X-Git-Url: https://perl5.git.perl.org/perl5.git/blobdiff_plain/8c9b4e639b8164433ff657146b42306a354ce3cf..4bab39bc1904f776c12d31a54ff5abe06fc9c103:/regcomp.h diff --git a/regcomp.h b/regcomp.h index 31b8d5e..520e60e 100644 --- a/regcomp.h +++ b/regcomp.h @@ -7,6 +7,10 @@ * License or the Artistic License, as specified in the README file. * */ + +#ifndef PERL_REGCOMP_H_ +#define PERL_REGCOMP_H_ + #include "regcharclass.h" /* Convert branch sequences to more efficient trie ops? */ @@ -90,8 +94,6 @@ /* This is the stuff that used to live in regexp.h that was truly private to the engine itself. It now lives here. */ - - typedef struct regexp_internal { int name_list_idx; /* Optional data index of an array of paren names */ union { @@ -109,8 +111,7 @@ Used to make it easier to clone and free arbitrary data that the regops need. Often the ARG field of a regop is an index into this structure */ - struct reg_code_block *code_blocks;/* positions of literal (?{}) */ - int num_code_blocks; /* size of code_blocks[] */ + struct reg_code_blocks *code_blocks;/* positions of literal (?{}) */ regnode program[1]; /* Unwarranted chumminess with compiler. */ } regexp_internal; @@ -141,6 +142,7 @@ #define PREGf_ANCH_MBOL 0x00000400 #define PREGf_ANCH_SBOL 0x00000800 #define PREGf_ANCH_GPOS 0x00001000 +#define PREGf_RECURSE_SEEN 0x00002000 #define PREGf_ANCH \ ( PREGf_ANCH_SBOL | PREGf_ANCH_GPOS | PREGf_ANCH_MBOL ) @@ -154,6 +156,14 @@ struct regnode_string { char string[1]; }; +struct regnode_lstring { /* Constructed this way to keep the string aligned. */ + U8 flags; + U8 type; + U16 next_off; + U32 str_len; /* Only 16 bits allowed before would overflow 'next_off' */ + char string[1]; +}; + /* Argument bearing node - workhorse, arg1 is often for the data field */ struct regnode_1 { @@ -192,7 +202,7 @@ struct regnode_2 { * Cyrillic, Greek, Hebrew, Indian subcontinent, Latin, and Thai; but not Han, * Japanese, nor Korean. (The regarglen structure in regnodes.h is a U8, and * the trie types TRIEC and AHOCORASICKC are larger than U8 for shift values - * below above 12.) Be sure to benchmark before changing, as larger sizes do + * above 12.) Be sure to benchmark before changing, as larger sizes do * significantly slow down the test suite */ #define NUM_ANYOF_CODE_POINTS (1 << 8) @@ -219,7 +229,7 @@ struct regnode_charclass { }; /* has runtime (locale) \d, \w, ..., [:posix:] classes */ -struct regnode_charclass_class { +struct regnode_charclass_posixl { U8 flags; /* ANYOF_MATCHES_POSIXL bit must go here */ U8 type; U16 next_off; @@ -267,22 +277,22 @@ struct regnode_ssc { Impose a limit of REG_INFTY on various pattern matching operations to limit stack growth and to avoid "infinite" recursions. */ -/* The default size for REG_INFTY is I16_MAX, which is the same as - SHORT_MAX (see perl.h). Unfortunately I16 isn't necessarily 16 bits - (see handy.h). On the Cray C90, sizeof(short)==4 and hence I16_MAX is - ((1<<31)-1), while on the Cray T90, sizeof(short)==8 and I16_MAX is - ((1<<63)-1). To limit stack growth to reasonable sizes, supply a +/* The default size for REG_INFTY is U16_MAX, which is the same as + USHORT_MAX (see perl.h). Unfortunately U16 isn't necessarily 16 bits + (see handy.h). On the Cray C90, sizeof(short)==4 and hence U16_MAX is + ((1<<32)-1), while on the Cray T90, sizeof(short)==8 and U16_MAX is + ((1<<64)-1). To limit stack growth to reasonable sizes, supply a smaller default. --Andy Dougherty 11 June 1998 */ #if SHORTSIZE > 2 # ifndef REG_INFTY -# define REG_INFTY ((1<<15)-1) +# define REG_INFTY ((1<<16)-1) # endif #endif #ifndef REG_INFTY -# define REG_INFTY I16_MAX +# define REG_INFTY U16_MAX #endif #define ARG_VALUE(arg) (arg) @@ -322,19 +332,57 @@ struct regnode_ssc { #undef OP #undef OPERAND -#undef MASK #undef STRING #define OP(p) ((p)->type) #define FLAGS(p) ((p)->flags) /* Caution: Doesn't apply to all \ regnode types. For some, it's the \ character set of the regnode */ -#define OPERAND(p) (((struct regnode_string *)p)->string) -#define MASK(p) ((char*)OPERAND(p)) -#define STR_LEN(p) (((struct regnode_string *)p)->str_len) -#define STRING(p) (((struct regnode_string *)p)->string) -#define STR_SZ(l) ((l + sizeof(regnode) - 1) / sizeof(regnode)) -#define NODE_SZ_STR(p) (STR_SZ(STR_LEN(p))+1) +#define