Compile (same as PCRE_NO_UTF8_CHECK and PCRE_NO_UTF32_CHECK) */
+#define PCRE_NO_UTF32_CHECK     0x00002000  /* Compile (same as PCRE_NO_UTF8_CHECK and PCRE_NO_UTF16_CHECK) */
 #define PCRE_AUTO_CALLOUT       0x00004000  /* Compile */
 #define PCRE_PARTIAL_SOFT       0x00008000  /* Exec, DFA exec */
 #define PCRE_PARTIAL            0x00008000  /* Backwards compatible synonym */
@@ -156,10 +158,12 @@ compiling). */
 #define PCRE_ERROR_NOSUBSTRING      (-7)
 #define PCRE_ERROR_MATCHLIMIT       (-8)
 #define PCRE_ERROR_CALLOUT          (-9)  /* Never used by PCRE itself */
-#define PCRE_ERROR_BADUTF8         (-10)  /* Same for 8/16 */
-#define PCRE_ERROR_BADUTF16        (-10)  /* Same for 8/16 */
-#define PCRE_ERROR_BADUTF8_OFFSET  (-11)  /* Same for 8/16 */
-#define PCRE_ERROR_BADUTF16_OFFSET (-11)  /* Same for 8/16 */
+#define PCRE_ERROR_BADUTF8         (-10)  /* Same for 8/16/32 */
+#define PCRE_ERROR_BADUTF16        (-10)  /* Same for 8/16/32 */
+#define PCRE_ERROR_BADUTF32        (-10)  /* Same for 8/16/32 */
+#define PCRE_ERROR_BADUTF8_OFFSET  (-11)  /* Same for 8/16/32 */
+#define PCRE_ERROR_BADUTF16_OFFSET (-11)  /* Same for 8/16/32 */
+#define PCRE_ERROR_BADUTF32_OFFSET (-11)  /* Same for 8/16/32 */
 #define PCRE_ERROR_PARTIAL         (-12)
 #define PCRE_ERROR_BADPARTIAL      (-13)
 #define PCRE_ERROR_INTERNAL        (-14)
@@ -214,6 +218,12 @@ compiling). */
 #define PCRE_UTF16_ERR3              3
 #define PCRE_UTF16_ERR4              4
 
+/* Specific error codes for UTF-32 validity checks */
+
+#define PCRE_UTF32_ERR0              0
+#define PCRE_UTF32_ERR1              1
+#define PCRE_UTF32_ERR2              2
+
 /* Request types for pcre_fullinfo() */
 
 #define PCRE_INFO_OPTIONS            0
@@ -252,6 +262,7 @@ compatible. */
 #define PCRE_CONFIG_JIT                     9
 #define PCRE_CONFIG_UTF16                  10
 #define PCRE_CONFIG_JITTARGET              11
+#define PCRE_CONFIG_UTF32                  12
 
 /* Request types for pcre_study(). Do not re-arrange, in order to remain
 compatible. */
@@ -261,7 +272,7 @@ compatible. */
 #define PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE   0x0004
 #define PCRE_STUDY_EXTRA_NEEDED               0x0008
 
-/* Bit flags for the pcre[16]_extra structure. Do not re-arrange or redefine
+/* Bit flags for the pcre[16|32]_extra structure. Do not re-arrange or redefine
 these bits, just add new ones on the end, in order to remain compatible. */
 
 #define PCRE_EXTRA_STUDY_DATA             0x0001
@@ -280,12 +291,18 @@ typedef struct real_pcre pcre;
 struct real_pcre16;               /* declaration; the definition is private  */
 typedef struct real_pcre16 pcre16;
 
+struct real_pcre32;               /* declaration; the definition is private  */
+typedef struct real_pcre32 pcre32;
+
 struct real_pcre_jit_stack;       /* declaration; the definition is private  */
 typedef struct real_pcre_jit_stack pcre_jit_stack;
 
 struct real_pcre16_jit_stack;     /* declaration; the definition is private  */
 typedef struct real_pcre16_jit_stack pcre16_jit_stack;
 
