summaryrefslogtreecommitdiff
path: root/src/regparse.h
diff options
context:
space:
mode:
authorJörg Frings-Fürst <debian@jff.email>2020-04-21 06:47:36 +0200
committerJörg Frings-Fürst <debian@jff.email>2020-04-21 06:47:36 +0200
commita792c4308d32f68f34131ff89e124190e4513c38 (patch)
tree68d5ba9857cafd597ea84011077a83e43f2d1620 /src/regparse.h
parent9e629c8f43b43617fa5b7d3654f7d81e81b8a427 (diff)
parentd1ffcb88ee95aded1bc4aef56f4f39951ad03ceb (diff)
Merge branch 'release/debian/6.9.5-1'debian/6.9.5-1
Diffstat (limited to 'src/regparse.h')
-rw-r--r--src/regparse.h180
1 files changed, 95 insertions, 85 deletions
diff --git a/src/regparse.h b/src/regparse.h
index 1525ccb..979e982 100644
--- a/src/regparse.h
+++ b/src/regparse.h
@@ -4,7 +4,7 @@
regparse.h - Oniguruma (regular expression library)
**********************************************************************/
/*-
- * Copyright (c) 2002-2019 K.Kosako
+ * Copyright (c) 2002-2020 K.Kosako
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
@@ -31,23 +31,23 @@
#include "regint.h"
-#define NODE_STRING_MARGIN 16
-#define NODE_STRING_BUF_SIZE 20 /* sizeof(CClassNode) - sizeof(int)*4 */
-#define NODE_BACKREFS_SIZE 6
+#define NODE_STRING_MARGIN 16
+#define NODE_STRING_BUF_SIZE 24 /* sizeof(CClassNode) - sizeof(int)*4 */
+#define NODE_BACKREFS_SIZE 6
/* node type */
typedef enum {
- NODE_STRING = 0,
- NODE_CCLASS = 1,
- NODE_CTYPE = 2,
- NODE_BACKREF = 3,
- NODE_QUANT = 4,
- NODE_BAG = 5,
- NODE_ANCHOR = 6,
- NODE_LIST = 7,
- NODE_ALT = 8,
- NODE_CALL = 9,
- NODE_GIMMICK = 10
+ NODE_STRING = 0,
+ NODE_CCLASS = 1,
+ NODE_CTYPE = 2,
+ NODE_BACKREF = 3,
+ NODE_QUANT = 4,
+ NODE_BAG = 5,
+ NODE_ANCHOR = 6,
+ NODE_LIST = 7,
+ NODE_ALT = 8,
+ NODE_CALL = 9,
+ NODE_GIMMICK = 10
} NodeType;
enum BagType {
@@ -67,10 +67,10 @@ enum GimmickType {
};
enum BodyEmptyType {
- BODY_IS_NOT_EMPTY = 0,
- BODY_IS_EMPTY_POSSIBILITY = 1,
- BODY_IS_EMPTY_POSSIBILITY_MEM = 2,
- BODY_IS_EMPTY_POSSIBILITY_REC = 3
+ BODY_IS_NOT_EMPTY = 0,
+ BODY_MAY_BE_EMPTY = 1,
+ BODY_MAY_BE_EMPTY_MEM = 2,
+ BODY_MAY_BE_EMPTY_REC = 3
};
struct _Node;
@@ -84,8 +84,7 @@ typedef struct {
UChar* end;
unsigned int flag;
UChar buf[NODE_STRING_BUF_SIZE];
- int capacity; /* (allocated size - 1) or 0: use buf[] */
- int case_min_len;
+ int capacity; /* (allocated size - 1) or 0: use buf[] */
} StrNode;
typedef struct {
@@ -110,7 +109,7 @@ typedef struct {
enum BodyEmptyType emptiness;
struct _Node* head_exact;
struct _Node* next_head_exact;
- int include_referred; /* include called node. don't eliminate even if {0} */