STR_LENs(p) (__ASSERT_(OP(p) != LEXACT && OP(p) != LEXACT_ONLY8) \ + ((struct regnode_string *)p)->str_len) +#define STRINGs(p) (__ASSERT_(OP(p) != LEXACT && OP(p) != LEXACT_ONLY8) \ + ((struct regnode_string *)p)->string) +#define OPERANDs(p) STRINGs(p) + +/* Long strings. Currently limited to length 18 bits, which handles a 262000 + * byte string. The limiting factor is the 16 bit 'next_off' field, which + * points to the next regnode, so the furthest away it can be is 2**16. On + * most architectures, regnodes are 2**2 bytes long, so that yields 2**18 + * bytes. Should a longer string be desired, we could increase it to 26 bits + * fairly easily, by changing this node to have longj type which causes the ARG + * field to be used for the link to the next regnode (although code would have + * to be changed to account for this), and then use a combination of the flags + * and next_off fields for the length. To get 34 bit length, also change the + * node to be an ARG2L, using the second 32 bit field for the length, and not + * using the flags nor next_off fields at all. One could have an llstring node + * and even an lllstring type. */ +#define STR_LENl(p) (__ASSERT_(OP(p) == LEXACT || OP(p) == LEXACT_ONLY8) \ + (((struct regnode_lstring *)p)->str_len)) +#define STRINGl(p) (__ASSERT_(OP(p) == LEXACT || OP(p) == LEXACT_ONLY8) \ + (((struct regnode_lstring *)p)->string)) +#define OPERANDl(p) STRINGl(p) + +#define STR_LEN(p) ((OP(p) == LEXACT || OP(p) == LEXACT_ONLY8) \ + ? STR_LENl(p) : STR_LENs(p)) +#define STRING(p) ((OP(p) == LEXACT || OP(p) == LEXACT_ONLY8) \ + ? STRINGl(p) : STRINGs(p)) +#define OPERAND(p) STRING(p) + +/* The number of (smallest) regnode equivalents that a string of length l bytes + * occupies */ +#define STR_SZ(l) (((l) + sizeof(regnode) - 1) / sizeof(regnode)) + +/* The number of (smallest) regnode equivalents that the EXACTISH node 'p' + * occupies */ +#define NODE_SZ_STR(p) (STR_SZ(STR_LEN(p)) + 1 + regarglen[(p)->type]) + +#define setSTR_LEN(p,v) \ + STMT_START{ \ + if (OP(p) == LEXACT || OP(p) == LEXACT_ONLY8) \ + ((struct regnode_lstring *)(p))->str_len = (v); \ + else \ + ((struct regnode_string *)(p))->str_len = (v); \ + } STMT_END #undef NODE_ALIGN #undef ARG_LOC @@ -355,71 +403,134 @@ struct regnode_ssc { #define NEXTOPER(p) ((p) + NODE_STEP_REGNODE) #define PREVOPER(p) ((p) - NODE_STEP_REGNODE) -#define FILL_ADVANCE_NODE(ptr, op) STMT_START { \ - (ptr)->type = op; (ptr)->next_off = 0; (ptr)++; } STMT_END -#define FILL_ADVANCE_NODE_ARG(ptr, op, arg) STMT_START { \ - ARG_SET(ptr, arg); FILL_ADVANCE_NODE(ptr, op); (ptr) += 1; } STMT_END -#define FILL_ADVANCE_NODE_2L_ARG(ptr, op, arg1, arg2) \ - STMT_START { \ - ARG_SET(ptr, arg1); \ - ARG2L_SET(ptr, arg2); \ - FILL_ADVANCE_NODE(ptr, op); \ - (ptr) += 2; \ - } STMT_END +#define FILL_NODE(offset, op) \ + STMT_START { \ + OP(REGNODE_p(offset)) = op; \ + NEXT_OFF(REGNODE_p(offset)) = 0; \ + } STMT_END +#define FILL_ADVANCE_NODE(offset, op) \ + STMT_START { \ + FILL_NODE(offset, op); \ + (offset)++; \ + } STMT_END +#define FILL_ADVANCE_NODE_ARG(offset, op, arg) \ + STMT_START { \ + ARG_SET(REGNODE_p(offset), arg); \ + FILL_ADVANCE_NODE(offset, op); \ + /* This is used generically for other operations \ + * that have a longer argument */ \ + (offset) += regarglen[op]; \ + } STMT_END +#define FILL_ADVANCE_NODE_2L_ARG(offset, op, arg1, arg2) \ + STMT_START { \ + ARG_SET(REGNODE_p(offset), arg1); \ + ARG2L_SET(REGNODE_p(offset), arg2); \ + FILL_ADVANCE_NODE(offset, op); \ + (offset) += 2; \ + } STMT_END #define REG_MAGIC 0234 -#define SIZE_ONLY (RExC_emit == (regnode *) & RExC_emit_dummy) -#define PASS1 SIZE_ONLY -#define PASS2 (! SIZE_ONLY) - -/* If the bitmap fully represents what this ANYOF node can match, the - * ARG is set to this special value (since 0, 1, ... are legal, but will never - * reach this high). */ +/* An ANYOF node is basically a bitmap with the index being a code point. If + * the bit for that code point is 1, the code point matches; if 0, it doesn't + * match (complemented if inverted). There is an additional mechanism to deal + * with cases where the bitmap is insufficient in and of itself. This #define + * indicates if the bitmap does fully represent what this ANYOF node can match. + * The ARG is set to this special value (since 0, 1, ... are