+struct real_pcre32_jit_stack;     /* declaration; the definition is private  */
+typedef struct real_pcre32_jit_stack pcre32_jit_stack;
+
 /* If PCRE is compiled with 16 bit character support, PCRE_UCHAR16 must contain
 a 16 bit wide signed data type. Otherwise it can be a dummy data type since
 pcre16 functions are not implemented. There is a check for this in pcre_internal.h. */
@@ -297,6 +314,17 @@ pcre16 functions are not implemented. There is a check for this in pcre_internal
 #define PCRE_SPTR16 const PCRE_UCHAR16 *
 #endif
 
+/* If PCRE is compiled with 32 bit character support, PCRE_UCHAR32 must contain
+a 32 bit wide signed data type. Otherwise it can be a dummy data type since
+pcre32 functions are not implemented. There is a check for this in pcre_internal.h. */
+#ifndef PCRE_UCHAR32
+#define PCRE_UCHAR32 unsigned int
+#endif
+
+#ifndef PCRE_SPTR32
+#define PCRE_SPTR32 const PCRE_UCHAR32 *
+#endif
+
 /* When PCRE is compiled as a C++ library, the subject pointer type can be
 replaced with a custom type. For conventional use, the public interface is a
 const char *. */
@@ -333,6 +361,19 @@ typedef struct pcre16_extra {
   void *executable_jit;           /* Contains a pointer to a compiled jit code */
 } pcre16_extra;
 
+/* Same structure as above, but with 32 bit char pointers. */
+
+typedef struct pcre32_extra {
+  unsigned long int flags;        /* Bits for which fields are set */
+  void *study_data;               /* Opaque data from pcre_study() */
+  unsigned long int match_limit;  /* Maximum number of calls to match() */
+  void *callout_data;             /* Data passed back in callouts */
+  const unsigned char *tables;    /* Pointer to character tables */
+  unsigned long int match_limit_recursion; /* Max recursive calls to match() */
+  PCRE_UCHAR32 **mark;            /* For passing back a mark pointer */
+  void *executable_jit;           /* Contains a pointer to a compiled jit code */
+} pcre32_extra;
+
 /* The structure for passing out data via the pcre_callout_function. We use a
 structure so that new fields can be added on the end in future versions,
 without changing the API of the function, thereby allowing old clients to work
@@ -380,6 +421,28 @@ typedef struct pcre16_callout_block {
   /* ------------------------------------------------------------------ */
 } pcre16_callout_block;
 