+ int include_referred; /* include called node. don't eliminate even if {0} */
} QuantNode;
typedef struct {
@@ -139,7 +138,8 @@ typedef struct {
/* for multiple call reference */
OnigLen min_len; /* min length (byte) */
OnigLen max_len; /* max length (byte) */
- int char_len; /* character length */
+ OnigLen min_char_len;
+ OnigLen max_char_len;
int opt_count; /* referenced count in optimize_nodes() */
} BagNode;
@@ -189,8 +189,10 @@ typedef struct {
struct _Node* body;
int type;
- int char_len;
+ OnigLen char_min_len;
+ OnigLen char_max_len;
int ascii_mode;
+ struct _Node* lead_node;
} AnchorNode;
typedef struct {
@@ -209,7 +211,6 @@ typedef struct {
int ctype;
int not;
- OnigOptionType options;
int ascii_mode;
} CtypeNode;
@@ -248,6 +249,11 @@ typedef struct _Node {
} u;
} Node;
+typedef struct {
+ int new_val;
+} GroupNumMap;
+
+
#define NULL_NODE ((Node* )0)
@@ -280,62 +286,59 @@ typedef struct _Node {
#define CALL_(node) (&((node)->u.call))
#define GIMMICK_(node) (&((node)->u.gimmick))
-#define NODE_CAR(node) (CONS_(node)->car)
-#define NODE_CDR(node) (CONS_(node)->cdr)
+#define NODE_CAR(node) (CONS_(node)->car)
+#define NODE_CDR(node) (CONS_(node)->cdr)
#define CTYPE_ANYCHAR -1
#define NODE_IS_ANYCHAR(node) \
(NODE_TYPE(node) == NODE_CTYPE && CTYPE_(node)->ctype == CTYPE_ANYCHAR)
-#define CTYPE_OPTION(node, reg) \
- (NODE_IS_FIXED_OPTION(node) ? CTYPE_(node)->options : reg->options)
-
#define ANCR_ANYCHAR_INF_MASK (ANCR_ANYCHAR_INF | ANCR_ANYCHAR_INF_ML)
#define ANCR_END_BUF_MASK (ANCR_END_BUF | ANCR_SEMI_END_BUF)
-#define NODE_STRING_CRUDE (1<<0)
-#define NODE_STRING_CASE_EXPANDED (1<<1)
-#define NODE_STRING_CASE_FOLD_MATCH (1<<2)
+#define NODE_STRING_CRUDE (1<<0)
+#define NODE_STRING_CASE_EXPANDED (1<<1)
#define NODE_STRING_LEN(node) (int )((node)->u.str.end - (node)->u.str.s)
#define NODE_STRING_SET_CRUDE(node) (node)->u.str.flag |= NODE_STRING_CRUDE
#define NODE_STRING_CLEAR_CRUDE(node) (node)->u.str.flag &= ~NODE_STRING_CRUDE
#define NODE_STRING_SET_CASE_EXPANDED(node) (node)->u.str.flag |= NODE_STRING_CASE_EXPANDED
-#define NODE_STRING_SET_CASE_FOLD_MATCH(node) (node)->u.str.flag |= NODE_STRING_CASE_FOLD_MATCH
#define NODE_STRING_IS_CRUDE(node) \
(((node)->u.str.flag & NODE_STRING_CRUDE) != 0)
#define NODE_STRING_IS_CASE_EXPANDED(node) \
(((node)->u.str.flag & NODE_STRING_CASE_EXPANDED) != 0)
-#define NODE_STRING_IS_CASE_FOLD_MATCH(node) \
- (((node)->u.str.flag & NODE_STRING_CASE_FOLD_MATCH) != 0)
#define BACKREFS_P(br) \