legal, but will + * never reach this high). */ #define ANYOF_ONLY_HAS_BITMAP ((U32) -1) -/* Below are the flags for node->flags of ANYOF. These are in short supply, - * with none currently available. The ABOVE_BITMAP_ALL bit could be freed up - * by resorting to creating a swash containing everything above 255. This - * seems likely to introduce a performance penalty (but actual numbers haven't - * been done), so its probably better do some of the other possibilities below - * in preference to this. +/* When the bitmap isn't completely sufficient for handling the ANYOF node, + * flags (in node->flags of the ANYOF node) get set to indicate this. These + * are perennially in short supply. Beyond several cases where warnings need + * to be raised under certain circumstances, currently, there are six cases + * where the bitmap alone isn't sufficient. We could use six flags to + * represent the 6 cases, but to save flags bits, we play some games. The + * cases are: + * + * 1) The bitmap has a compiled-in very finite size. So something else needs + * to be used to specify if a code point that is too large for the bitmap + * actually matches. The mechanism currently is an inversion + * list. ANYOF_ONLY_HAS_BITMAP, described above, being TRUE indicates + * there are no matches of too-large code points. But if it is FALSE, + * then almost certainly there are matches too large for the bitmap. (The + * other cases, described below, either imply this one or are extremely + * rare in practice.) So we can just assume that a too-large code point + * will need something beyond the bitmap if ANYOF_ONLY_HAS_BITMAP is + * FALSE, instead of having a separate flag for this. + * 2) A subset of item 1) is if all possible code points outside the bitmap + * match. This is a common occurrence when the class is complemented, + * like /[^ij]/. Therefore a bit is reserved to indicate this, + * rather than having a more expensive inversion list created, + * ANYOF_MATCHES_ALL_ABOVE_BITMAP. + * 3) Under /d rules, it can happen that code points that are in the upper + * latin1 range (\x80-\xFF or their equivalents on EBCDIC platforms) match + * only if the runtime target string being matched against is UTF-8. For + * example /[\w[:punct:]]/d. This happens only for posix classes (with a + * couple of exceptions, like \d where it doesn't happen), and all such + * ones also have above-bitmap matches. Thus, 3) implies 1) as well. + * Note that /d rules are no longer encouraged; 'use 5.14' or higher + * deselects them. But a flag is required so that they can be properly + * handled. But it can be a shared flag: see 5) below. + * 4) Also under /d rules, something like /[\Wfoo]/ will match everything in + * the \x80-\xFF range, unless the string being matched against is UTF-8. + * An inversion list could be created for this case, but this is + * relatively common, and it turns out that it's all or nothing: if any + * one of these code points matches, they all do. Hence a single bit + * suffices. We use a shared flag that doesn't take up space by itself: + * ANYOF_SHARED_d_MATCHES_ALL_NON_UTF8_NON_ASCII_non_d_WARN_SUPER. This + * also implies 1), with one exception: [:^cntrl:]. + * 5) A user-defined \p{} property may not have been defined by the time the + * regex is compiled. In this case, we don't know until runtime what it + * will match, so we have to assume it could match anything, including + * code points that ordinarily would be in the bitmap. A flag bit is + * necessary to indicate this , though it can be shared with the item 3) + * flag, as that only occurs under /d, and this only occurs under non-d. + * This case is quite uncommon in the field, and the /(?[ ...])