From 7aad0205e1d04932a2a35b39c5014b199f4298ce Mon Sep 17 00:00:00 2001 From: Vladimir Dubrovin <3proxy@3proxy.ru> Date: Mon, 17 Mar 2025 19:44:48 +0300 Subject: [PATCH] Remove legacy NTLMv1 code --- src/Makefile.inc | 10 +- src/auth.c | 16 - src/libs/regex.c | 3821 --------------------------------------------- src/libs/regex.h | 74 - src/libs/smbdes.c | 321 ---- src/ntlm.c | 88 -- src/proxy.c | 48 - 7 files changed, 2 insertions(+), 4376 deletions(-) delete mode 100644 src/libs/regex.c delete mode 100644 src/libs/regex.h delete mode 100644 src/libs/smbdes.c delete mode 100644 src/ntlm.c diff --git a/src/Makefile.inc b/src/Makefile.inc index d7f4b1e..922b20a 100644 --- a/src/Makefile.inc +++ b/src/Makefile.inc @@ -144,18 +144,12 @@ $(BUILDDIR)mycrypt$(EXESUFFICS): md4$(OBJSUFFICS) md5$(OBJSUFFICS) mycryptmain$( md4$(OBJSUFFICS): libs/md4.h libs/md4.c $(CC) $(COUT)md4$(OBJSUFFICS) $(CFLAGS) libs/md4.c -smbdes$(OBJSUFFICS): libs/smbdes.c - $(CC) $(COUT)smbdes$(OBJSUFFICS) $(CFLAGS) libs/smbdes.c - md5$(OBJSUFFICS): libs/md5.h libs/md5.c $(CC) $(COUT)md5$(OBJSUFFICS) $(CFLAGS) libs/md5.c -ntlm$(OBJSUFFICS): ntlm.c - $(CC) $(COUT)ntlm$(OBJSUFFICS) $(CFLAGS) ntlm.c - stringtable$(OBJSUFFICS): stringtable.c $(CC) $(COUT)stringtable$(OBJSUFFICS) $(CFLAGS) stringtable.c -$(BUILDDIR)3proxy$(EXESUFFICS): 3proxy$(OBJSUFFICS) mainfunc$(OBJSUFFICS) srvproxy$(OBJSUFFICS) srvpop3p$(OBJSUFFICS) srvsmtpp$(OBJSUFFICS) srvftppr$(OBJSUFFICS) srvsocks$(OBJSUFFICS) srvtcppm$(OBJSUFFICS) srvtlspr$(OBJSUFFICS) srvauto$(OBJSUFFICS) srvudppm$(OBJSUFFICS) sockmap$(OBJSUFFICS) sockgetchar$(OBJSUFFICS) common$(OBJSUFFICS) auth$(OBJSUFFICS) authradius$(OBJSUFFICS) conf$(OBJSUFFICS) log$(OBJSUFFICS) datatypes$(OBJSUFFICS) md4$(OBJSUFFICS) md5$(OBJSUFFICS) mycrypt$(OBJSUFFICS) base64$(OBJSUFFICS) ftp$(OBJSUFFICS) smbdes$(OBJSUFFICS) ntlm$(OBJSUFFICS) stringtable$(OBJSUFFICS) srvwebadmin$(OBJSUFFICS) srvdnspr$(OBJSUFFICS) plugins$(OBJSUFFICS) $(COMPATLIBS) $(VERSIONDEP) - $(LN) $(LNOUT)$(BUILDDIR)3proxy$(EXESUFFICS) $(LDFLAGS) $(VERFILE) 3proxy$(OBJSUFFICS) mainfunc$(OBJSUFFICS) auth$(OBJSUFFICS) authradius$(OBJSUFFICS) conf$(OBJSUFFICS) datatypes$(OBJSUFFICS) srvauto$(OBJSUFFICS) srvproxy$(OBJSUFFICS) srvpop3p$(OBJSUFFICS) srvsmtpp$(OBJSUFFICS) srvftppr$(OBJSUFFICS) srvsocks$(OBJSUFFICS) srvtcppm$(OBJSUFFICS) srvtlspr$(OBJSUFFICS) srvudppm$(OBJSUFFICS) sockmap$(OBJSUFFICS) sockgetchar$(OBJSUFFICS) common$(OBJSUFFICS) log$(OBJSUFFICS) mycrypt$(OBJSUFFICS) md5$(OBJSUFFICS) md4$(OBJSUFFICS) base64$(OBJSUFFICS) ftp$(OBJSUFFICS) smbdes$(OBJSUFFICS) ntlm$(OBJSUFFICS) stringtable$(OBJSUFFICS) srvwebadmin$(OBJSUFFICS) srvdnspr$(OBJSUFFICS) plugins$(OBJSUFFICS) $(COMPATLIBS) $(LIBS) +$(BUILDDIR)3proxy$(EXESUFFICS): 3proxy$(OBJSUFFICS) mainfunc$(OBJSUFFICS) srvproxy$(OBJSUFFICS) srvpop3p$(OBJSUFFICS) srvsmtpp$(OBJSUFFICS) srvftppr$(OBJSUFFICS) srvsocks$(OBJSUFFICS) srvtcppm$(OBJSUFFICS) srvtlspr$(OBJSUFFICS) srvauto$(OBJSUFFICS) srvudppm$(OBJSUFFICS) sockmap$(OBJSUFFICS) sockgetchar$(OBJSUFFICS) common$(OBJSUFFICS) auth$(OBJSUFFICS) authradius$(OBJSUFFICS) conf$(OBJSUFFICS) log$(OBJSUFFICS) datatypes$(OBJSUFFICS) md4$(OBJSUFFICS) md5$(OBJSUFFICS) mycrypt$(OBJSUFFICS) base64$(OBJSUFFICS) ftp$(OBJSUFFICS) stringtable$(OBJSUFFICS) srvwebadmin$(OBJSUFFICS) srvdnspr$(OBJSUFFICS) plugins$(OBJSUFFICS) $(COMPATLIBS) $(VERSIONDEP) + $(LN) $(LNOUT)$(BUILDDIR)3proxy$(EXESUFFICS) $(LDFLAGS) $(VERFILE) 3proxy$(OBJSUFFICS) mainfunc$(OBJSUFFICS) auth$(OBJSUFFICS) authradius$(OBJSUFFICS) conf$(OBJSUFFICS) datatypes$(OBJSUFFICS) srvauto$(OBJSUFFICS) srvproxy$(OBJSUFFICS) srvpop3p$(OBJSUFFICS) srvsmtpp$(OBJSUFFICS) srvftppr$(OBJSUFFICS) srvsocks$(OBJSUFFICS) srvtcppm$(OBJSUFFICS) srvtlspr$(OBJSUFFICS) srvudppm$(OBJSUFFICS) sockmap$(OBJSUFFICS) sockgetchar$(OBJSUFFICS) common$(OBJSUFFICS) log$(OBJSUFFICS) mycrypt$(OBJSUFFICS) md5$(OBJSUFFICS) md4$(OBJSUFFICS) base64$(OBJSUFFICS) ftp$(OBJSUFFICS) stringtable$(OBJSUFFICS) srvwebadmin$(OBJSUFFICS) srvdnspr$(OBJSUFFICS) plugins$(OBJSUFFICS) $(COMPATLIBS) $(LIBS) diff --git a/src/auth.c b/src/auth.c index eadcc74..95c30f9 100644 --- a/src/auth.c +++ b/src/auth.c @@ -971,15 +971,6 @@ int strongauth(struct clientparam * param){ else if (!param->pwtype && param->password && !strcmp((char *)param->password, (char *)pwl->password)){ break; } -#ifndef NOCRYPT - else if (param->pwtype == 2 && param->password) { - ntpwdhash(buf, pwl->password, 0); - mschap(buf, param->password, buf + 16); - if(!memcmp(buf+16, param->password+8, 24)) { - break; - } - } -#endif pthread_mutex_unlock(&pwl_mutex); return 6; #ifndef NOCRYPT @@ -993,13 +984,6 @@ int strongauth(struct clientparam * param){ if(param->password && !param->pwtype && !memcmp(pwl->password, ntpwdhash(buf,param->password, 1), 32)) { break; } - else if (param->pwtype == 2){ - fromhex(pwl->password, buf, 16); - mschap(buf, param->password, buf + 16); - if(!memcmp(buf + 16, param->password+8, 24)) { - break; - } - } pthread_mutex_unlock(&pwl_mutex); return 8; #endif diff --git a/src/libs/regex.c b/src/libs/regex.c deleted file mode 100644 index 82ae70a..0000000 --- a/src/libs/regex.c +++ /dev/null @@ -1,3821 +0,0 @@ -#include -#include -#include -#include -#include -#include -#include - -#include "regex.h" - -/* utility definitions */ -#ifdef _POSIX2_RE_DUP_MAX -#define DUPMAX _POSIX2_RE_DUP_MAX -#else -#define DUPMAX 255 -#endif -#define INFINITY (DUPMAX + 1) -#define NC (CHAR_MAX - CHAR_MIN + 1) -typedef unsigned char uch; - -/* for old systems with bcopy() but no memmove() */ -#ifdef USEBCOPY -#define memmove(d, s, c) bcopy(s, d, c) -#endif - -#define MAGIC1 ((('r'^0200)<<8) | 'e') - -/* - * The internal representation is a *strip*, a sequence of - * operators ending with an endmarker. (Some terminology etc. is a - * historical relic of earlier versions which used multiple strips.) - * Certain oddities in the representation are there to permit running - * the machinery backwards; in particular, any deviation from sequential - * flow must be marked at both its source and its destination. Some - * fine points: - * - * - OPLUS_ and O_PLUS are *inside* the loop they create. - * - OQUEST_ and O_QUEST are *outside* the bypass they create. - * - OCH_ and O_CH are *outside* the multi-way branch they create, while - * OOR1 and OOR2 are respectively the end and the beginning of one of - * the branches. Note that there is an implicit OOR2 following OCH_ - * and an implicit OOR1 preceding O_CH. - * - * In state representations, an operator's bit is on to signify a state - * immediately *preceding* "execution" of that operator. - */ -typedef long sop; /* strip operator */ -typedef long sopno; -#define OPRMASK 0x7c000000 -#define OPDMASK 0x03ffffff -#define OPSHIFT (26) -#define OP(n) ((n)&OPRMASK) -#define OPND(n) ((n)&OPDMASK) -#define SOP(op, opnd) ((op)|(opnd)) -/* operators meaning operand */ -/* (back, fwd are offsets) */ -#define OEND (1< uch [csetsize] */ - uch mask; /* bit within array */ - uch hash; /* hash code */ - size_t smultis; - char *multis; /* -> char[smulti] ab\0cd\0ef\0\0 */ -} cset; -/* note that CHadd and CHsub are unsafe, and CHIN doesn't yield 0/1 */ -#define CHadd(cs, c) ((cs)->ptr[(uch)(c)] |= (cs)->mask, (cs)->hash += (c)) -#define CHsub(cs, c) ((cs)->ptr[(uch)(c)] &= ~(cs)->mask, (cs)->hash -= (c)) -#define CHIN(cs, c) ((cs)->ptr[(uch)(c)] & (cs)->mask) -#define MCadd(p, cs, cp) mcadd(p, cs, cp) /* regcomp() internal fns */ - -/* stuff for character categories */ -typedef unsigned char cat_t; - -/* - * main compiled-expression structure - */ -struct re_guts { - int magic; -# define MAGIC2 ((('R'^0200)<<8)|'E') - sop *strip; /* malloced area for strip */ - int csetsize; /* number of bits in a cset vector */ - int ncsets; /* number of csets in use */ - cset *sets; /* -> cset [ncsets] */ - uch *setbits; /* -> uch[csetsize][ncsets/CHAR_BIT] */ - int cflags; /* copy of regcomp() cflags argument */ - sopno nstates; /* = number of sops */ - sopno firststate; /* the initial OEND (normally 0) */ - sopno laststate; /* the final OEND */ - int iflags; /* internal flags */ -# define USEBOL 01 /* used ^ */ -# define USEEOL 02 /* used $ */ -# define BAD 04 /* something wrong */ - int nbol; /* number of ^ used */ - int neol; /* number of $ used */ - int ncategories; /* how many character categories */ - cat_t *categories; /* ->catspace[-CHAR_MIN] */ - char *must; /* match must contain this string */ - int mlen; /* length of must */ - size_t nsub; /* copy of re_nsub */ - int backrefs; /* does it use back references? */ - sopno nplus; /* how deep does it nest +s? */ - /* catspace must be last */ - cat_t catspace[1]; /* actually [NC] */ -}; - -/* misc utilities */ -#define OUT (CHAR_MAX+1) /* a non-character value */ -#define ISWORD(c) (isalnum(c) || (c) == '_') - - -/* character-class table */ -static struct cclass { - char *name; - char *chars; - char *multis; -} cclasses[] = { - {"alnum","ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789",""}, - {"alpha", "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz", ""}, - {"blank"," \t", ""}, - {"cntrl","\007\b\t\n\v\f\r\1\2\3\4\5\6\16\17\20\21\22\23\24\25\26\27\30\31\32\33\34\35\36\37\177",""}, - {"digit","0123456789",""}, - {"graph","ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~",""}, - {"lower","abcdefghijklmnopqrstuvwxyz",""}, - {"print","ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~ ",""}, - {"punct","!