(IS_NOT_NULL((br)->back_dynamic) ? (br)->back_dynamic : (br)->back_static)
/* node status bits */
-#define NODE_ST_MIN_FIXED (1<<0)
-#define NODE_ST_MAX_FIXED (1<<1)
-#define NODE_ST_CLEN_FIXED (1<<2)
-#define NODE_ST_MARK1 (1<<3)
-#define NODE_ST_MARK2 (1<<4)
-#define NODE_ST_STRICT_REAL_REPEAT (1<<5)
-#define NODE_ST_RECURSION (1<<6)
-#define NODE_ST_CALLED (1<<7)
-#define NODE_ST_ADDR_FIXED (1<<8)
-#define NODE_ST_NAMED_GROUP (1<<9)
-#define NODE_ST_IN_REAL_REPEAT (1<<10) /* STK_REPEAT is nested in stack. */
-#define NODE_ST_IN_ZERO_REPEAT (1<<11) /* (....){0} */
-#define NODE_ST_IN_MULTI_ENTRY (1<<12)
-#define NODE_ST_NEST_LEVEL (1<<13)
-#define NODE_ST_BY_NUMBER (1<<14) /* {n,m} */
-#define NODE_ST_BY_NAME (1<<15) /* backref by name */
-#define NODE_ST_BACKREF (1<<16)
-#define NODE_ST_CHECKER (1<<17)
-#define NODE_ST_FIXED_OPTION (1<<18)
-#define NODE_ST_PROHIBIT_RECURSION (1<<19)
-#define NODE_ST_SUPER (1<<20)
-#define NODE_ST_EMPTY_STATUS_CHECK (1<<21)
+#define NODE_ST_FIXED_MIN (1<<0)
+#define NODE_ST_FIXED_MAX (1<<1)
+#define NODE_ST_FIXED_CLEN (1<<2)
+#define NODE_ST_MARK1 (1<<3)
+#define NODE_ST_MARK2 (1<<4)
+#define NODE_ST_STRICT_REAL_REPEAT (1<<5)
+#define NODE_ST_RECURSION (1<<6)
+#define NODE_ST_CALLED (1<<7)
+#define NODE_ST_FIXED_ADDR (1<<8)
+#define NODE_ST_NAMED_GROUP (1<<9)
+#define NODE_ST_IN_REAL_REPEAT (1<<10) /* STK_REPEAT is nested in stack. */
+#define NODE_ST_IN_ZERO_REPEAT (1<<11) /* (....){0} */
+#define NODE_ST_IN_MULTI_ENTRY (1<<12)
+#define NODE_ST_NEST_LEVEL (1<<13)
+#define NODE_ST_BY_NUMBER (1<<14) /* {n,m} */
+#define NODE_ST_BY_NAME (1<<15) /* backref by name */
+#define NODE_ST_BACKREF (1<<16)
+#define NODE_ST_CHECKER (1<<17)
+#define NODE_ST_PROHIBIT_RECURSION (1<<18)
+#define NODE_ST_SUPER (1<<19)
+#define NODE_ST_EMPTY_STATUS_CHECK (1<<20)
+#define NODE_ST_IGNORECASE (1<<21)
+#define NODE_ST_MULTILINE (1<<22)
+#define NODE_ST_TEXT_SEGMENT_WORD (1<<23)
+#define NODE_ST_ABSENT_WITH_SIDE_EFFECTS (1<<24) /* stopper or clear */
+#define NODE_ST_FIXED_CLEN_MIN_SURE (1<<25)
#define NODE_STATUS(node) (((Node* )node)->u.base.status)
@@ -349,17 +352,16 @@ typedef struct _Node {
#define NODE_IS_RECURSION(node) ((NODE_STATUS(node) & NODE_ST_RECURSION) != 0)
#define NODE_IS_IN_ZERO_REPEAT(node) ((NODE_STATUS(node) & NODE_ST_IN_ZERO_REPEAT) != 0)
#define NODE_IS_NAMED_GROUP(node) ((NODE_STATUS(node) & NODE_ST_NAMED_GROUP) != 0)
-#define NODE_IS_ADDR_FIXED(node) ((NODE_STATUS(node) & NODE_ST_ADDR_FIXED) != 0)
-#define NODE_IS_CLEN_FIXED(node) ((NODE_STATUS(node) & NODE_ST_CLEN_FIXED) != 0)