/ construct + * is a better way to accomplish what this feature does. This case also + * implies 1). + * ANYOF_SHARED_d_UPPER_LATIN1_UTF8_STRING_MATCHES_non_d_RUNTIME_USER_PROP + * is the shared flag. + * 6) /[foo]/il may have folds that are only valid if the runtime locale is a + * UTF-8 one. These are quite rare, so it would be good to avoid the + * expense of looking for them. But /l matching is slow anyway, and we've + * traditionally not worried too much about its performance. And this + * condition requires the ANYOFL_FOLD flag to be set, so testing for + * that flag would be sufficient to rule out most cases of this. So it is + * unclear if this should have a flag or not. But, this flag can be + * shared with another, so it doesn't occupy extra space. + * + * At the moment, there is one spare bit, but this could be increased by + * various tricks: + * + * If just one more bit is needed, as of this writing it seems to khw that the + * best choice would be to make ANYOF_MATCHES_ALL_ABOVE_BITMAP not a flag, but + * something like + * + * #define ANYOF_MATCHES_ALL_ABOVE_BITMAP ((U32) -2) + * + * and access it through the ARG like ANYOF_ONLY_HAS_BITMAP is. This flag is + * used by all ANYOF node types, and it could be used to avoid calling the + * handler function, as the macro REGINCLASS in regexec.c does now for other + * cases. * - * If just one bit is required, it seems to me (khw) that the best option would - * be to turn the ANYOF_LOC_REQ_UTF8 bit into a separate node type: a - * specialization of the ANYOFL type, freeing up the currently occupied bit. - * When turning a bit into a node type, one has to take into consideration that - * a SSC may use that bit -- not just a regular ANYOF[DL]?. In the case of - * ANYOF_LOC_REQ_UTF8, the only likely problem is accurately settting the SSC - * node-type to the new one, which would likely involve S_ssc_or and S_ssc_and, - * and not how the SSC currently gets set to ANYOFL. This bit is a natural - * candidate for being a separate node type because it is a specialization of - * the current ANYOFL, and because no other ANYOFL-only bits are set when it - * is; also most of its uses are actually outside the reginclass() function, so - * this could be done with no performance penalty. The other potential bits - * seem to me to have a potential issue with a combinatorial explosion of node - * types, because of not having that mutual exclusivity, where you may end up - * having to have a node type for bitX being set, one for bitY, and one for - * both bitXY. + * Another possibility is based on the fact that ANYOF_MATCHES_POSIXL is + * redundant with the node type ANYOFPOSIXL. That flag could be removed, but + * at the expense of extra code in regexec.c. The flag has been retained + * because it allows us to see if we need to call reginsert, or just use the + * bitmap in one test. * - * If you don't want to do this, or two bits are required, one could instead - * rename the ANYOF_POSIXL bit to be ANYOFL_LARGE, to mean that the ANYOF node - * has an extra 32 bits beyond what a regular one does. That's what it - * effectively means now, with the extra space all for the POSIX class bits. - * But those classes actually only occupy 30 bits, so the ANYOF_LOC_REQ_BIT (if - * an extra node type for it hasn't been created) and/or the ANYOF_LOC_FOLD - * bits could be moved there. The downside of this is that ANYOFL nodes with - * whichever of the bits get moved would have to have the extra space always - * allocated. + * If this is done, an extension would be to make all ANYOFL nodes contain the + * extra 32 bits that ANYOFPOSIXL ones do. The posix flags only occupy 30 + * bits, so the ANYOFL_SHARED_UTF8_LOCALE_fold_HAS_MATCHES_nonfold_REQD flags + * and ANYOFL_FOLD could be moved to that extra space, but it would mean extra + * instructions, as there are currently places in the code that assume those + * two bits are zero. * - * If three bits are required, one could additionally make a node type for - * ANYOFL_LARGE, removing that as a bit, and move both the above bits to that - * extra word. There isn't an SSC problem as all SSCs are this large anyway, - * and the SSC could be set to this node type. REGINCLASS would have to be - * modified so that if the node type were this, it would call reginclass(). - * as the flag bit that does this now would be gone. + * All told, 5 bits could be available for other uses if all of the above were + * done. * - * Several flags are not used in synthetic start class (SSC) nodes, so could be + * Some flags are not used in synthetic start class (SSC) nodes, so could be * shared should new flags be needed for SSCs, like SSC_MATCHES_EMPTY_STRING * now. */ @@ -441,17 +552,32 @@ struct regnode_ssc { * time. However under locale, the actual folding varies depending on * what the locale is at the time of execution, so it has to be