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~",""}, - {"space","\t\n\v\f\r ", ""}, - {"upper","ABCDEFGHIJKLMNOPQRSTUVWXYZ",""}, - {"xdigit","0123456789ABCDEFabcdef",""}, - {NULL,NULL,""} -}; - -/* character-name table */ -static struct cname { - char *name; - char code; -} cnames[] = { - {"NUL", '\0'}, - {"SOH", '\001'}, - {"STX", '\002'}, - {"ETX", '\003'}, - {"EOT", '\004'}, - {"ENQ", '\005'}, - {"ACK", '\006'}, - {"BEL", '\007'}, - {"alert", '\007'}, - {"BS", '\010'}, - {"backspace", '\b'}, - {"HT", '\011'}, - {"tab", '\t'}, - {"LF", '\012'}, - {"newline", '\n'}, - {"VT", '\013'}, - {"vertical-tab", '\v'}, - {"FF", '\014'}, - {"form-feed", '\f'}, - {"CR", '\015'}, - {"carriage-return", '\r'}, - {"SO", '\016'}, - {"SI", '\017'}, - {"DLE", '\020'}, - {"DC1", '\021'}, - {"DC2", '\022'}, - {"DC3", '\023'}, - {"DC4", '\024'}, - {"NAK", '\025'}, - {"SYN", '\026'}, - {"ETB", '\027'}, - {"CAN", '\030'}, - {"EM", '\031'}, - {"SUB", '\032'}, - {"ESC", '\033'}, - {"IS4", '\034'}, - {"FS", '\034'}, - {"IS3", '\035'}, - {"GS", '\035'}, - {"IS2", '\036'}, - {"RS", '\036'}, - {"IS1", '\037'}, - {"US", '\037'}, - {"space", ' '}, - {"exclamation-mark", '!'}, - {"quotation-mark", '"'}, - {"number-sign", '#'}, - {"dollar-sign", '$'}, - {"percent-sign", '%'}, - {"ampersand", '&'}, - {"apostrophe", '\''}, - {"left-parenthesis", '('}, - {"right-parenthesis", ')'}, - {"asterisk", '*'}, - {"plus-sign", '+'}, - {"comma", ','}, - {"hyphen", '-'}, - {"hyphen-minus", '-'}, - {"period", '.'}, - {"full-stop", '.'}, - {"slash", '/'}, - {"solidus", '/'}, - {"zero", '0'}, - {"one", '1'}, - {"two", '2'}, - {"three", '3'}, - {"four", '4'}, - {"five", '5'}, - {"six", '6'}, - {"seven", '7'}, - {"eight", '8'}, - {"nine", '9'}, - {"colon", ':'}, - {"semicolon", ';'}, - {"less-than-sign", '<'}, - {"equals-sign", '='}, - {"greater-than-sign", '>'}, - {"question-mark", '?'}, - {"commercial-at", '@'}, - {"left-square-bracket", '['}, - {"backslash", '\\'}, - {"reverse-solidus", '\\'}, - {"right-square-bracket", ']'}, - {"circumflex", '^'}, - {"circumflex-accent", '^'}, - {"underscore", '_'}, - {"low-line", '_'}, - {"grave-accent", '`'}, - {"left-brace", '{'}, - {"left-curly-bracket", '{'}, - {"vertical-line", '|'}, - {"right-brace", '}'}, - {"right-curly-bracket", '}'}, - {"tilde", '~'}, - {"DEL", '\177'}, - {NULL, 0} -}; - - -/* - * parse structure, passed up and down to avoid global variables and - * other clumsinesses - */ -struct parse { - char *next; /* next character in RE */ - char *end; /* end of string (-> NUL normally) */ - int error; /* has an error been seen? */ - sop *strip; /* malloced strip */ - sopno ssize; /* malloced strip size (allocated) */ - sopno slen; /* malloced strip length (used) */ - int ncsalloc; /* number of csets allocated */ - struct re_guts *g; -# define NPAREN 10 /* we need to remember () 1-9 for back refs */ - sopno pbegin[NPAREN]; /* -> ( ([0] unused) */ - sopno pend[NPAREN]; /* -> ) ([0] unused) */ -}; - -#ifdef __cplusplus -extern "C" { -#endif - -static void p_ere(register struct parse *p, int stop); -static void p_ere_exp(register struct parse *p); -static void p_str(register struct parse *p); -static void p_bre(register struct parse *p, register int end1, register int end2); -static int p_simp_re(register struct parse *p, int starordinary); -static int p_count(register struct parse *p); -static void p_bracket(register struct parse *p); -static void p_b_term(register struct parse *p, register cset *cs); -static void p_b_cclass(register struct parse *p, register cset *cs); -static void p_b_eclass(register struct parse *p, register cset *cs); -static char p_b_symbol(register struct parse *p); -static char p_b_coll_elem(register struct parse *p, int endc); -static char othercase(int ch); -static void bothcases(register struct parse *p, int ch); -static void ordinary(register struct parse *p, register int ch); -static void nonnewline(register struct parse *p); -static void repeat(register struct parse *p, sopno start, int from, int to); -static int seterr(register struct parse *p, int e); -static cset *allocset(register struct parse *p); -static void freeset(register struct parse *p, register cset *cs); -static int freezeset(register struct parse *p, register cset *cs); -static int firstch(register struct parse *p, register cset *cs); -static int nch(register struct parse *p, register cset *cs); -static void mcadd(register struct parse *p, register cset *cs, register char *cp); -static void mcinvert(register struct parse *p, register cset *cs); -static void mccase(register struct parse *p, register cset *cs); -static int isinsets(register struct re_guts *g, int c); -static int samesets(register struct re_guts *g, int c1, int c2); -static void categorize(struct parse *p, register struct re_guts *g); -static sopno dupl(register struct parse *p, sopno start, sopno finish); -static void doemit(register struct parse *p, sop op, size_t opnd); -static void doinsert(register struct parse *p, sop op, size_t opnd, sopno pos); -static void dofwd(register struct parse *p, sopno pos, sop value); -static void enlarge(register struct parse *p, sopno size); -static void stripsnug(register struct parse *p, register struct re_guts *g); -static void findmust(register struct parse *p, register struct re_guts *g); -static sopno pluscount(register struct parse *p, register struct re_guts *g); - -#ifdef __cplusplus -} -#endif - -static char nuls[10]; /* place to point scanner in event of error */ - -/* - * macros for use with parse structure - * BEWARE: these know that the parse structure is named `p' !!! - */ -#define PEEK() (*p->next) -#define PEEK2() (*(p->next+1)) -#define MORE() (p->next < p->end) -#define MORE2() (p->next+1 < p->end) -#define SEE(c) (MORE() && PEEK() == (c)) -#define SEETWO(a, b) (MORE() && MORE2() && PEEK() == (a) && PEEK2() == (b)) -#define EAT(c) ((SEE(c)) ? (NEXT(), 1) : 0) -#define EATTWO(a, b) ((SEETWO(a, b)) ? (NEXT2(), 1) : 0) -#define NEXT() (p->next++) -#define NEXT2() (p->next += 2) -#define NEXTn(n) (p->next += (n)) -#define GETNEXT() (*p->next++) -#define SETERROR(e) seterr(p, (e)) -#define REQUIRE(co, e) ((co) || SETERROR(e)) -#define MUSTSEE(c, e) (REQUIRE(MORE() && PEEK() == (c), e)) -#define MUSTEAT(c, e) (REQUIRE(MORE() && GETNEXT() == (c), e)) -#define MUSTNOTSEE(c, e) (REQUIRE(!MORE() || PEEK() != (c), e)) -#define EMIT(op, sopnd) doemit(p, (sop)(op), (size_t)(sopnd)) -#define INSERT(op, pos) doinsert(p, (sop)(op), HERE()-(pos)+1, pos) -#define AHEAD(pos) dofwd(p, pos, HERE()-(pos)) -#define ASTERN(sop, pos) EMIT(sop, HERE()-pos) -#define HERE() (p->slen) -#define THERE() (p->slen - 1) -#define THERETHERE() (p->slen - 2) -#define DROP(n) (p->slen -= (n)) - -#define never 0 /* some s have bugs too */ - -int /* 0 success, otherwise REG_something */ -regcomp(preg, pattern, cflags) -regex_t *preg; -const char *pattern; -int cflags; -{ - struct parse pa; - register struct re_guts *g; - register struct parse *p = &pa; - register int i; - register size_t len; -#define GOODFLAGS(f) ((f)&~REG_DUMP) - - cflags = GOODFLAGS(cflags); - if ((cflags®_EXTENDED) && (cflags®_NOSPEC)) - return(REG_INVARG); - - if (cflags®_PEND) { - if (preg->re_endp < pattern) - return(REG_INVARG); - len = preg->re_endp - pattern; - } else - len = strlen((char *)pattern); - - /* do the mallocs early so failure handling is easy */ - g = (struct re_guts *)malloc(sizeof(struct re_guts) + - (NC-1)*sizeof(cat_t)); - if (g == NULL) - return(REG_ESPACE); - p->ssize = len/(size_t)2*(size_t)3 + (size_t)1; /* ugh */ - p->strip = (sop *)malloc(p->ssize * sizeof(sop)); - p->slen = 0; - if (p->strip == NULL) { - free((char *)g); - return(REG_ESPACE); - } - - /* set things up */ - p->g = g; - p->next = (char *)pattern; /* convenience; we do not modify it */ - p->end = p->next + len; - p->error = 0; - p->ncsalloc = 0; - for (i = 0; i < NPAREN; i++) { - p->pbegin[i] = 0; - p->pend[i] = 0; - } - g->csetsize = NC; - g->sets = NULL; - g->setbits = NULL; - g->ncsets = 0; - g->cflags = cflags; - g->iflags = 0; - g->nbol = 0; - g->neol = 0; - g->must = NULL; - g->mlen = 0; - g->nsub = 0; - g->ncategories = 1; /* category 0 is "everything else" */ - g->categories = &g->catspace[-(CHAR_MIN)]; - (void) memset((char *)g->catspace, 0, NC*sizeof(cat_t)); - g->backrefs = 0; - - /* do it */ - EMIT(OEND, 0); - g->firststate = THERE(); - if (cflags®_EXTENDED) - p_ere(p, OUT); - else if (cflags®_NOSPEC) - p_str(p); - else - p_bre(p, OUT, OUT); - EMIT(OEND, 0); - g->laststate = THERE(); - - /* tidy up loose ends and fill things in */ - categorize(p, g); - stripsnug(p, g); - findmust(p, g); - g->nplus = pluscount(p, g); - g->magic = MAGIC2; - preg->re_nsub = g->nsub; - preg->re_g = g; - preg->re_magic = MAGIC1; - /* not debugging, so can't rely on the assert() in regexec() */ - if (g->iflags&BAD) - SETERROR(REG_ASSERT); - - /* win or lose, we're done */ - if (p->error != 0) /* lose */ - regfree(preg); - return(p->error); -#undef GOODFLAGS -} - -/* - - p_ere - ERE parser top level, concatenation and alternation - == static void p_ere(register struct parse *p, int stop); - */ -static void -p_ere(p, stop) -register struct parse *p; -int stop; /* character this ERE should end at */ -{ - register char c; - register sopno prevback; - register sopno prevfwd; - register sopno conc; - register int first = 1; /* is this the first alternative? */ - - for (;;) { - /* do a bunch of concatenated expressions */ - conc = HERE(); - while (MORE() && (c = PEEK()) != '|' && c != stop) - p_ere_exp(p); - REQUIRE(HERE() != conc, REG_EMPTY); /* require nonempty */ - - if (!EAT('|')) - break; /* NOTE BREAK OUT */ - - if (first) { - INSERT(OCH_, conc); /* offset is wrong */ - prevfwd = conc; - prevback = conc; - first = 0; - } - ASTERN(OOR1, prevback); - prevback = THERE(); - AHEAD(prevfwd); /* fix previous offset */ - prevfwd = HERE(); - EMIT(OOR2, 0); /* offset is very wrong */ - } - - if (!first) { /* tail-end fixups */ - AHEAD(prevfwd); - ASTERN(O_CH, prevback); - } - - assert(!MORE() || SEE(stop)); -} - -/* - - p_ere_exp - parse one subERE, an atom possibly followed by a repetition op - == static void p_ere_exp(register struct parse *p); - */ -static void -p_ere_exp(p) -register struct parse *p; -{ - register char c; - register sopno pos; - register int count; - register int count2; - register sopno subno; - int wascaret = 0; - - assert(MORE()); /* caller should have ensured this */ - c = GETNEXT(); - - pos = HERE(); - switch (c) { - case '(': - REQUIRE(MORE(), REG_EPAREN); - p->g->nsub++; - subno = p->g->nsub; - if (subno < NPAREN) - p->pbegin[subno] = HERE(); - EMIT(OLPAREN, subno); - if (!SEE(')')) - p_ere(p, ')'); - if (subno < NPAREN) { - p->pend[subno] = HERE(); - assert(p->pend[subno] != 0); - } - EMIT(ORPAREN, subno); - MUSTEAT(')', REG_EPAREN); - break; - case '^': - EMIT(OBOL, 0); - p->g->iflags |= USEBOL; - p->g->nbol++; - wascaret = 1; - break; - case '$': - EMIT(OEOL, 0); - p->g->iflags |= USEEOL; - p->g->neol++; - break; - case '|': - SETERROR(REG_EMPTY); - break; - case '*': - case '+': - case '?': - SETERROR(REG_BADRPT); - break; - case '.': - if (p->g->cflags®_NEWLINE) - nonnewline(p); - else - EMIT(OANY, 0); - break; - case '[': - p_bracket(p); - break; - case '\\': - REQUIRE(MORE(), REG_EESCAPE); - c = GETNEXT(); - ordinary(p, c); - break; - case '{': /* okay as ordinary except if digit follows */ - REQUIRE(!MORE() || !isdigit(PEEK()), REG_BADRPT); - /* FALLTHROUGH */ - default: - ordinary(p, c); - break; - } - - if (!MORE()) - return; - c = PEEK(); - /* we call { a repetition if followed by a digit */ - if (!( c == '*' || c == '+' || c == '?' || - (c == '{' && MORE2() && isdigit(PEEK2())) )) - return; /* no repetition, we're done */ - NEXT(); - - REQUIRE(!wascaret, REG_BADRPT); - switch (c) { - case '*': /* implemented as +? */ - /* this case does not require the (y|) trick, noKLUDGE */ - INSERT(OPLUS_, pos); - ASTERN(O_PLUS, pos); - INSERT(OQUEST_, pos); - ASTERN(O_QUEST, pos); - break; - case '+': - INSERT(OPLUS_, pos); - ASTERN(O_PLUS, pos); - break; - case '?': - /* KLUDGE: emit y? as (y|) until subtle bug gets fixed */ - INSERT(OCH_, pos); /* offset slightly wrong */ - ASTERN(OOR1, pos); /* this one's right */ - AHEAD(pos); /* fix the OCH_ */ - EMIT(OOR2, 0); /* offset very wrong... */ - AHEAD(THERE()); /* ...so fix it */ - ASTERN(O_CH, THERETHERE()); - break; - case '{': - count = p_count(p); - if (EAT(',')) { - if (isdigit(PEEK())) { - count2 = p_count(p); - REQUIRE(count <= count2, REG_BADBR); - } else /* single number with comma */ - count2 = INFINITY; - } else /* just a single number */ - count2 = count; - repeat(p, pos, count, count2); - if (!EAT('}')) { /* error heuristics */ - while (MORE() && PEEK() != '}') - NEXT(); - REQUIRE(MORE(), REG_EBRACE); - SETERROR(REG_BADBR); - } - break; - } - - if (!MORE()) - return; - c = PEEK(); - if (!( c == '*' || c == '+' || c == '?' || - (c == '{' && MORE2() && isdigit(PEEK2())) ) ) - return; - SETERROR(REG_BADRPT); -} - -/* - - p_str - string (no metacharacters) "parser" - == static void p_str(register struct parse *p); - */ -static void -p_str(p) -register struct parse *p; -{ - REQUIRE(MORE(), REG_EMPTY); - while (MORE()) - ordinary(p, GETNEXT()); -} - -/* - - p_bre - BRE parser top level, anchoring and concatenation - == static void p_bre(register struct parse *p, register int end1, \ - == register int end2); - * Giving end1 as OUT essentially eliminates the end1/end2 check. - * - * This implementation is a bit of a kludge, in that a trailing $ is first - * taken as an ordinary character and then revised to be an anchor. The - * only undesirable side effect is that '$' gets included as a character - * category in such cases. This is fairly harmless; not worth fixing. - * The amount of lookahead needed to avoid this kludge is excessive. - */ -static void -p_bre(p, end1, end2) -register struct parse *p; -register int end1; /* first terminating character */ -register int end2; /* second terminating character */ -{ - register sopno start = HERE(); - register int first = 1; /* first subexpression? */ - register int wasdollar = 0; - - if (EAT('^')) { - EMIT(OBOL, 0); - p->g->iflags |= USEBOL; - p->g->nbol++; - } - while (MORE() && !SEETWO(end1, end2)) { - wasdollar = p_simp_re(p, first); - first = 0; - } - if (wasdollar) { /* oops, that was a trailing anchor */ - DROP(1); - EMIT(OEOL, 0); - p->g->iflags |= USEEOL; - p->g->neol++; - } - - REQUIRE(HERE() != start, REG_EMPTY); /* require nonempty */ -} - -/* - - p_simp_re - parse a simple RE, an atom possibly followed by a repetition - == static int p_simp_re(register struct parse *p, int starordinary); - */ -static int /* was the simple RE an unbackslashed $? */ -p_simp_re(p, starordinary) -register struct parse *p; -int starordinary; /* is a leading * an ordinary character? */ -{ - register int c; - register int count; - register int count2; - register sopno pos; - register int i; - register sopno subno; -# define BACKSL (1<g->cflags®_NEWLINE) - nonnewline(p); - else - EMIT(OANY, 0); - break; - case '[': - p_bracket(p); - break; - case BACKSL|'{': - SETERROR(REG_BADRPT); - break; - case BACKSL|'(': - p->g->nsub++; - subno = p->g->nsub; - if (subno < NPAREN) - p->pbegin[subno] = HERE(); - EMIT(OLPAREN, subno); - /* the MORE here is an error heuristic */ - if (MORE() && !SEETWO('\\', ')')) - p_bre(p, '\\', ')'); - if (subno < NPAREN) { - p->pend[subno] = HERE(); - assert(p->pend[subno] != 0); - } - EMIT(ORPAREN, subno); - REQUIRE(EATTWO('\\', ')'), REG_EPAREN); - break; - case BACKSL|')': /* should not get here -- must be user */ - case BACKSL|'}': - SETERROR(REG_EPAREN); - break; - case BACKSL|'1': - case BACKSL|'2': - case BACKSL|'3': - case BACKSL|'4': - case BACKSL|'5': - case BACKSL|'6': - case BACKSL|'7': - case BACKSL|'8': - case BACKSL|'9': - i = (c&~BACKSL) - '0'; - assert(i < NPAREN); - if (p->pend[i] != 0) { - assert(i <= p->g->nsub); - EMIT(OBACK_, i); - assert(p->pbegin[i] != 0); - assert(OP(p->strip[p->pbegin[i]]) == OLPAREN); - assert(OP(p->strip[p->pend[i]]) == ORPAREN); - (void) dupl(p, p->pbegin[i]+1, p->pend[i]); - EMIT(O_BACK, i); - } else - SETERROR(REG_ESUBREG); - p->g->backrefs = 1; - break; - case '*': - REQUIRE(starordinary, REG_BADRPT); - /* FALLTHROUGH */ - default: - ordinary(p, (char)c); /* takes off BACKSL, if any */ - break; - } - - if (EAT('*')) { /* implemented as +? */ - /* this case does not require the (y|) trick, noKLUDGE */ - INSERT(OPLUS_, pos); - ASTERN(O_PLUS, pos); - INSERT(OQUEST_, pos); - ASTERN(O_QUEST, pos); - } else if (EATTWO('\\', '{')) { - count = p_count(p); - if (EAT(',')) { - if (MORE() && isdigit(PEEK())) { - count2 = p_count(p); - REQUIRE(count <= count2, REG_BADBR); - } else /* single number with comma */ - count2 = INFINITY; - } else /* just a single number */ - count2 = count; - repeat(p, pos, count, count2); - if (!EATTWO('\\', '}')) { /* error heuristics */ - while (MORE() && !SEETWO('\\', '}')) - NEXT(); - REQUIRE(MORE(), REG_EBRACE); - SETERROR(REG_BADBR); - } - } else if (c == (unsigned char)'$') /* $ (but not \$) ends it */ - return(1); - - return(0); -} - -/* - - p_count - parse a repetition count - == static int p_count(register struct parse *p); - */ -static int /* the value */ -p_count(p) -register struct parse *p; -{ - register int count = 0; - register int ndigits = 0; - - while (MORE() && isdigit(PEEK()) && count <= DUPMAX) { - count = count*10 + (GETNEXT() - '0'); - ndigits++; - } - - REQUIRE(ndigits > 0 && count <= DUPMAX, REG_BADBR); - return(count); -} - -/* - - p_bracket - parse a bracketed character list - == static void p_bracket(register struct parse *p); - * - * Note a significant property of this code: if the allocset() did SETERROR, - * no set operations are done. - */ -static void -p_bracket(p) -register struct parse *p; -{ - register cset *cs = allocset(p); - register int invert = 0; - - /* Dept of Truly Sickening Special-Case Kludges */ - if (p->next + 5 < p->end && strncmp(p->next, "[:<:]]", 6) == 0) { - EMIT(OBOW, 0); - NEXTn(6); - return; - } - if (p->next + 5 < p->end && strncmp(p->next, "[:>:]]", 6) == 0) { - EMIT(OEOW, 0); - NEXTn(6); - return; - } - - if (EAT('^')) - invert++; /* make note to invert set at end */ - if (EAT(']')) - CHadd(cs, ']'); - else if (EAT('-')) - CHadd(cs, '-'); - while (MORE() && PEEK() != ']' && !SEETWO('-', ']')) - p_b_term(p, cs); - if (EAT('-')) - CHadd(cs, '-'); - MUSTEAT(']', REG_EBRACK); - - if (p->error != 0) /* don't mess things up further */ - return; - - if (p->g->cflags®_ICASE) { - register int i; - register int ci; - - for (i = p->g->csetsize - 1; i >= 0; i--) - if (CHIN(cs, i) && isalpha(i)) { - ci = othercase(i); - if (ci != i) - CHadd(cs, ci); - } - if (cs->multis != NULL) - mccase(p, cs); - } - if (invert) { - register int i; - - for (i = p->g->csetsize - 1; i >= 0; i--) - if (CHIN(cs, i)) - CHsub(cs, i); - else - CHadd(cs, i); - if (p->g->cflags®_NEWLINE) - CHsub(cs, '\n'); - if (cs->multis != NULL) - mcinvert(p, cs); - } - - assert(cs->multis == NULL); /* xxx */ - - if (nch(p, cs) == 1) { /* optimize singleton sets */ - ordinary(p, firstch(p, cs)); - freeset(p, cs); - } else - EMIT(OANYOF, freezeset(p, cs)); -} - -/* - - p_b_term - parse one term of a bracketed character list - == static void p_b_term(register struct parse *p, register cset *cs); - */ -static void -p_b_term(p, cs) -register struct parse *p; -register cset *cs; -{ - register char c; - register char start, finish; - register int