-#define NODE_IS_MIN_FIXED(node) ((NODE_STATUS(node) & NODE_ST_MIN_FIXED) != 0)
-#define NODE_IS_MAX_FIXED(node) ((NODE_STATUS(node) & NODE_ST_MAX_FIXED) != 0)
+#define NODE_IS_FIXED_ADDR(node) ((NODE_STATUS(node) & NODE_ST_FIXED_ADDR) != 0)
+#define NODE_IS_FIXED_CLEN(node) ((NODE_STATUS(node) & NODE_ST_FIXED_CLEN) != 0)
+#define NODE_IS_FIXED_MIN(node) ((NODE_STATUS(node) & NODE_ST_FIXED_MIN) != 0)
+#define NODE_IS_FIXED_MAX(node) ((NODE_STATUS(node) & NODE_ST_FIXED_MAX) != 0)
#define NODE_IS_MARK1(node) ((NODE_STATUS(node) & NODE_ST_MARK1) != 0)
#define NODE_IS_MARK2(node) ((NODE_STATUS(node) & NODE_ST_MARK2) != 0)
#define NODE_IS_NEST_LEVEL(node) ((NODE_STATUS(node) & NODE_ST_NEST_LEVEL) != 0)
#define NODE_IS_BY_NAME(node) ((NODE_STATUS(node) & NODE_ST_BY_NAME) != 0)
#define NODE_IS_BACKREF(node) ((NODE_STATUS(node) & NODE_ST_BACKREF) != 0)
#define NODE_IS_CHECKER(node) ((NODE_STATUS(node) & NODE_ST_CHECKER) != 0)
-#define NODE_IS_FIXED_OPTION(node) ((NODE_STATUS(node) & NODE_ST_FIXED_OPTION) != 0)
#define NODE_IS_SUPER(node) ((NODE_STATUS(node) & NODE_ST_SUPER) != 0)
#define NODE_IS_PROHIBIT_RECURSION(node) \
((NODE_STATUS(node) & NODE_ST_PROHIBIT_RECURSION) != 0)
@@ -367,6 +369,11 @@ typedef struct _Node {
((NODE_STATUS(node) & NODE_ST_STRICT_REAL_REPEAT) != 0)
#define NODE_IS_EMPTY_STATUS_CHECK(node) \
((NODE_STATUS(node) & NODE_ST_EMPTY_STATUS_CHECK) != 0)
+#define NODE_IS_IGNORECASE(node) ((NODE_STATUS(node) & NODE_ST_IGNORECASE) != 0)
+#define NODE_IS_MULTILINE(node) ((NODE_STATUS(node) & NODE_ST_MULTILINE) != 0)
+#define NODE_IS_TEXT_SEGMENT_WORD(node) ((NODE_STATUS(node) & NODE_ST_TEXT_SEGMENT_WORD) != 0)
+#define NODE_IS_ABSENT_WITH_SIDE_EFFECTS(node) ((NODE_STATUS(node) & NODE_ST_ABSENT_WITH_SIDE_EFFECTS) != 0)
+#define NODE_IS_FIXED_CLEN_MIN_SURE(node) ((NODE_STATUS(node) & NODE_ST_FIXED_CLEN_MIN_SURE) != 0)
#define NODE_PARENT(node) ((node)->u.base.parent)
#define NODE_BODY(node) ((node)->u.base.body)
@@ -375,11 +382,20 @@ typedef struct _Node {
#define NODE_CALL_BODY(node) ((node)->body)
#define NODE_ANCHOR_BODY(node) ((node)->body)
-#define SCANENV_MEMENV_SIZE 8
+#define SCANENV_MEMENV_SIZE 8
#define SCANENV_MEMENV(senv) \
(IS_NOT_NULL((senv)->mem_env_dynamic) ? \
(senv)->mem_env_dynamic : (senv)->mem_env_static)
+#define IS_SYNTAX_OP(syn, opm) (((syn)->op & (opm)) != 0)
+#define IS_SYNTAX_OP2(syn, opm) (((syn)->op2 & (opm)) != 0)
+#define IS_SYNTAX_BV(syn, bvm) (((syn)->behavior & (bvm)) != 0)
+
+#define ID_ENTRY(env, id) do {\
+ id = (env)->id_num++;\
+} while(0)
+
+