+/* Same structure as above, but with 32 bit char pointers. */
+
+typedef struct pcre32_callout_block {
+  int          version;           /* Identifies version of block */
+  /* ------------------------ Version 0 ------------------------------- */
+  int          callout_number;    /* Number compiled into pattern */
+  int         *offset_vector;     /* The offset vector */
+  PCRE_SPTR32  subject;           /* The subject being matched */
+  int          subject_length;    /* The length of the subject */
+  int          start_match;       /* Offset to start of this match attempt */
+  int          current_position;  /* Where we currently are in the subject */
+  int          capture_top;       /* Max current capture */
+  int          capture_last;      /* Most recently closed capture */
+  void        *callout_data;      /* Data passed in with the call */
+  /* ------------------- Added for Version 1 -------------------------- */
+  int          pattern_position;  /* Offset to next item in the pattern */
+  int          next_item_length;  /* Length of next item in the pattern */
+  /* ------------------- Added for Version 2 -------------------------- */
+  const PCRE_UCHAR32 *mark;       /* Pointer to current mark or NULL    */
+  /* ------------------------------------------------------------------ */
+} pcre32_callout_block;
+
 /* Indirection for store get and free functions. These can be set to
 alternative malloc/free functions if required. Special ones are used in the
 non-recursive case for "frames". There is also an optional callout function
@@ -398,6 +461,12 @@ PCRE_EXP_DECL void  (*pcre16_free)(void *);
 PCRE_EXP_DECL void *(*pcre16_stack_malloc)(size_t);
 PCRE_EXP_DECL void  (*pcre16_stack_free)(void *);
 PCRE_EXP_DECL int   (*pcre16_callout)(pcre16_callout_block *);
+
+PCRE_EXP_DECL void *(*pcre32_malloc)(size_t);
+PCRE_EXP_DECL void  (*pcre32_free)(void *);
+PCRE_EXP_DECL void *(*pcre32_stack_malloc)(size_t);
+PCRE_EXP_DECL void  (*pcre32_stack_free)(void *);
+PCRE_EXP_DECL int   (*pcre32_callout)(pcre32_callout_block *);
 #else   /* VPCOMPAT */
 PCRE_EXP_DECL void *pcre_malloc(size_t);
 PCRE_EXP_DECL void  pcre_free(void *);
@@ -410,12 +479,19 @@ PCRE_EXP_DECL void  pcre16_free(void *);
 PCRE_EXP_DECL void *pcre16_stack_malloc(size_t);
 PCRE_EXP_DECL void  pcre16_stack_free(void *);
 PCRE_EXP_DECL int   pcre16_callout(pcre16_callout_block *);
+
+PCRE_EXP_DECL void *pcre32_malloc(size_t);
+PCRE_EXP_DECL void  pcre32_free(void *);
+PCRE_EXP_DECL void *pcre32_stack_malloc(size_t);
+PCRE_EXP_DECL void  pcre32_stack_free(void *);
+PCRE_EXP_DECL int   pcre32_callout(pcre32_callout_block *);
 #endif  /* VPCOMPAT */
 
 /* User defined callback which provides a stack just before the match starts. */
 
 typedef pcre_jit_stack *(*pcre_jit_callback)(void *);
 typedef pcre16_jit_stack *(*pcre16_jit_callback)(void *);
+typedef pcre32_jit_stack *(*pcre32_jit_callback)(void *);
 