i; - - /* classify what we've got */ - switch ((MORE()) ? PEEK() : '\0') { - case '[': - c = (MORE2()) ? PEEK2() : '\0'; - break; - case '-': - SETERROR(REG_ERANGE); - return; /* NOTE RETURN */ - break; - default: - c = '\0'; - break; - } - - switch (c) { - case ':': /* character class */ - NEXT2(); - REQUIRE(MORE(), REG_EBRACK); - c = PEEK(); - REQUIRE(c != '-' && c != ']', REG_ECTYPE); - p_b_cclass(p, cs); - REQUIRE(MORE(), REG_EBRACK); - REQUIRE(EATTWO(':', ']'), REG_ECTYPE); - break; - case '=': /* equivalence class */ - NEXT2(); - REQUIRE(MORE(), REG_EBRACK); - c = PEEK(); - REQUIRE(c != '-' && c != ']', REG_ECOLLATE); - p_b_eclass(p, cs); - REQUIRE(MORE(), REG_EBRACK); - REQUIRE(EATTWO('=', ']'), REG_ECOLLATE); - break; - default: /* symbol, ordinary character, or range */ -/* xxx revision needed for multichar stuff */ - start = p_b_symbol(p); - if (SEE('-') && MORE2() && PEEK2() != ']') { - /* range */ - NEXT(); - if (EAT('-')) - finish = '-'; - else - finish = p_b_symbol(p); - } else - finish = start; -/* xxx what about signed chars here... */ - REQUIRE(start <= finish, REG_ERANGE); - for (i = start; i <= finish; i++) - CHadd(cs, i); - break; - } -} - -/* - - p_b_cclass - parse a character-class name and deal with it - == static void p_b_cclass(register struct parse *p, register cset *cs); - */ -static void -p_b_cclass(p, cs) -register struct parse *p; -register cset *cs; -{ - register char *sp = p->next; - register struct cclass *cp; - register size_t len; - register char *u; - register char c; - - while (MORE() && isalpha(PEEK())) - NEXT(); - len = p->next - sp; - for (cp = cclasses; cp->name != NULL; cp++) - if (strncmp(cp->name, sp, len) == 0 && cp->name[len] == '\0') - break; - if (cp->name == NULL) { - /* oops, didn't find it */ - SETERROR(REG_ECTYPE); - return; - } - - u = cp->chars; - while ((c = *u++) != '\0') - CHadd(cs, c); - for (u = cp->multis; *u != '\0'; u += strlen(u) + 1) - MCadd(p, cs, u); -} - -/* - - p_b_eclass - parse an equivalence-class name and deal with it - == static void p_b_eclass(register struct parse *p, register cset *cs); - * - * This implementation is incomplete. xxx - */ -static void -p_b_eclass(p, cs) -register struct parse *p; -register cset *cs; -{ - register char c; - - c = p_b_coll_elem(p, '='); - CHadd(cs, c); -} - -/* - - p_b_symbol - parse a character or [..]ed multicharacter collating symbol - == static char p_b_symbol(register struct parse *p); - */ -static char /* value of symbol */ -p_b_symbol(p) -register struct parse *p; -{ - register char value; - - REQUIRE(MORE(), REG_EBRACK); - if (!EATTWO('[', '.')) - return(GETNEXT()); - - /* collating symbol */ - value = p_b_coll_elem(p, '.'); - REQUIRE(EATTWO('.', ']'), REG_ECOLLATE); - return(value); -} - -/* - - p_b_coll_elem - parse a collating-element name and look it up - == static char p_b_coll_elem(register struct parse *p, int endc); - */ -static char /* value of collating element */ -p_b_coll_elem(p, endc) -register struct parse *p; -int endc; /* name ended by endc,']' */ -{ - register char *sp = p->next; - register struct cname *cp; - register int len; - - while (MORE() && !SEETWO(endc, ']')) - NEXT(); - if (!MORE()) { - SETERROR(REG_EBRACK); - return(0); - } - len = p->next - sp; - for (cp = cnames; cp->name != NULL; cp++) - if (strncmp(cp->name, sp, len) == 0 && cp->name[len] == '\0') - return(cp->code); /* known name */ - if (len == 1) - return(*sp); /* single character */ - SETERROR(REG_ECOLLATE); /* neither */ - return(0); -} - -/* - - othercase - return the case counterpart of an alphabetic - == static char othercase(int ch); - */ -static char /* if no counterpart, return ch */ -othercase(ch) -int ch; -{ - assert(isalpha(ch)); - if (isupper(ch)) - return(tolower(ch)); - else if (islower(ch)) - return(toupper(ch)); - else /* peculiar, but could happen */ - return(ch); -} - -/* - - bothcases - emit a dualcase version of a two-case character - == static void bothcases(register struct parse *p, int ch); - * - * Boy, is this implementation ever a kludge... - */ -static void -bothcases(p, ch) -register struct parse *p; -int ch; -{ - register char *oldnext = p->next; - register char *oldend = p->end; - char bracket[3]; - - assert(othercase(ch) != ch); /* p_bracket() would recurse */ - p->next = bracket; - p->end = bracket+2; - bracket[0] = ch; - bracket[1] = ']'; - bracket[2] = '\0'; - p_bracket(p); - assert(p->next == bracket+2); - p->next = oldnext; - p->end = oldend; -} - -/* - - ordinary - emit an ordinary character - == static void ordinary(register struct parse *p, register int ch); - */ -static void -ordinary(p, ch) -register struct parse *p; -register int ch; -{ - register cat_t *cap = p->g->categories; - - if ((p->g->cflags®_ICASE) && isalpha(ch) && othercase(ch) != ch) - bothcases(p, ch); - else { - EMIT(OCHAR, (unsigned char)ch); - if (cap[ch] == 0) - cap[ch] = p->g->ncategories++; - } -} - -/* - - nonnewline - emit REG_NEWLINE version of OANY - == static void nonnewline(register struct parse *p); - * - * Boy, is this implementation ever a kludge... - */ -static void -nonnewline(p) -register struct parse *p; -{ - register char *oldnext = p->next; - register char *oldend = p->end; - char bracket[4]; - - p->next = bracket; - p->end = bracket+3; - bracket[0] = '^'; - bracket[1] = '\n'; - bracket[2] = ']'; - bracket[3] = '\0'; - p_bracket(p); - assert(p->next == bracket+3); - p->next = oldnext; - p->end = oldend; -} - -/* - - repeat - generate code for a bounded repetition, recursively if needed - == static void repeat(register struct parse *p, sopno start, int from, int to); - */ -static void -repeat(p, start, from, to) -register struct parse *p; -sopno start; /* operand from here to end of strip */ -int from; /* repeated from this number */ -int to; /* to this number of times (maybe INFINITY) */ -{ - register sopno finish = HERE(); -# define N 2 -# define INF 3 -# define REP(f, t) ((f)*8 + (t)) -# define MAP(n) (((n) <= 1) ? (n) : ((n) == INFINITY) ? INF : N) - register sopno copy; - - if (p->error != 0) /* head off possible runaway recursion */ - return; - - assert(from <= to); - - switch (REP(MAP(from), MAP(to))) { - case REP(0, 0): /* must be user doing this */ - DROP(finish-start); /* drop the operand */ - break; - case REP(0, 1): /* as x{1,1}? */ - case REP(0, N): /* as x{1,n}? */ - case REP(0, INF): /* as x{1,}? */ - /* KLUDGE: emit y? as (y|) until subtle bug gets fixed */ - INSERT(OCH_, start); /* offset is wrong... */ - repeat(p, start+1, 1, to); - ASTERN(OOR1, start); - AHEAD(start); /* ... fix it */ - EMIT(OOR2, 0); - AHEAD(THERE()); - ASTERN(O_CH, THERETHERE()); - break; - case REP(1, 1): /* trivial case */ - /* done */ - break; - case REP(1, N): /* as x?x{1,n-1} */ - /* KLUDGE: emit y? as (y|) until subtle bug gets fixed */ - INSERT(OCH_, start); - ASTERN(OOR1, start); - AHEAD(start); - EMIT(OOR2, 0); /* offset very wrong... */ - AHEAD(THERE()); /* ...so fix it */ - ASTERN(O_CH, THERETHERE()); - copy = dupl(p, start+1, finish+1); - assert(copy == finish+4); - repeat(p, copy, 1, to-1); - break; - case REP(1, INF): /* as x+ */ - INSERT(OPLUS_, start); - ASTERN(O_PLUS, start); - break; - case REP(N, N): /* as xx{m-1,n-1} */ - copy = dupl(p, start, finish); - repeat(p, copy, from-1, to-1); - break; - case REP(N, INF): /* as xx{n-1,INF} */ - copy = dupl(p, start, finish); - repeat(p, copy, from-1, to); - break; - default: /* "can't happen" */ - SETERROR(REG_ASSERT); /* just in case */ - break; - } -} - -/* - - seterr - set an error condition - == static int seterr(register struct parse *p, int e); - */ -static int /* useless but makes type checking happy */ -seterr(p, e) -register struct parse *p; -int e; -{ - if (p->error == 0) /* keep earliest error condition */ - p->error = e; - p->next = nuls; /* try to bring things to a halt */ - p->end = nuls; - return(0); /* make the return value well-defined */ -} - -/* - - allocset - allocate a set of characters for [] - == static cset *allocset(register struct parse *p); - */ -static cset * -allocset(p) -register struct parse *p; -{ - register int no = p->g->ncsets++; - register size_t nc; - register size_t nbytes; - register cset *cs; - register size_t css = (size_t)p->g->csetsize; - register int i; - - if (no >= p->ncsalloc) { /* need another column of space */ - p->ncsalloc += CHAR_BIT; - nc = p->ncsalloc; - assert(nc % CHAR_BIT == 0); - nbytes = nc / CHAR_BIT * css; - if (p->g->sets == NULL) - p->g->sets = (cset *)malloc(nc * sizeof(cset)); - else - p->g->sets = (cset *)realloc((char *)p->g->sets, - nc * sizeof(cset)); - if (p->g->setbits == NULL) - p->g->setbits = (uch *)malloc(nbytes); - else { - p->g->setbits = (uch *)realloc((char *)p->g->setbits, - nbytes); - /* xxx this isn't right if setbits is now NULL */ - for (i = 0; i < no; i++) - p->g->sets[i].ptr = p->g->setbits + css*(i/CHAR_BIT); - } - if (p->g->sets != NULL && p->g->setbits != NULL) - (void) memset((char *)p->g->setbits + (nbytes - css), - 0, css); - else { - no = 0; - SETERROR(REG_ESPACE); - /* caller's responsibility not to do set ops */ - } - } - - assert(p->g->sets != NULL); /* xxx */ - cs = &p->g->sets[no]; - cs->ptr = p->g->setbits + css*((no)/CHAR_BIT); - cs->mask = 1 << ((no) % CHAR_BIT); - cs->hash = 0; - cs->smultis = 0; - cs->multis = NULL; - - return(cs); -} - -/* - - freeset - free a now-unused set - == static void freeset(register struct parse *p, register cset *cs); - */ -static void -freeset(p, cs) -register struct parse *p; -register cset *cs; -{ - register int i; - register cset *top = &p->g->sets[p->g->ncsets]; - register size_t css = (size_t)p->g->csetsize; - - for (i = 0; i < css; i++) - CHsub(cs, i); - if (cs == top-1) /* recover only the easy case */ - p->g->ncsets--; -} - -/* - - freezeset - final processing on a set of characters - == static int freezeset(register struct parse *p, register cset *cs); - * - * The main task here is merging identical sets. This is usually a waste - * of time (although the hash code minimizes the overhead), but can win - * big if REG_ICASE is being used. REG_ICASE, by the way, is why the hash - * is done using addition rather than xor -- all ASCII [aA] sets xor to - * the same value! - */ -static int /* set number */ -freezeset(p, cs) -register struct parse *p; -register cset *cs; -{ - register uch h = cs->hash; - register int i; - register cset *top = &p->g->sets[p->g->ncsets]; - register cset *cs2; - register size_t css = (size_t)p->g->csetsize; - - /* look for an earlier one which is the same */ - for (cs2 = &p->g->sets[0]; cs2 < top; cs2++) - if (cs2->hash == h && cs2 != cs) { - /* maybe */ - for (i = 0; i < css; i++) - if (!!CHIN(cs2, i) != !!CHIN(cs, i)) - break; /* no */ - if (i == css) - break; /* yes */ - } - - if (cs2 < top) { /* found one */ - freeset(p, cs); - cs = cs2; - } - - return((int)(cs - p->g->sets)); -} - -/* - - firstch - return first character in a set (which must have at least one) - == static int firstch(register struct parse *p, register cset *cs); - */ -static int /* character; there is no "none" value */ -firstch(p, cs) -register struct parse *p; -register cset *cs; -{ - register int i; - register size_t css = (size_t)p->g->csetsize; - - for (i = 0; i < css; i++) - if (CHIN(cs, i)) - return((char)i); - assert(never); - return(0); /* arbitrary */ -} - -/* - - nch - number of characters in a set - == static int nch(register struct parse *p, register cset *cs); - */ -static int -nch(p, cs) -register struct parse *p; -register cset *cs; -{ - register int i; - register size_t css = (size_t)p->g->csetsize; - register int n = 0; - - for (i = 0; i < css; i++) - if (CHIN(cs, i)) - n++; - return(n); -} - -/* - - mcadd - add a collating element to a cset - == static void mcadd(register struct parse *p, register cset *cs, \ - == register char *cp); - */ -static void -mcadd(p, cs, cp) -register struct parse *p; -register cset *cs; -register char *cp; -{ - register size_t oldend = cs->smultis; - - cs->smultis += strlen(cp) + 1; - if (cs->multis == NULL) - cs->multis = malloc(cs->smultis); - else - cs->multis = realloc(cs->multis, cs->smultis); - if (cs->multis == NULL) { - SETERROR(REG_ESPACE); - return; - } - - (void) strcpy(cs->multis + oldend - 1, cp); - cs->multis[cs->smultis - 1] = '\0'; -} - -/* - - mcinvert - invert the list of collating elements in a cset - == static void mcinvert(register struct parse *p, register cset *cs); - * - * This would have to know the set of possibilities. Implementation - * is deferred. - */ -static void -mcinvert(p, cs) -register struct parse *p; -register cset *cs; -{ - assert(cs->multis == NULL); /* xxx */ -} - -/* - - mccase - add case counterparts of the list of collating elements in a cset - == static void mccase(register struct parse *p, register cset *cs); - * - * This would have to know the set of possibilities. Implementation - * is deferred. - */ -static void -mccase(p, cs) -register struct parse *p; -register cset *cs; -{ - assert(cs->multis == NULL); /* xxx */ -} - -/* - - isinsets - is this character in any sets? - == static int isinsets(register struct re_guts *g, int c); - */ -static int /* predicate */ -isinsets(g, c) -register struct re_guts *g; -int c; -{ - register uch *col; - register int i; - register int ncols = (g->ncsets+(CHAR_BIT-1)) / CHAR_BIT; - register unsigned uc = (unsigned char)c; - - for (i = 0, col = g->setbits; i < ncols; i++, col += g->csetsize) - if (col[uc] != 0) - return(1); - return(0); -} - -/* - - samesets - are these two characters in exactly the same sets? - == static int samesets(register struct re_guts *g, int c1, int c2); - */ -static int /* predicate */ -samesets(g, c1, c2) -register struct re_guts *g; -int c1; -int c2; -{ - register uch *col; - register int i; - register int ncols = (g->ncsets+(CHAR_BIT-1)) / CHAR_BIT; - register unsigned uc1 = (unsigned char)c1; - register unsigned uc2 = (unsigned char)c2; - - for (i = 0, col = g->setbits; i < ncols; i++, col += g->csetsize) - if (col[uc1] != col[uc2]) - return(0); - return(1); -} - -/* - - categorize - sort out character categories - == static void categorize(struct parse *p, register struct re_guts *g); - */ -static void -categorize(p, g) -struct parse *p; -register struct re_guts *g; -{ - register cat_t *cats = g->categories; - register int c; - register int c2; - register cat_t cat; - - /* avoid making error situations worse */ - if (p->error != 0) - return; - - for (c = CHAR_MIN; c <= CHAR_MAX; c++) - if (cats[c] == 0 && isinsets(g, c)) { - cat = g->ncategories++; - cats[c] = cat; - for (c2 = c+1; c2 <= CHAR_MAX; c2++) - if (cats[c2] == 0 && samesets(g, c, c2)) - cats[c2] = cat; - } -} - -/* - - dupl - emit a duplicate of a bunch of sops - == static sopno dupl(register struct parse *p, sopno start, sopno finish); - */ -static sopno /* start of duplicate */ -dupl(p, start, finish) -register struct parse *p; -sopno start; /* from here */ -sopno finish; /* to this less one */ -{ - register sopno ret = HERE(); - register sopno len = finish - start; - - assert(finish >= start); - if (len == 0) - return(ret); - enlarge(p, p->ssize + len); /* this many unexpected additions */ - assert(p->ssize >= p->slen + len); - (void) memcpy((char *)(p->strip + p->slen), - (char *)(p->strip + start), (size_t)len*sizeof(sop)); - p->slen += len; - return(ret); -} - -/* - - doemit - emit a strip operator - == static void doemit(register struct parse *p, sop op, size_t opnd); - * - * It might seem better to implement this as a macro with a function as - * hard-case backup, but it's just too big and messy unless there are - * some changes to the data structures. Maybe later. - */ -static void -doemit(p, op, opnd) -register struct parse *p; -sop op; -size_t opnd; -{ - /* avoid making error situations worse */ - if (p->error != 0) - return; - - /* deal with oversize operands ("can't happen", more or less) */ - assert(opnd < 1<slen >= p->ssize) - enlarge(p, (p->ssize+1) / 2 * 3); /* +50% */ - assert(p->slen < p->ssize); - - /* finally, it's all reduced to the easy case */ - p->strip[p->slen++] = SOP(op, opnd); -} - -/* - - doinsert - insert a sop into the strip - == static void doinsert(register struct parse *p, sop op, size_t opnd, sopno pos); - */ -static void -doinsert(p, op, opnd, pos) -register struct parse *p; -sop op; -size_t opnd; -sopno pos; -{ - register sopno sn; - register sop s; - register int i; - - /* avoid making error situations worse */ - if (p->error != 0) - return; - - sn = HERE(); - EMIT(op, opnd); /* do checks, ensure space */ - assert(HERE() == sn+1); - s = p->strip[sn]; - - /* adjust paren pointers */ - assert(pos > 0); - for (i = 1; i < NPAREN; i++) { - if (p->pbegin[i] >= pos) { - p->pbegin[i]++; - } - if (p->pend[i] >= pos) { - p->pend[i]++; - } - } - - memmove((char *)&p->strip[pos+1], (char *)&p->strip[pos], - (HERE()-pos-1)*sizeof(sop)); - p->strip[pos] = s; -} - -/* - - dofwd - complete a forward reference - == static void dofwd(register struct parse *p, sopno pos, sop value); - */ -static void -dofwd(p, pos, value) -register struct parse *p; -register sopno pos; -sop value; -{ - /* avoid making error situations worse */ - if (p->error != 0) - return; - - assert(value < 1<strip[pos] = OP(p->strip[pos]) | value; -} - -/* - - enlarge - enlarge the strip - == static void enlarge(register struct parse *p, sopno size); - */ -static void -enlarge(p, size) -register struct parse *p; -register sopno size; -{ - register sop *sp; - - if (p->ssize >= size) - return; - - sp = (sop *)realloc(p->strip, size*sizeof(sop)); - if (sp == NULL) { - SETERROR(REG_ESPACE); - return; - } - p->strip = sp; - p->ssize = size; -} - -/* - - stripsnug - compact the strip - == static void stripsnug(register struct parse *p, register struct re_guts *g); - */ -static void -stripsnug(p, g) -register struct parse *p; -register struct re_guts *g; -{ - g->nstates = p->slen; - g->strip = (sop *)realloc((char *)p->strip, p->slen * sizeof(sop)); - if (g->strip == NULL) { - SETERROR(REG_ESPACE); - g->strip = p->strip; - } -} - -/* - - findmust - fill in must and mlen with longest mandatory literal string - == static void findmust(register struct parse *p, register struct re_guts *g); - * - * This algorithm could do fancy things like analyzing the operands of | - * for common subsequences. Someday. This code is simple and finds most - * of the interesting cases. - * - * Note that must and mlen got initialized during setup. - */ -static void -findmust(p, g) -struct parse *p; -register struct re_guts *g; -{ - register sop *scan; - sop *start; - register sop *newstart; - register sopno newlen; - register sop s; - register char *cp; - register sopno i; - - /* avoid making error situations worse */ - if (p->error != 0) - return; - - /* find the longest OCHAR sequence in strip */ - newlen = 0; - scan = g->strip + 1; - do { - s = *scan++; - switch (OP(s)) { - case OCHAR: /* sequence member */ - if (newlen == 0) /* new sequence */ - newstart = scan - 1; - newlen++; - break; - case OPLUS_: /* things that don't break one */ - case OLPAREN: - case ORPAREN: - break; - case OQUEST_: /* things that must be skipped */ - case OCH_: - scan--; - do { - scan += OPND(s); - s = *scan; - /* assert() interferes w debug printouts */ - if (OP(s) != O_QUEST && OP(s) != O_CH && - OP(s) != OOR2) { - g->iflags |= BAD; - return; - } - } while (OP(s) != O_QUEST && OP(s) != O_CH); - /* fallthrough */ - default: /* things that break a sequence */ - if (newlen > g->mlen) { /* ends one */ - start = newstart; - g->mlen = newlen; - } - newlen = 0; - break; - } - } while (OP(s) != OEND); - - if (g->mlen == 0) /* there isn't one */ - return; - - /* turn it into a character string */ - g->must = malloc((size_t)g->mlen + 1); - if (g->must == NULL) { /* argh; just forget it */ - g->mlen = 0; - return; - } - cp = g->must; - scan = start; - for (i = g->mlen; i > 0; i--) { - while (OP(s = *scan++) != OCHAR) - continue; - assert(cp < g->must + g->mlen); - *cp++ = (char)OPND(s); - } - assert(cp == g->must + g->mlen); - *cp++ = '\0'; /* just on general principles */ -} - -/* - - pluscount - count + nesting - == static sopno pluscount(register struct parse *p, register struct re_guts *g); - */ -static sopno /* nesting depth */ -pluscount(p, g) -struct parse *p; -register struct re_guts *g; -{ - register sop *scan; - register sop s; - register sopno plusnest = 0; - register sopno maxnest = 0; - - if (p->error != 0) - return(0); /* there may not be an OEND */ - - scan = g->strip + 1; - do { - s = *scan++; - switch (OP(s)) { - case OPLUS_: - plusnest++; - break; - case O_PLUS: - if (plusnest > maxnest) - maxnest = plusnest; - plusnest--; - break; - } - } while (OP(s) != OEND); - if (plusnest != 0) - g->iflags |= BAD; - return(maxnest); -} - -static int nope = 0; /* for use in asserts; shuts lint up */ - -/* macros for manipulating states, small version */ -#define states unsigned -#define states1 unsigned /* for later use in regexec() decision */ -#define CLEAR(v) ((v) = 0) -#define SET0(v, n) ((v) &= ~((unsigned)1 << (n))) -#define SET1(v, n) ((v) |= (unsigned)1 << (n)) -#define ISSET(v, n) ((v) & ((unsigned)1 << (n))) -#define ASSIGN(d, s) ((d) = (s)) -#define EQ(a, b) ((a) == (b)) -#define STATEVARS int dummy /* dummy version */ -#define STATESETUP(m, n) /* nothing */ -#define STATETEARDOWN(m) /* nothing */ -#define SETUP(v) ((v) = 0) -#define onestate unsigned -#define INIT(o, n) ((o) = (unsigned)1 << (n)) -#define INC(o) ((o) <<= 1) -#define ISSTATEIN(v, o) ((v) & (o)) -/* some abbreviations; note that some of these know variable names! */ -/* do "if I'm here, I can also be there" etc without branches */ -#define FWD(dst, src, n) ((dst) |= ((unsigned)(src)&(here)) << (n)) -#define BACK(dst, src, n) ((dst) |= ((unsigned)(src)&(here)) >> (n)) -#define ISSETBACK(v, n) ((v) & ((unsigned)here >> (n))) -/* function names */ -#define SNAMES /* engine.c looks after details */ - -/* - * The matching engine and friends. This file is #included by regexec.c - * after suitable #defines of a variety of macros used herein, so that - * different state representations can be used without duplicating masses - * of code. - */ - -#ifdef SNAMES -#define matcher smatcher -#define fast sfast -#define slow sslow -#define dissect sdissect -#define backref sbackref -#define step sstep -#define print sprint -#define at sat -#define match smat -#endif -#ifdef LNAMES -#define matcher lmatcher -#define fast lfast -#define slow lslow -#define dissect ldissect -#define backref lbackref -#define step lstep -#define print lprint -#define at lat -#define match lmat -#endif - -/* another structure passed up and down to avoid zillions of parameters */ -struct match { - struct re_guts *g; - int eflags; - regmatch_t *pmatch; /* [nsub+1] (0 element unused) */ - char *offp; /* offsets work from here */ - char *beginp; /* start of string -- virtual NUL precedes */ - char *endp; /* end of string -- virtual NUL here */ - char *coldp; /* can be no match starting before here */ - char **lastpos; /* [nplus+1] */ - STATEVARS; - states st; /* current states */ - states fresh; /* states for a fresh start */ - states tmp; /* temporary */ - states empty; /* empty set of states */ -}; - -static int matcher(register struct re_guts *g, char *string, size_t nmatch, regmatch_t pmatch[], int eflags); -static char *dissect(register struct match *m, char *start, char *stop, sopno startst, sopno stopst); -static char *backref(register struct match *m, char *start, char *stop, sopno startst, sopno stopst, sopno lev); -static char *fast(register struct match *m, char *start, char *stop, sopno startst, sopno stopst); -static char *slow(register struct match *m, char *start, char *stop, sopno startst, sopno stopst); -static states step(register struct re_guts *g, sopno start, sopno stop, register states bef, int ch, register states aft); -#define BOL (OUT+1) -#define EOL (BOL+1) -#define BOLEOL (BOL+2) -#define NOTHING (BOL+3) -#define BOW (BOL+4) -#define EOW (BOL+5) -#define CODEMAX (BOL+5) /* highest code used */ -#define NONCHAR(c) ((c) > CHAR_MAX) -#define NNONCHAR (CODEMAX-CHAR_MAX) -#define SP(t, s, c) /* nothing */ -#define AT(t, p1, p2, s1, s2) /* nothing */ -#define NOTE(s) /* nothing */ - -/* - - matcher - the actual matching engine - == static int matcher(register struct re_guts *g, char *string, \ - == size_t nmatch, regmatch_t pmatch[], int eflags); - */ -static int /* 0 success, REG_NOMATCH failure */ -matcher(g, string, nmatch, pmatch, eflags) -register struct re_guts *g; -char *string; -size_t nmatch; -regmatch_t pmatch[]; -int eflags; -{ - register char *endp; - register int i; - struct match mv; - register struct match *m = &mv; - register char *dp; - const register sopno gf = g->firststate+1; /* +1 for OEND */ - const register sopno gl = g->laststate; - char *start; - char *stop; - - /* simplify the situation where possible */ - if (g->cflags®_NOSUB) - nmatch = 0; - if (eflags®_STARTEND) { - start = string + pmatch[0].rm_so; - stop = string + pmatch[0].rm_eo; - } else { - start = string; - stop = start + strlen(start); - } - if (stop < start) - return(REG_INVARG); - - /* prescreening; this does wonders for this rather slow code */ - if (g->must != NULL) { - for (dp = start; dp < stop; dp++) - if (*dp == g->must[0] && stop - dp >= g->mlen && - memcmp(dp, g->must, (size_t)g->mlen) == 0) - break; - if (dp == stop) /* we didn't find g->must */ - return(REG_NOMATCH); - } - - /* match struct setup */ - m->g = g; - m->eflags = eflags; - m->pmatch = NULL; - m->lastpos = NULL; - m->offp = string; - m->beginp = start; - m->endp = stop; - STATESETUP(m, 4); - SETUP(m->st); - SETUP(m->fresh); - SETUP(m->tmp); - SETUP(m->empty); - CLEAR(m->empty); - - /* this loop does only one repetition except for backrefs */ - for (;;) { - endp = fast(m, start, stop, gf, gl); - if (endp == NULL) { /* a miss */ - STATETEARDOWN(m); - return(REG_NOMATCH); - } - if (nmatch == 0 && !g->backrefs) - break; /* no further info needed */ - - /* where? */ - assert(m->coldp != NULL); - for (;;) { - NOTE("finding start"); - endp = slow(m, m->coldp, stop, gf, gl); - if (endp != NULL) - break; - assert(m->coldp < m->endp); - m->coldp++; - } - if (nmatch == 1 && !g->backrefs) - break; /* no further info needed */ - - /* oh my, he wants the subexpressions... */ - if (m->pmatch == NULL) - m->pmatch = (regmatch_t *)malloc((m->g->nsub + 1) * - sizeof(regmatch_t)); - if (m->pmatch == NULL) { - STATETEARDOWN(m); - return(REG_ESPACE); - } - for (i = 1; i <= m->g->nsub; i++) - m->pmatch[i].rm_so = m->pmatch[i].rm_eo = -1; - if (!g->backrefs && !(m->eflags®_BACKR)) { - NOTE("dissecting"); - dp = dissect(m, m->coldp, endp, gf, gl); - } else { - if (g->nplus > 0 && m->lastpos == NULL) - m->lastpos = (char **)malloc((g->nplus+1) * - sizeof(char *)); - if (g->nplus > 0 && m->lastpos == NULL) { - free(m->pmatch); - STATETEARDOWN(m); - return(REG_ESPACE); - } - NOTE("backref dissect"); - dp = backref(m, m->coldp, endp, gf, gl, (sopno)0); - } - if (dp != NULL) - break; - - /* uh-oh... we couldn't find a subexpression-level match */ - assert(g->backrefs); /* must be back references doing it */ - assert(g->nplus == 0 || m->lastpos != NULL); - for (;;) { - if (dp != NULL || endp <= m->coldp) - break; /* defeat */ - NOTE("backoff"); - endp = slow(m, m->coldp, endp-1, gf, gl); - if (endp == NULL) - break; /* defeat */ - /* try it on a shorter possibility */ - NOTE("backoff dissect"); - dp = backref(m, m->coldp, endp, gf, gl, (sopno)0); - } - assert(dp == NULL || dp == endp); - if (dp != NULL) /* found a shorter one */ - break; - - /* despite initial appearances, there is no match here */ - NOTE("false alarm"); - start = m->coldp + 1; /* recycle starting later */ - assert(start <= stop); - } - - /* fill in the details if requested */ - if (nmatch > 0) { - pmatch[0].rm_so = m->coldp - m->offp; - pmatch[0].rm_eo = endp - m->offp; - } - if (nmatch > 1) { - assert(m->pmatch != NULL); - for (i = 1; i < nmatch; i++) - if (i <= m->g->nsub) - pmatch[i] = m->pmatch[i]; - else { - pmatch[i].rm_so = -1; - pmatch[i].rm_eo = -1; - } - } - - if (m->pmatch != NULL) - free((char *)m->pmatch); - if (m->lastpos != NULL) - free((char *)m->lastpos); - STATETEARDOWN(m); - return(0); -} - -/* - - dissect - figure out what matched what, no back references - == static char *dissect(register struct match *m, char *start, \ - == char *stop, sopno startst, sopno stopst); - */ -static char * /* == stop (success) always */ -dissect(m, start, stop, startst, stopst) -register struct match *m; -char *start; -char *stop; -sopno startst; -sopno stopst; -{ - register int i; - register sopno ss; /* start sop of current subRE */ - register sopno es; /* end sop of current subRE */ - register char *sp; /* start of string matched by it */ - register char *stp; /* string matched by it cannot pass here */ - register char *rest; /* start of rest of string */ - register