deferred until * then. Only set under /l; never in an SSC */ -#define ANYOF_LOC_FOLD 0x04 - -/* If set, ANYOF_LOC_FOLD is also set, and there are potential matches that - * will be valid only if the locale is a UTF-8 one. */ -#define ANYOF_ONLY_UTF8_LOC_FOLD_MATCHES 0x08 - -/* If set, means to warn if runtime locale isn't a UTF-8 one. Only under /l. - * If set, none of INVERT, LOC_FOLD, POSIXL, - * ANYOF_SHARED_d_UPPER_LATIN1_UTF8_STRING_MATCHES_non_d_RUNTIME_USER_PROP can - * be set. Can be in an SSC */ -#define ANYOF_LOC_REQ_UTF8 0x10 +#define ANYOFL_FOLD 0x04 + +/* Shared bit set only with ANYOFL and SSC nodes: + * If ANYOFL_FOLD is set, this flag indicates there are potential matches + * valid only if the locale is a UTF-8 one. + * If ANYOFL_FOLD is NOT set, this flag means to warn if the runtime locale + * isn't a UTF-8 one (and the generated node assumes a UTF-8 locale). + * None of INVERT, POSIXL, + * ANYOF_SHARED_d_UPPER_LATIN1_UTF8_STRING_MATCHES_non_d_RUNTIME_USER_PROP + * can be set. */ +#define ANYOFL_SHARED_UTF8_LOCALE_fold_HAS_MATCHES_nonfold_REQD 0x08 + +/* Convenience macros for teasing apart the meanings when reading the above bit + * */ +#define ANYOFL_SOME_FOLDS_ONLY_IN_UTF8_LOCALE(flags) \ + ((flags & ( ANYOFL_FOLD /* Both bits are set */ \ + |ANYOFL_SHARED_UTF8_LOCALE_fold_HAS_MATCHES_nonfold_REQD)) \ + == ( ANYOFL_FOLD \ + |ANYOFL_SHARED_UTF8_LOCALE_fold_HAS_MATCHES_nonfold_REQD)) + +#define ANYOFL_UTF8_LOCALE_REQD(flags) \ + ((flags & ( ANYOFL_FOLD /* Only REQD bit is set */ \ + |ANYOFL_SHARED_UTF8_LOCALE_fold_HAS_MATCHES_nonfold_REQD)) \ + == ANYOFL_SHARED_UTF8_LOCALE_fold_HAS_MATCHES_nonfold_REQD) + +/* Spare: Be sure to change ANYOF_FLAGS_ALL if this gets used 0x10 */ /* If set, the node matches every code point NUM_ANYOF_CODE_POINTS and above. * Can be in an SSC */ @@ -460,10 +586,11 @@ struct regnode_ssc { /* Shared bit: * Under /d it means the ANYOFD node matches more things if the target * string is encoded in UTF-8; any such things will be non-ASCII, - * characters that are < 256, and can be accessed via the swash. + * characters that are < 256, and can be accessed via the inversion + * list. * When not under /d, it means the ANYOF node contains a user-defined * property that wasn't yet defined at the time the regex was compiled, - * and so must be looked up at runtime, by creating a swash + * and so must be looked up at runtime, by creating an inversion list. * (These uses are mutually exclusive because a user-defined property is * specified by \p{}, and \p{} implies /u which deselects /d). The long macro * name is to make sure that you are cautioned about its shared nature. Only @@ -481,14 +608,14 @@ struct regnode_ssc { * are cautioned about its shared nature */ #define ANYOF_SHARED_d_MATCHES_ALL_NON_UTF8_NON_ASCII_non_d_WARN_SUPER 0x80 -#define ANYOF_FLAGS_ALL (0xff) +#define ANYOF_FLAGS_ALL (0xff & ~0x10) -#define ANYOF_LOCALE_FLAGS (ANYOF_LOC_FOLD | ANYOF_MATCHES_POSIXL) +#define ANYOF_LOCALE_FLAGS (ANYOFL_FOLD | ANYOF_MATCHES_POSIXL) /* These are the flags that apply to both regular ANYOF nodes and synthetic * start class nodes during construction of the SSC. During finalization of * the SSC, other of the flags may get added to it */ -#define ANYOF_COMMON_FLAGS ANYOF_LOC_REQ_UTF8 +#define ANYOF_COMMON_FLAGS 0 /* Character classes for node->classflags of ANYOF */ /* Should be synchronized with a table in regprop() */ @@ -569,17 +696,22 @@ struct regnode_ssc { #define ANYOF_BIT(c) (1U << ((c) & 7)) -#define ANYOF_POSIXL_SET(p, c) (((regnode_charclass_posixl*) (p))->classflags |= (1U << (c))) -#define ANYOF_CLASS_SET(p, c) ANYOF_POSIXL_SET((p), (c)) +#define POSIXL_SET(field, c) ((field) |= (1U << (c))) +#define ANYOF_POSIXL_SET(p, c) POSIXL_SET(((regnode_charclass_posixl*) (p))->classflags, (c)) -#define ANYOF_POSIXL_CLEAR(p, c) (((regnode_charclass_posixl*) (p))->classflags &= ~ (1U <<(c))) -#define ANYOF_CLASS_CLEAR(p, c) ANYOF_POSIXL_CLEAR((p), (c)) +#define POSIXL_CLEAR(field, c) ((field) &= ~ (1U <<(c))) +#define ANYOF_POSIXL_CLEAR(p, c) POSIXL_CLEAR(((regnode_charclass_posixl*) (p))->classflags, (c)) -#define ANYOF_POSIXL_TEST(p, c) (((regnode_charclass_posixl*) (p))->classflags & (1U << (c))) -#define ANYOF_CLASS_TEST(p, c) ANYOF_POSIXL_TEST((p), (c)) +#define