 /* Exported PCRE functions */
 
@@ -423,83 +499,122 @@ PCRE_EXP_DECL pcre *pcre_compile(const char *, int, const char **, int *,
                   const unsigned char *);
 PCRE_EXP_DECL pcre16 *pcre16_compile(PCRE_SPTR16, int, const char **, int *,
                   const unsigned char *);
+PCRE_EXP_DECL pcre32 *pcre32_compile(PCRE_SPTR32, int, const char **, int *,
+                  const unsigned char *);
 PCRE_EXP_DECL pcre *pcre_compile2(const char *, int, int *, const char **,
                   int *, const unsigned char *);
 PCRE_EXP_DECL pcre16 *pcre16_compile2(PCRE_SPTR16, int, int *, const char **,
                   int *, const unsigned char *);
+PCRE_EXP_DECL pcre32 *pcre32_compile2(PCRE_SPTR32, int, int *, const char **,
+                  int *, const unsigned char *);
 PCRE_EXP_DECL int  pcre_config(int, void *);
 PCRE_EXP_DECL int  pcre16_config(int, void *);
+PCRE_EXP_DECL int  pcre32_config(int, void *);
 PCRE_EXP_DECL int  pcre_copy_named_substring(const pcre *, const char *,
                   int *, int, const char *, char *, int);
 PCRE_EXP_DECL int  pcre16_copy_named_substring(const pcre16 *, PCRE_SPTR16,
                   int *, int, PCRE_SPTR16, PCRE_UCHAR16 *, int);
+PCRE_EXP_DECL int  pcre32_copy_named_substring(const pcre32 *, PCRE_SPTR32,
+                  int *, int, PCRE_SPTR32, PCRE_UCHAR32 *, int);
 PCRE_EXP_DECL int  pcre_copy_substring(const char *, int *, int, int,
                   char *, int);
 PCRE_EXP_DECL int  pcre16_copy_substring(PCRE_SPTR16, int *, int, int,
                   PCRE_UCHAR16 *, int);
+PCRE_EXP_DECL int  pcre32_copy_substring(PCRE_SPTR32, int *, int, int,
+                  PCRE_UCHAR32 *, int);
 PCRE_EXP_DECL int  pcre_dfa_exec(const pcre *, const pcre_extra *,
                   const char *, int, int, int, int *, int , int *, int);
 PCRE_EXP_DECL int  pcre16_dfa_exec(const pcre16 *, const pcre16_extra *,
                   PCRE_SPTR16, int, int, int, int *, int , int *, int);
+PCRE_EXP_DECL int  pcre32_dfa_exec(const pcre32 *, const pcre32_extra *,
+                  PCRE_SPTR32, int, int, int, int *, int , int *, int);
 PCRE_EXP_DECL int  pcre_exec(const pcre *, const pcre_extra *, PCRE_SPTR,
                    int, int, int, int *, int);
 PCRE_EXP_DECL int  pcre16_exec(const pcre16 *, const pcre16_extra *,
                    PCRE_SPTR16, int, int, int, int *, int);
+PCRE_EXP_DECL int  pcre32_exec(const pcre32 *, const pcre32_extra *,
+                   PCRE_SPTR32, int, int, int, int *, int);
 PCRE_EXP_DECL void pcre_free_substring(const char *);
 PCRE_EXP_DECL void pcre16_free_substring(PCRE_SPTR16);
+PCRE_EXP_DECL void pcre32_free_substring(PCRE_SPTR32);
 PCRE_EXP_DECL void pcre_free_substring_list(const char **);
 PCRE_EXP_DECL void pcre16_free_substring_list(PCRE_SPTR16 *);
+PCRE_EXP_DECL void pcre32_free_substring_list(PCRE_SPTR32 *);
 PCRE_EXP_DECL int  pcre_fullinfo(const pcre *, const pcre_extra *, int,
                   void *);
 PCRE_EXP_DECL int  pcre16_fullinfo(const pcre16 *, const pcre16_extra *, int,
                   void *);
+PCRE_EXP_DECL int  pcre32_fullinfo(const pcre32 *, const pcre32_extra *, int,
+                  void *);
 PCRE_EXP_DECL int  pcre_get_named_substring(const pcre *, const char *,
                   int *, int, const char *, const char **);
 PCRE_EXP_DECL int  pcre16_get_named_substring(const pcre16 *, PCRE_SPTR16,
                   int *, int, PCRE_SPTR16, PCRE_SPTR16 *);
+PCRE_EXP_DECL int  pcre32_get_named_substring(const pcre32 *, PCRE_SPTR32,
+                  int *, int, PCRE_SPTR32, PCRE_SPTR32 *);
 PCRE_EXP_DECL int  pcre_get_stringnumber(const pcre *, const char *);
 PCRE_EXP_DECL int  pcre16_get_stringnumber(const pcre16 *, PCRE_SPTR16);
+PCRE_EXP_DECL int  pcre32_get_stringnumber(const pcre32 *, PCRE_SPTR32);
 PCRE_EXP_DECL int  pcre_get_stringtable_entries(const pcre *, const char *,
                   char **, char **);
 PCRE_EXP_DECL int  pcre16_get_stringtable_entries(const pcre16 *, PCRE_SPTR16,
                   PCRE_UCHAR16 **, PCRE_UCHAR16 **);
+PCRE_EXP_DECL int  pcre32_get_stringtable_entries(const pcre32 *, PCRE_SPTR32,
+                  PCRE_UCHAR32 **, PCRE_UCHAR32 **);
 PCRE_EXP_DECL int  pcre_get_substring(const char *, int *, int, int,
                   const char **);
 PCRE_EXP_DECL int  pcre16_get_substring(PCRE_SPTR16, int *, int, int,
                   PCRE_SPTR16 *);
+PCRE_EXP_DECL int  pcre32_get_substring(PCRE_SPTR32, int *, int, int,
+                  PCRE_SPTR32 *);
 PCRE_EXP_DECL int  pcre_get_substring_list(const char *, int *, int,
                   const char ***);
 PCRE_EXP_DECL int  pcre16_get_substring_list(PCRE_SPTR16, int *, int,
                   PCRE_SPTR16 **);
+PCRE_EXP_DECL int  pcre32_get_substring_list(PCRE_SPTR32, int *, int,
+                  PCRE_SPTR32 **);
 PCRE_EXP_DECL const unsigned char *pcre_maketables(void);
 PCRE_EXP_DECL const unsigned char *pcre16_maketables(void);
+PCRE_EXP_DECL const unsigned char *pcre32_maketables(void);
 PCRE_EXP_DECL int  pcre_refcount(pcre *, int);
 PCRE_EXP_DECL int  pcre16_refcount(pcre16 *, int);
+PCRE_EXP_DECL int  pcre32_refcount(pcre32 *, int);
 PCRE_EXP_DECL pcre_extra *pcre_study(const pcre *, int, const char **);
 PCRE_EXP_DECL pcre16_extra *pcre16_study(const pcre16 *, int, const char **);
+PCRE_EXP_DECL pcre32_extra *pcre32_study(const pcre32 *, int, const char **);
 PCRE_EXP_DECL void pcre_free_study(pcre_extra *);
 PCRE_EXP_DECL void pcre16_free_study(pcre16_extra *);
+PCRE_EXP_DECL void pcre32_free_study(pcre32_extra *);
 PCRE_EXP_DECL const char *pcre_version(void);
 PCRE_EXP_DECL const char *pcre16_version(void);
+PCRE_EXP_DECL const char *pcre32_version(void);
 