char *tail; /* string unmatched by rest of RE */ - register sopno ssub; /* start sop of subsubRE */ - register sopno esub; /* end sop of subsubRE */ - register char *ssp; /* start of string matched by subsubRE */ - register char *sep; /* end of string matched by subsubRE */ - register char *oldssp; /* previous ssp */ - register char *dp; - - AT("diss", start, stop, startst, stopst); - sp = start; - for (ss = startst; ss < stopst; ss = es) { - /* identify end of subRE */ - es = ss; - switch (OP(m->g->strip[es])) { - case OPLUS_: - case OQUEST_: - es += OPND(m->g->strip[es]); - break; - case OCH_: - while (OP(m->g->strip[es]) != O_CH) - es += OPND(m->g->strip[es]); - break; - } - es++; - - /* figure out what it matched */ - switch (OP(m->g->strip[ss])) { - case OEND: - assert(nope); - break; - case OCHAR: - sp++; - break; - case OBOL: - case OEOL: - case OBOW: - case OEOW: - break; - case OANY: - case OANYOF: - sp++; - break; - case OBACK_: - case O_BACK: - assert(nope); - break; - /* cases where length of match is hard to find */ - case OQUEST_: - stp = stop; - for (;;) { - /* how long could this one be? */ - rest = slow(m, sp, stp, ss, es); - assert(rest != NULL); /* it did match */ - /* could the rest match the rest? */ - tail = slow(m, rest, stop, es, stopst); - if (tail == stop) - break; /* yes! */ - /* no -- try a shorter match for this one */ - stp = rest - 1; - assert(stp >= sp); /* it did work */ - } - ssub = ss + 1; - esub = es - 1; - /* did innards match? */ - if (slow(m, sp, rest, ssub, esub) != NULL) { - dp = dissect(m, sp, rest, ssub, esub); - assert(dp == rest); - } else /* no */ - assert(sp == rest); - sp = rest; - break; - case OPLUS_: - stp = stop; - for (;;) { - /* how long could this one be? */ - rest = slow(m, sp, stp, ss, es); - assert(rest != NULL); /* it did match */ - /* could the rest match the rest? */ - tail = slow(m, rest, stop, es, stopst); - if (tail == stop) - break; /* yes! */ - /* no -- try a shorter match for this one */ - stp = rest - 1; - assert(stp >= sp); /* it did work */ - } - ssub = ss + 1; - esub = es - 1; - ssp = sp; - oldssp = ssp; - for (;;) { /* find last match of innards */ - sep = slow(m, ssp, rest, ssub, esub); - if (sep == NULL || sep == ssp) - break; /* failed or matched null */ - oldssp = ssp; /* on to next try */ - ssp = sep; - } - if (sep == NULL) { - /* last successful match */ - sep = ssp; - ssp = oldssp; - } - assert(sep == rest); /* must exhaust substring */ - assert(slow(m, ssp, sep, ssub, esub) == rest); - dp = dissect(m, ssp, sep, ssub, esub); - assert(dp == sep); - sp = rest; - break; - case OCH_: - stp = stop; - for (;;) { - /* how long could this one be? */ - rest = slow(m, sp, stp, ss, es); - assert(rest != NULL); /* it did match */ - /* could the rest match the rest? */ - tail = slow(m, rest, stop, es, stopst); - if (tail == stop) - break; /* yes! */ - /* no -- try a shorter match for this one */ - stp = rest - 1; - assert(stp >= sp); /* it did work */ - } - ssub = ss + 1; - esub = ss + OPND(m->g->strip[ss]) - 1; - assert(OP(m->g->strip[esub]) == OOR1); - for (;;) { /* find first matching branch */ - if (slow(m, sp, rest, ssub, esub) == rest) - break; /* it matched all of it */ - /* that one missed, try next one */ - assert(OP(m->g->strip[esub]) == OOR1); - esub++; - assert(OP(m->g->strip[esub]) == OOR2); - ssub = esub + 1; - esub += OPND(m->g->strip[esub]); - if (OP(m->g->strip[esub]) == OOR2) - esub--; - else - assert(OP(m->g->strip[esub]) == O_CH); - } - dp = dissect(m, sp, rest, ssub, esub); - assert(dp == rest); - sp = rest; - break; - case O_PLUS: - case O_QUEST: - case OOR1: - case OOR2: - case O_CH: - assert(nope); - break; - case OLPAREN: - i = OPND(m->g->strip[ss]); - assert(0 < i && i <= m->g->nsub); - m->pmatch[i].rm_so = sp - m->offp; - break; - case ORPAREN: - i = OPND(m->g->strip[ss]); - assert(0 < i && i <= m->g->nsub); - m->pmatch[i].rm_eo = sp - m->offp; - break; - default: /* uh oh */ - assert(nope); - break; - } - } - - assert(sp == stop); - return(sp); -} - -/* - - backref - figure out what matched what, figuring in back references - == static char *backref(register struct match *m, char *start, \ - == char *stop, sopno startst, sopno stopst, sopno lev); - */ -static char * /* == stop (success) or NULL (failure) */ -backref(m, start, stop, startst, stopst, lev) -register struct match *m; -char *start; -char *stop; -sopno startst; -sopno stopst; -sopno lev; /* PLUS nesting level */ -{ - register int i; - register sopno ss; /* start sop of current subRE */ - register char *sp; /* start of string matched by it */ - register sopno ssub; /* start sop of subsubRE */ - register sopno esub; /* end sop of subsubRE */ - register char *ssp; /* start of string matched by subsubRE */ - register char *dp; - register size_t len; - register int hard; - register sop s; - register regoff_t offsave; - register cset *cs; - - AT("back", start, stop, startst, stopst); - sp = start; - - /* get as far as we can with easy stuff */ - hard = 0; - for (ss = startst; !hard && ss < stopst; ss++) - switch (OP(s = m->g->strip[ss])) { - case OCHAR: - if (sp == stop || *sp++ != (char)OPND(s)) - return(NULL); - break; - case OANY: - if (sp == stop) - return(NULL); - sp++; - break; - case OANYOF: - cs = &m->g->sets[OPND(s)]; - if (sp == stop || !CHIN(cs, *sp++)) - return(NULL); - break; - case OBOL: - if ( (sp == m->beginp && !(m->eflags®_NOTBOL)) || - (sp < m->endp && *(sp-1) == '\n' && - (m->g->cflags®_NEWLINE)) ) - { /* yes */ } - else - return(NULL); - break; - case OEOL: - if ( (sp == m->endp && !(m->eflags®_NOTEOL)) || - (sp < m->endp && *sp == '\n' && - (m->g->cflags®_NEWLINE)) ) - { /* yes */ } - else - return(NULL); - break; - case OBOW: - if (( (sp == m->beginp && !(m->eflags®_NOTBOL)) || - (sp < m->endp && *(sp-1) == '\n' && - (m->g->cflags®_NEWLINE)) || - (sp > m->beginp && - !ISWORD(*(sp-1))) ) && - (sp < m->endp && ISWORD(*sp)) ) - { /* yes */ } - else - return(NULL); - break; - case OEOW: - if (( (sp == m->endp && !(m->eflags®_NOTEOL)) || - (sp < m->endp && *sp == '\n' && - (m->g->cflags®_NEWLINE)) || - (sp < m->endp && !ISWORD(*sp)) ) && - (sp > m->beginp && ISWORD(*(sp-1))) ) - { /* yes */ } - else - return(NULL); - break; - case O_QUEST: - break; - case OOR1: /* matches null but needs to skip */ - ss++; - s = m->g->strip[ss]; - do { - assert(OP(s) == OOR2); - ss += OPND(s); - } while (OP(s = m->g->strip[ss]) != O_CH); - /* note that the ss++ gets us past the O_CH */ - break; - default: /* have to make a choice */ - hard = 1; - break; - } - if (!hard) { /* that was it! */ - if (sp != stop) - return(NULL); - return(sp); - } - ss--; /* adjust for the for's final increment */ - - /* the hard stuff */ - AT("hard", sp, stop, ss, stopst); - s = m->g->strip[ss]; - switch (OP(s)) { - case OBACK_: /* the vilest depths */ - i = OPND(s); - assert(0 < i && i <= m->g->nsub); - if (m->pmatch[i].rm_eo == -1) - return(NULL); - assert(m->pmatch[i].rm_so != -1); - len = m->pmatch[i].rm_eo - m->pmatch[i].rm_so; - assert(stop - m->beginp >= len); - if (sp > stop - len) - return(NULL); /* not enough left to match */ - ssp = m->offp + m->pmatch[i].rm_so; - if (memcmp(sp, ssp, len) != 0) - return(NULL); - while (m->g->strip[ss] != SOP(O_BACK, i)) - ss++; - return(backref(m, sp+len, stop, ss+1, stopst, lev)); - break; - case OQUEST_: /* to null or not */ - dp = backref(m, sp, stop, ss+1, stopst, lev); - if (dp != NULL) - return(dp); /* not */ - return(backref(m, sp, stop, ss+OPND(s)+1, stopst, lev)); - break; - case OPLUS_: - assert(m->lastpos != NULL); - assert(lev+1 <= m->g->nplus); - m->lastpos[lev+1] = sp; - return(backref(m, sp, stop, ss+1, stopst, lev+1)); - break; - case O_PLUS: - if (sp == m->lastpos[lev]) /* last pass matched null */ - return(backref(m, sp, stop, ss+1, stopst, lev-1)); - /* try another pass */ - m->lastpos[lev] = sp; - dp = backref(m, sp, stop, ss-OPND(s)+1, stopst, lev); - if (dp == NULL) - return(backref(m, sp, stop, ss+1, stopst, lev-1)); - else - return(dp); - break; - case OCH_: /* find the right one, if any */ - ssub = ss + 1; - esub = ss + OPND(s) - 1; - assert(OP(m->g->strip[esub]) == OOR1); - for (;;) { /* find first matching branch */ - dp = backref(m, sp, stop, ssub, esub, lev); - if (dp != NULL) - return(dp); - /* that one missed, try next one */ - if (OP(m->g->strip[esub]) == O_CH) - return(NULL); /* there is none */ - esub++; - assert(OP(m->g->strip[esub]) == OOR2); - ssub = esub + 1; - esub += OPND(m->g->strip[esub]); - if (OP(m->g->strip[esub]) == OOR2) - esub--; - else - assert(OP(m->g->strip[esub]) == O_CH); - } - break; - case OLPAREN: /* must undo assignment if rest fails */ - i = OPND(s); - assert(0 < i && i <= m->g->nsub); - offsave = m->pmatch[i].rm_so; - m->pmatch[i].rm_so = sp - m->offp; - dp = backref(m, sp, stop, ss+1, stopst, lev); - if (dp != NULL) - return(dp); - m->pmatch[i].rm_so = offsave; - return(NULL); - break; - case ORPAREN: /* must undo assignment if rest fails */ - i = OPND(s); - assert(0 < i && i <= m->g->nsub); - offsave = m->pmatch[i].rm_eo; - m->pmatch[i].rm_eo = sp - m->offp; - dp = backref(m, sp, stop, ss+1, stopst, lev); - if (dp != NULL) - return(dp); - m->pmatch[i].rm_eo = offsave; - return(NULL); - break; - default: /* uh oh */ - assert(nope); - break; - } - - /* "can't happen" */ - assert(nope); - /* NOTREACHED */ - return((char *)NULL); /* dummy */ -} - -/* - - fast - step through the string at top speed - == static char *fast(register struct match *m, char *start, \ - == char *stop, sopno startst, sopno stopst); - */ -static char * /* where tentative match ended, or NULL */ -fast(m, start, stop, startst, stopst) -register struct match *m; -char *start; -char *stop; -sopno startst; -sopno stopst; -{ - register states st = m->st; - register states fresh = m->fresh; - register states tmp = m->tmp; - register char *p = start; - register int c = (start == m->beginp) ? OUT : *(start-1); - register int lastc; /* previous c */ - register int flagch; - register int i; - register char *coldp; /* last p after which no match was underway */ - - CLEAR(st); - SET1(st, startst); - st = step(m->g, startst, stopst, st, NOTHING, st); - ASSIGN(fresh, st); - SP("start", st, *p); - coldp = NULL; - for (;;) { - /* next character */ - lastc = c; - c = (p == m->endp) ? OUT : *p; - if (EQ(st, fresh)) - coldp = p; - - /* is there an EOL and/or BOL between lastc and c? */ - flagch = '\0'; - i = 0; - if ( (lastc == '\n' && m->g->cflags®_NEWLINE) || - (lastc == OUT && !