typedef struct {
Node* mem_node;
Node* empty_repeat_node;
@@ -403,50 +419,44 @@ typedef struct {
UChar* error_end;
regex_t* reg; /* for reg->names only */
int num_call;
-#ifdef USE_CALL
- UnsetAddrList* unset_addr_list;
- int has_call_zero;
-#endif
int num_mem;
int num_named;
int mem_alloc;
MemEnv mem_env_static[SCANENV_MEMENV_SIZE];
MemEnv* mem_env_dynamic;
+ int backref_num;
+ int keep_num;
+ int id_num;
+ int save_alloc_num;
+ SaveItem* saves;
+#ifdef USE_CALL
+ UnsetAddrList* unset_addr_list;
+ int has_call_zero;
+#endif
unsigned int parse_depth;
#ifdef ONIG_DEBUG_PARSE
unsigned int max_parse_depth;
#endif
- int backref_num;
- int keep_num;
- int save_num;
- int save_alloc_num;
- SaveItem* saves;
} ScanEnv;
-#define IS_SYNTAX_OP(syn, opm) (((syn)->op & (opm)) != 0)
-#define IS_SYNTAX_OP2(syn, opm) (((syn)->op2 & (opm)) != 0)
-#define IS_SYNTAX_BV(syn, bvm) (((syn)->behavior & (bvm)) != 0)
-
-typedef struct {
- int new_val;
-} GroupNumRemap;
-
-extern int onig_renumber_name_table P_((regex_t* reg, GroupNumRemap* map));
+extern int onig_renumber_name_table P_((regex_t* reg, GroupNumMap* map));
extern int onig_strncmp P_((const UChar* s1, const UChar* s2, int n));
extern void onig_strcpy P_((UChar* dest, const UChar* src, const UChar* end));
extern void onig_scan_env_set_error_string P_((ScanEnv* env, int ecode, UChar* arg, UChar* arg_end));
extern int onig_reduce_nested_quantifier P_((Node* pnode));
+extern int onig_node_copy(Node** rcopy, Node* from);
extern int onig_node_str_cat P_((Node* node, const UChar* s, const UChar* end));
-extern int onig_node_str_set P_((Node* node, const UChar* s, const UChar* end));
+extern int onig_node_str_set P_((Node* node, const UChar* s, const UChar* end, int need_free));
+extern void onig_node_str_clear P_((Node* node, int need_free));
extern void onig_node_free P_((Node* node));
+extern int onig_node_reset_empty P_((Node* node));
+extern int onig_node_reset_fail P_((Node* node));
extern Node* onig_node_new_bag P_((enum BagType type));
-extern Node* onig_node_new_anchor P_((int type, int ascii_mode));
extern Node* onig_node_new_str P_((const UChar* s, const UChar* end));
extern Node* onig_node_new_list P_((Node* left, Node* right));
extern Node* onig_node_new_alt P_((Node* left, Node* right));
-extern void onig_node_str_clear P_((Node* node));
extern int onig_names_free P_((regex_t* reg));
extern int onig_parse_tree P_((Node** root, const UChar* pattern, const UChar* end, regex_t* reg, ScanEnv* env));
extern int onig_free_shared_cclass_table P_((void));