POSIXL_TEST(field, c) ((field) & (1U << (c))) +#define ANYOF_POSIXL_TEST(p, c) POSIXL_TEST(((regnode_charclass_posixl*) (p))->classflags, (c)) -#define ANYOF_POSIXL_ZERO(ret) STMT_START { ((regnode_charclass_posixl*) (ret))->classflags = 0; } STMT_END -#define ANYOF_CLASS_ZERO(ret) ANYOF_POSIXL_ZERO(ret) +#define POSIXL_ZERO(field) STMT_START { (field) = 0; } STMT_END +#define ANYOF_POSIXL_ZERO(ret) POSIXL_ZERO(((regnode_charclass_posixl*) (ret))->classflags) + +#define ANYOF_POSIXL_SET_TO_BITMAP(p, bits) \ + STMT_START { \ + ((regnode_charclass_posixl*) (p))->classflags = (bits); \ + } STMT_END /* Shifts a bit to get, eg. 0x4000_0000, then subtracts 1 to get 0x3FFF_FFFF */ #define ANYOF_POSIXL_SETALL(ret) STMT_START { ((regnode_charclass_posixl*) (ret))->classflags = ((1U << ((ANYOF_POSIXL_MAX) - 1))) - 1; } STMT_END @@ -608,9 +740,9 @@ struct regnode_ssc { #define ANYOF_POSIXL_AND(source, dest) STMT_START { (dest)->classflags &= (source)->classflags ; } STMT_END -#define ANYOF_BITMAP_ZERO(ret) Zero(((struct regnode_charclass*)(ret))->bitmap, ANYOF_BITMAP_SIZE, char) -#define ANYOF_BITMAP(p) (((struct regnode_charclass*)(p))->bitmap) -#define ANYOF_BITMAP_BYTE(p, c) (ANYOF_BITMAP(p)[(((U8)(c)) >> 3) & 31]) +#define ANYOF_BITMAP_ZERO(ret) Zero(((regnode_charclass*)(ret))->bitmap, ANYOF_BITMAP_SIZE, char) +#define ANYOF_BITMAP(p) ((regnode_charclass*)(p))->bitmap +#define ANYOF_BITMAP_BYTE(p, c) BITMAP_BYTE(ANYOF_BITMAP(p), c) #define ANYOF_BITMAP_SET(p, c) (ANYOF_BITMAP_BYTE(p, c) |= ANYOF_BIT(c)) #define ANYOF_BITMAP_CLEAR(p,c) (ANYOF_BITMAP_BYTE(p, c) &= ~ANYOF_BIT(c)) #define ANYOF_BITMAP_TEST(p, c) cBOOL(ANYOF_BITMAP_BYTE(p, c) & ANYOF_BIT(c)) @@ -620,9 +752,6 @@ struct regnode_ssc { #define ANYOF_BITMAP_CLEARALL(p) \ Zero (ANYOF_BITMAP(p), ANYOF_BITMAP_SIZE) -#define ANYOF_SKIP EXTRA_SIZE(struct regnode_charclass) -#define ANYOF_POSIXL_SKIP EXTRA_SIZE(regnode_charclass_posixl) - /* * Utility definitions. */ @@ -632,6 +761,8 @@ struct regnode_ssc { # define UCHARAT(p) ((int)*(p)&CHARMASK) #endif +/* Number of regnode equivalents that 'guy' occupies beyond the size of the + * smallest regnode. */ #define EXTRA_SIZE(guy) ((sizeof(guy)-1)/sizeof(struct regnode)) #define REG_ZERO_LEN_SEEN 0x00000001 @@ -644,7 +775,7 @@ struct regnode_ssc { #define REG_CUTGROUP_SEEN 0x00000100 #define REG_RUN_ON_COMMENT_SEEN 0x00000200 #define REG_UNFOLDED_MULTI_SEEN 0x00000400 -#define REG_GOSTART_SEEN 0x00000800 +/* spare */ #define REG_UNBOUNDED_QUANTIFIER_SEEN 0x00001000 @@ -691,9 +822,9 @@ END_EXTERN_C * l - start op for literal (?{EVAL}) item * L - start op for literal (?{EVAL}) item, with separate CV (qr//) * r - pointer to an embedded code-containing qr, e.g. /ab$qr/ - * s - swash for Unicode-style character class, and the multicharacter - * strings resulting from casefolding the single-character entries - * in the character class + * s - inversion list for Unicode-style character class, and the + * multicharacter strings resulting from casefolding the single-character + * entries in the character class * t - trie struct * u - trie struct's widecharmap (a HV, so can't share, must dup) * also used for revcharmap and words under DEBUGGING @@ -836,7 +967,7 @@ typedef struct _reg_ac_data reg_ac_data; three different sets... */ #define TRIE_BITMAP(p) (((reg_trie_data *)(p))->bitmap) -#define TRIE_BITMAP_BYTE(p, c) (TRIE_BITMAP(p)[(((U8)(c)) >> 3) & 31]) +#define TRIE_BITMAP_BYTE(p, c) BITMAP_BYTE(TRIE_BITMAP(p), c) #define TRIE_BITMAP_SET(p, c) (TRIE_BITMAP_BYTE(p, c) |= ANYOF_BIT((U8)c)) #define TRIE_BITMAP_CLEAR(p,c) (TRIE_BITMAP_BYTE(p, c) &= ~ANYOF_BIT((U8)c)) #define TRIE_BITMAP_TEST(p, c) (TRIE_BITMAP_BYTE(p, c) & ANYOF_BIT((U8)c)) @@ -845,7 +976,7 @@ typedef struct _reg_ac_data reg_ac_data; #define IS_TRIE_AC(op) ((op)>=AHOCORASICK) -#define BITMAP_BYTE(p, c) (((U8*)p)[(((U8)(c)) >> 3) & 31]) +#define BITMAP_BYTE(p, c) (( (U8*) p) [ ( ( (UV) (c)) >> 3) ] ) #define BITMAP_TEST(p, c) (BITMAP_BYTE(p, c) & ANYOF_BIT((U8)c)) /* these defines assume uniquecharcount is the correct variable, and state may be evaluated twice */ @@ -863,6 +994,9 @@ typedef struct _reg_ac_data reg_ac_data; #define RE_TRIE_MAXBUF_NAME "\022E_TRIE_MAXBUF" #define RE_DEBUG_FLAGS "\022E_DEBUG_FLAGS" +#define RE_COMPILE_RECURSION_INIT 1000 +#define RE_COMPILE_RECURSION_LIMIT "\022E_COMPILE_RECURSION_LIMIT" + /* RE_DEBUG_FLAGS is used to control what debug output is emitted @@ -911,79 +1045,81 @@ re.pm, especially to the documentation. #define RE_DEBUG_EXECUTE_TRIE 0x000400 /* Extra */ -#define RE_DEBUG_EXTRA_MASK 0xFF0000 -#define RE_DEBUG_EXTRA_TRIE 0x010000 -#define RE_DEBUG_EXTRA_OFFSETS 0x020000 -#define RE_DEBUG_EXTRA_OFFDEBUG 0x040000 -#define RE_DEBUG_EXTRA_STATE 0x080000 -#define RE_DEBUG_EXTRA_OPTIMISE 0x100000 -#define RE_DEBUG_EXTRA_BUFFERS 0x400000 -#define RE_DEBUG_EXTRA_GPOS 0x800000 +#define RE_DEBUG_EXTRA_MASK 0x1FF0000 +#define RE_DEBUG_EXTRA_TRIE 0x0010000 +#define RE_DEBUG_EXTRA_OFFSETS 0x0020000 +#define RE_DEBUG_EXTRA_OFFDEBUG 0x0040000 +#define RE_DEBUG_EXTRA_STATE 0x0080000 +#define RE_DEBUG_EXTRA_OPTIMISE 0x0100000 +#define RE_DEBUG_EXTRA_BUFFERS 0x0400000 +#define RE_DEBUG_EXTRA_GPOS 0x0800000 +#define RE_DEBUG_EXTRA_DUMP_PRE_OPTIMIZE 0x1000000 /* combined */ -#define RE_DEBUG_EXTRA_STACK 0x280000 +#define RE_DEBUG_EXTRA_STACK 0x0280000 #define RE_DEBUG_FLAG(x) (re_debug_flags & x) /* Compile */ #define DEBUG_COMPILE_r(x) DEBUG_r( \ - if (re_debug_flags & RE_DEBUG_COMPILE_MASK) x ) + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_COMPILE_MASK)) x ) #define DEBUG_PARSE_r(x) DEBUG_r( \ - if (re_debug_flags & RE_DEBUG_COMPILE_PARSE) x ) + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_COMPILE_PARSE)) x ) #define DEBUG_OPTIMISE_r(x) DEBUG_r( \ - if (re_debug_flags & RE_DEBUG_COMPILE_OPTIMISE) x ) -#define DEBUG_PARSE_r(x) DEBUG_r( \ - if (re_debug_flags & RE_DEBUG_COMPILE_PARSE) x ) + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_COMPILE_OPTIMISE)) x ) #define DEBUG_DUMP_r(x) DEBUG_r( \ - if (re_debug_flags & RE_DEBUG_COMPILE_DUMP) x ) + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_COMPILE_DUMP)) x ) #define DEBUG_TRIE_COMPILE_r(x) DEBUG_r( \ - if (re_debug_flags & RE_DEBUG_COMPILE_TRIE) x ) + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_COMPILE_TRIE)) x ) #define DEBUG_FLAGS_r(x) DEBUG_r( \ - if (re_debug_flags & RE_DEBUG_COMPILE_FLAGS) x ) + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_COMPILE_FLAGS)) x ) #define DEBUG_TEST_r(x) DEBUG_r( \ - if (re_debug_flags & RE_DEBUG_COMPILE_TEST) x ) + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_COMPILE_TEST)) x ) /* Execute */ #define DEBUG_EXECUTE_r(x) DEBUG_r( \ - if (re_debug_flags & RE_DEBUG_EXECUTE_MASK) x ) + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_EXECUTE_MASK)) x ) #define DEBUG_INTUIT_r(x) DEBUG_r( \ - if (re_debug_flags & RE_DEBUG_EXECUTE_INTUIT) x ) + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_EXECUTE_INTUIT)) x ) #define DEBUG_MATCH_r(x) DEBUG_r( \ - if (re_debug_flags & RE_DEBUG_EXECUTE_MATCH) x ) + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_EXECUTE_MATCH)) x ) #define DEBUG_TRIE_EXECUTE_r(x) DEBUG_r( \ - if (re_debug_flags & RE_DEBUG_EXECUTE_TRIE) x ) + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_EXECUTE_TRIE)) x ) /* Extra */ #define DEBUG_EXTRA_r(x) DEBUG_r( \ - if (re_debug_flags & RE_DEBUG_EXTRA_MASK) x ) + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_EXTRA_MASK)) x ) #define DEBUG_OFFSETS_r(x) DEBUG_r( \ - if (re_debug_flags & RE_DEBUG_EXTRA_OFFSETS) x ) + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_EXTRA_OFFSETS)) x ) #define DEBUG_STATE_r(x) DEBUG_r( \ - if (re_debug_flags & RE_DEBUG_EXTRA_STATE) x ) + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_EXTRA_STATE)) x ) #define DEBUG_STACK_r(x) DEBUG_r( \ - if (re_debug_flags & RE_DEBUG_EXTRA_STACK) x ) + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_EXTRA_STACK)) x ) #define DEBUG_BUFFERS_r(x) DEBUG_r( \ - if (re_debug_flags & RE_DEBUG_EXTRA_BUFFERS) x ) + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_EXTRA_BUFFERS)) x ) #define DEBUG_OPTIMISE_MORE_r(x) DEBUG_r( \ - if ((RE_DEBUG_EXTRA_OPTIMISE|RE_DEBUG_COMPILE_OPTIMISE) == \ - (re_debug_flags & (RE_DEBUG_EXTRA_OPTIMISE|RE_DEBUG_COMPILE_OPTIMISE)) ) x ) + if (DEBUG_v_TEST || ((RE_DEBUG_EXTRA_OPTIMISE|RE_DEBUG_COMPILE_OPTIMISE) == \ + (re_debug_flags & (RE_DEBUG_EXTRA_OPTIMISE|RE_DEBUG_COMPILE_OPTIMISE)))) x ) #define MJD_OFFSET_DEBUG(x) DEBUG_r( \ - if (re_debug_flags & RE_DEBUG_EXTRA_OFFDEBUG) \ + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_EXTRA_OFFDEBUG)) \ Perl_warn_nocontext x ) #define DEBUG_TRIE_COMPILE_MORE_r(x) DEBUG_TRIE_COMPILE_r( \ - if (re_debug_flags & RE_DEBUG_EXTRA_TRIE) x ) + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_EXTRA_TRIE)) x ) #define DEBUG_TRIE_EXECUTE_MORE_r(x) DEBUG_TRIE_EXECUTE_r( \ - if (re_debug_flags & RE_DEBUG_EXTRA_TRIE) x ) + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_EXTRA_TRIE)) x ) #define DEBUG_TRIE_r(x) DEBUG_r( \ - if (re_debug_flags & (RE_DEBUG_COMPILE_TRIE \ - | RE_DEBUG_EXECUTE_TRIE )) x ) + if (DEBUG_v_TEST || (re_debug_flags & (RE_DEBUG_COMPILE_TRIE \ + | RE_DEBUG_EXECUTE_TRIE ))) x ) #define DEBUG_GPOS_r(x) DEBUG_r( \ - if (re_debug_flags & RE_DEBUG_EXTRA_GPOS) x ) + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_EXTRA_GPOS)) x ) + +#define DEBUG_DUMP_PRE_OPTIMIZE_r(x) DEBUG_r( \ + if (DEBUG_v_TEST || (re_debug_flags & RE_DEBUG_EXTRA_DUMP_PRE_OPTIMIZE)) x ) /* initialization */ /* get_sv() can return NULL during global destruction. */ #define GET_RE_DEBUG_FLAGS DEBUG_r({ \ SV * re_debug_flags_sv = NULL; \ - re_debug_flags_sv = get_sv(RE_DEBUG_FLAGS, 1); \ + re_debug_flags_sv = PL_curcop ? get_sv(RE_DEBUG_FLAGS, GV_ADD) : NULL; \ if (re_debug_flags_sv) { \ if (!SvIOK(re_debug_flags_sv)) \ sv_setuv(re_debug_flags_sv, RE_DEBUG_COMPILE_DUMP | RE_DEBUG_EXECUTE_MASK ); \ @@ -993,26 +1129,27 @@ re.pm, especially to the documentation. #ifdef DEBUGGING -#define GET_RE_DEBUG_FLAGS_DECL VOL IV re_debug_flags = 0; \ +#define GET_RE_DEBUG_FLAGS_DECL volatile IV re_debug_flags = 0; \ PERL_UNUSED_VAR(re_debug_flags); GET_RE_DEBUG_FLAGS; -#define RE_PV_COLOR_DECL(rpv,rlen,isuni,dsv,pv,l,m,c1,c2) \ - const char * const rpv = \ - pv_pretty((dsv), (pv), (l), (m), \ - PL_colors[(c1)],PL_colors[(c2)], \ +#define RE_PV_COLOR_DECL(rpv,rlen,isuni,dsv,pv,l,m,c1,c2) \ + const char * const rpv = \ + pv_pretty((dsv), (pv), (l), (m), \ + PL_colors[(c1)],PL_colors[(c2)], \ PERL_PV_ESCAPE_RE|PERL_PV_ESCAPE_NONASCII |((isuni) ? PERL_PV_ESCAPE_UNI : 0) ); \ const int rlen = SvCUR(dsv) -#define RE_SV_ESCAPE(rpv,isuni,dsv,sv,m) \ - const char * const rpv = \ - pv_pretty((dsv), (SvPV_nolen_const(sv)), (SvCUR(sv)), (m), \ - PL_colors[(c1)],PL_colors[(c2)], \ +/* This is currently unsed in the core */ +#define RE_SV_ESCAPE(rpv,isuni,dsv,sv,m) \ + const char * const rpv = \ + pv_pretty((dsv), (SvPV_nolen_const(sv)), (SvCUR(sv)), (m), \ + PL_colors[(c1)],PL_colors[(c2)], \ PERL_PV_ESCAPE_RE|PERL_PV_ESCAPE_NONASCII |((isuni) ? PERL_PV_ESCAPE_UNI : 0) ) -#define RE_PV_QUOTED_DECL(rpv,isuni,dsv,pv,l,m) \ - const char * const rpv = \ - pv_pretty((dsv), (pv), (l), (m), \ - PL_colors[0], PL_colors[1], \ +#define RE_PV_QUOTED_DECL(rpv,isuni,dsv,pv,l,m) \ + const char * const rpv = \ + pv_pretty((dsv), (pv), (l), (m), \ + PL_colors[0], PL_colors[1], \ ( PERL_PV_PRETTY_QUOTE | PERL_PV_ESCAPE_RE | PERL_PV_ESCAPE_NONASCII | PERL_PV_PRETTY_ELLIPSES | \ ((isuni) ? PERL_PV_ESCAPE_UNI : 0)) \ ) @@ -1031,13 +1168,43 @@ re.pm, especially to the documentation. #endif /* DEBUG RELATED DEFINES */ +#define FIRST_NON_ASCII_DECIMAL_DIGIT 0x660 /* ARABIC_INDIC_DIGIT_ZERO */ + typedef enum { TRADITIONAL_BOUND = _CC_WORDCHAR, GCB_BOUND, + LB_BOUND, SB_BOUND, WB_BOUND } bound_type; +/* This unpacks the FLAGS field of ANYOFHx nodes. The value it contains + * gives the strict lower bound for the UTF-8 start byte of any code point + * matchable by the node, and a loose upper bound as well. + * + * The low bound is stored in the upper 6 bits, plus 0xC0. + * The loose upper bound is determined from the lowest 2 bits and the low bound + * (called x) as follows: + * + * 11 The upper limit of the range can be as much as (EF - x) / 8 + * 10 The upper limit of the range can be as much as (EF - x) / 4 + * 01 The upper limit of the range can be as much as (EF - x) / 2 + * 00 The upper limit of the range can be as much as EF + * + * For motivation of this design, see commit message in + * 3146c00a633e9cbed741e10146662fbcedfdb8d3 */ +#ifdef EBCDIC +# define MAX_ANYOF_HRx_BYTE 0xF4 +#else +# define MAX_ANYOF_HRx_BYTE 0xEF +#endif +#define LOWEST_ANYOF_HRx_BYTE(b) (((b) >> 2) + 0xC0) +#define HIGHEST_ANYOF_HRx_BYTE(b) \ + (LOWEST_ANYOF_HRx_BYTE(b) \ + + ((MAX_ANYOF_HRx_BYTE - LOWEST_ANYOF_HRx_BYTE(b)) >> ((b) & 3))) + +#endif /* PERL_REGCOMP_H_ */ + /* * ex: set ts=8 sts=4 sw=4 et: */