 /* Utility functions for byte order swaps. */
 PCRE_EXP_DECL int  pcre_pattern_to_host_byte_order(pcre *, pcre_extra *,
                   const unsigned char *);
 PCRE_EXP_DECL int  pcre16_pattern_to_host_byte_order(pcre16 *, pcre16_extra *,
                   const unsigned char *);
+PCRE_EXP_DECL int  pcre32_pattern_to_host_byte_order(pcre32 *, pcre32_extra *,
+                  const unsigned char *);
 PCRE_EXP_DECL int  pcre16_utf16_to_host_byte_order(PCRE_UCHAR16 *,
                   PCRE_SPTR16, int, int *, int);
+PCRE_EXP_DECL int  pcre32_utf32_to_host_byte_order(PCRE_UCHAR32 *,
+                  PCRE_SPTR32, int, int *, int);
 
 /* JIT compiler related functions. */
 
 PCRE_EXP_DECL pcre_jit_stack *pcre_jit_stack_alloc(int, int);
 PCRE_EXP_DECL pcre16_jit_stack *pcre16_jit_stack_alloc(int, int);
+PCRE_EXP_DECL pcre32_jit_stack *pcre32_jit_stack_alloc(int, int);
 PCRE_EXP_DECL void pcre_jit_stack_free(pcre_jit_stack *);
 PCRE_EXP_DECL void pcre16_jit_stack_free(pcre16_jit_stack *);
+PCRE_EXP_DECL void pcre32_jit_stack_free(pcre32_jit_stack *);
 PCRE_EXP_DECL void pcre_assign_jit_stack(pcre_extra *,
                   pcre_jit_callback, void *);
 PCRE_EXP_DECL void pcre16_assign_jit_stack(pcre16_extra *,
                   pcre16_jit_callback, void *);
+PCRE_EXP_DECL void pcre32_assign_jit_stack(pcre32_extra *,
+                  pcre32_jit_callback, void *);
 
 #ifdef __cplusplus
 }  /* extern "C" */