(m->eflags®_NOTBOL)) ) { - flagch = BOL; - i = m->g->nbol; - } - if ( (c == '\n' && m->g->cflags®_NEWLINE) || - (c == OUT && !(m->eflags®_NOTEOL)) ) { - flagch = (flagch == BOL) ? BOLEOL : EOL; - i += m->g->neol; - } - if (i != 0) { - for (; i > 0; i--) - st = step(m->g, startst, stopst, st, flagch, st); - SP("boleol", st, c); - } - - /* how about a word boundary? */ - if ( (flagch == BOL || (lastc != OUT && !ISWORD(lastc))) && - (c != OUT && ISWORD(c)) ) { - flagch = BOW; - } - if ( (lastc != OUT && ISWORD(lastc)) && - (flagch == EOL || (c != OUT && !ISWORD(c))) ) { - flagch = EOW; - } - if (flagch == BOW || flagch == EOW) { - st = step(m->g, startst, stopst, st, flagch, st); - SP("boweow", st, c); - } - - /* are we done? */ - if (ISSET(st, stopst) || p == stop) - break; /* NOTE BREAK OUT */ - - /* no, we must deal with this character */ - ASSIGN(tmp, st); - ASSIGN(st, fresh); - assert(c != OUT); - st = step(m->g, startst, stopst, tmp, c, st); - SP("aft", st, c); - assert(EQ(step(m->g, startst, stopst, st, NOTHING, st), st)); - p++; - } - - assert(coldp != NULL); - m->coldp = coldp; - if (ISSET(st, stopst)) - return(p+1); - else - return(NULL); -} - -/* - - slow - step through the string more deliberately - == static char *slow(register struct match *m, char *start, \ - == char *stop, sopno startst, sopno stopst); - */ -static char * /* where it ended */ -slow(m, start, stop, startst, stopst) -register struct match *m; -char *start; -char *stop; -sopno startst; -sopno stopst; -{ - register states st = m->st; - register states empty = m->empty; - register states tmp = m->tmp; - register char *p = start; - register int c = (start == m->beginp) ? OUT : *(start-1); - register int lastc; /* previous c */ - register int flagch; - register int i; - register char *matchp; /* last p at which a match ended */ - - AT("slow", start, stop, startst, stopst); - CLEAR(st); - SET1(st, startst); - SP("sstart", st, *p); - st = step(m->g, startst, stopst, st, NOTHING, st); - matchp = NULL; - for (;;) { - /* next character */ - lastc = c; - c = (p == m->endp) ? OUT : *p; - - /* is there an EOL and/or BOL between lastc and c? */ - flagch = '\0'; - i = 0; - if ( (lastc == '\n' && m->g->cflags®_NEWLINE) || - (lastc == OUT && !(m->eflags®_NOTBOL)) ) { - flagch = BOL; - i = m->g->nbol; - } - if ( (c == '\n' && m->g->cflags®_NEWLINE) || - (c == OUT && !(m->eflags®_NOTEOL)) ) { - flagch = (flagch == BOL) ? BOLEOL : EOL; - i += m->g->neol; - } - if (i != 0) { - for (; i > 0; i--) - st = step(m->g, startst, stopst, st, flagch, st); - SP("sboleol", st, c); - } - - /* how about a word boundary? */ - if ( (flagch == BOL || (lastc != OUT && !ISWORD(lastc))) && - (c != OUT && ISWORD(c)) ) { - flagch = BOW; - } - if ( (lastc != OUT && ISWORD(lastc)) && - (flagch == EOL || (c != OUT && !ISWORD(c))) ) { - flagch = EOW; - } - if (flagch == BOW || flagch == EOW) { - st = step(m->g, startst, stopst, st, flagch, st); - SP("sboweow", st, c); - } - - /* are we done? */ - if (ISSET(st, stopst)) - matchp = p; - if (EQ(st, empty) || p == stop) - break; /* NOTE BREAK OUT */ - - /* no, we must deal with this character */ - ASSIGN(tmp, st); - ASSIGN(st, empty); - assert(c != OUT); - st = step(m->g, startst, stopst, tmp, c, st); - SP("saft", st, c); - assert(EQ(step(m->g, startst, stopst, st, NOTHING, st), st)); - p++; - } - - return(matchp); -} - - -/* - - step - map set of states reachable before char to set reachable after - == static states step(register struct re_guts *g, sopno start, sopno stop, \ - == register states bef, int ch, register states aft); - == #define BOL (OUT+1) - == #define EOL (BOL+1) - == #define BOLEOL (BOL+2) - == #define NOTHING (BOL+3) - == #define BOW (BOL+4) - == #define EOW (BOL+5) - == #define CODEMAX (BOL+5) // highest code used - == #define NONCHAR(c) ((c) > CHAR_MAX) - == #define NNONCHAR (CODEMAX-CHAR_MAX) - */ -static states -step(g, start, stop, bef, ch, aft) -register struct re_guts *g; -sopno start; /* start state within strip */ -sopno stop; /* state after stop state within strip */ -register states bef; /* states reachable before */ -int ch; /* character or NONCHAR code */ -register states aft; /* states already known reachable after */ -{ - register cset *cs; - register sop s; - register sopno pc; - register onestate here; /* note, macros know this name */ - register sopno look; - register long i; - - for (pc = start, INIT(here, pc); pc != stop; pc++, INC(here)) { - s = g->strip[pc]; - switch (OP(s)) { - case OEND: - assert(pc == stop-1); - break; - case OCHAR: - /* only characters can match */ - assert(!NONCHAR(ch) || ch != (char)OPND(s)); - if (ch == (char)OPND(s)) - FWD(aft, bef, 1); - break; - case OBOL: - if (ch == BOL || ch == BOLEOL) - FWD(aft, bef, 1); - break; - case OEOL: - if (ch == EOL || ch == BOLEOL) - FWD(aft, bef, 1); - break; - case OBOW: - if (ch == BOW) - FWD(aft, bef, 1); - break; - case OEOW: - if (ch == EOW) - FWD(aft, bef, 1); - break; - case OANY: - if (!NONCHAR(ch)) - FWD(aft, bef, 1); - break; - case OANYOF: - cs = &g->sets[OPND(s)]; - if (!NONCHAR(ch) && CHIN(cs, ch)) - FWD(aft, bef, 1); - break; - case OBACK_: /* ignored here */ - case O_BACK: - FWD(aft, aft, 1); - break; - case OPLUS_: /* forward, this is just an empty */ - FWD(aft, aft, 1); - break; - case O_PLUS: /* both forward and back */ - FWD(aft, aft, 1); - i = ISSETBACK(aft, OPND(s)); - BACK(aft, aft, OPND(s)); - if (!i && ISSETBACK(aft, OPND(s))) { - /* oho, must reconsider loop body */ - pc -= OPND(s) + 1; - INIT(here, pc); - } - break; - case OQUEST_: /* two branches, both forward */ - FWD(aft, aft, 1); - FWD(aft, aft, OPND(s)); - break; - case O_QUEST: /* just an empty */ - FWD(aft, aft, 1); - break; - case OLPAREN: /* not significant here */ - case ORPAREN: - FWD(aft, aft, 1); - break; - case OCH_: /* mark the first two branches */ - FWD(aft, aft, 1); - assert(OP(g->strip[pc+OPND(s)]) == OOR2); - FWD(aft, aft, OPND(s)); - break; - case OOR1: /* done a branch, find the O_CH */ - if (ISSTATEIN(aft, here)) { - for (look = 1; - OP(s = g->strip[pc+look]) != O_CH; - look += OPND(s)) - assert(OP(s) == OOR2); - FWD(aft, aft, look); - } - break; - case OOR2: /* propagate OCH_'s marking */ - FWD(aft, aft, 1); - if (OP(g->strip[pc+OPND(s)]) != O_CH) { - assert(OP(g->strip[pc+OPND(s)]) == OOR2); - FWD(aft, aft, OPND(s)); - } - break; - case O_CH: /* just empty */ - FWD(aft, aft, 1); - break; - default: /* ooooops... */ - assert(nope); - break; - } - } - - return(aft); -} - -#undef matcher -#undef fast -#undef slow -#undef dissect -#undef backref -#undef step -#undef print -#undef at -#undef match - -/* now undo things */ -#undef states -#undef CLEAR -#undef SET0 -#undef SET1 -#undef ISSET -#undef ASSIGN -#undef EQ -#undef STATEVARS -#undef STATESETUP -#undef STATETEARDOWN -#undef SETUP -#undef onestate -#undef INIT -#undef INC -#undef ISSTATEIN -#undef FWD -#undef BACK -#undef ISSETBACK -#undef SNAMES - -/* macros for manipulating states, large version */ -#define states char * -#define CLEAR(v) memset(v, 0, m->g->nstates) -#define SET0(v, n) ((v)[n] = 0) -#define SET1(v, n) ((v)[n] = 1) -#define ISSET(v, n) ((v)[n]) -#define ASSIGN(d, s) memcpy(d, s, m->g->nstates) -#define EQ(a, b) (memcmp(a, b, m->g->nstates) == 0) -#define STATEVARS int vn; char *space -#define STATESETUP(m, nv) { (m)->space = malloc((nv)*(m)->g->nstates); \ - if ((m)->space == NULL) return(REG_ESPACE); \ - (m)->vn = 0; } -#define STATETEARDOWN(m) { free((m)->space); } -#define SETUP(v) ((v) = &m->space[m->vn++ * m->g->nstates]) -#define onestate int -#define INIT(o, n) ((o) = (n)) -#define INC(o) ((o)++) -#define ISSTATEIN(v, o) ((v)[o]) -/* some abbreviations; note that some of these know variable names! */ -/* do "if I'm here, I can also be there" etc without branches */ -#define FWD(dst, src, n) ((dst)[here+(n)] |= (src)[here]) -#define BACK(dst, src, n) ((dst)[here-(n)] |= (src)[here]) -#define ISSETBACK(v, n) ((v)[here - (n)]) -/* function names */ -#define LNAMES /* flag */ - -/* - * The matching engine and friends. This file is #included by regexec.c - * after suitable #defines of a variety of macros used herein, so that - * different state representations can be used without duplicating masses - * of code. - */ - -#ifdef SNAMES -#define matcher smatcher -#define fast sfast -#define slow sslow -#define dissect sdissect -#define backref sbackref -#define step sstep -#define print sprint -#define at sat -#define match smat -#endif -#ifdef LNAMES -#define matcher lmatcher -#define fast lfast -#define slow lslow -#define dissect ldissect -#define backref lbackref -#define step lstep -#define print lprint -#define at lat -#define match lmat -#endif - -/* another structure passed up and down to avoid zillions of parameters */ -struct match { - struct re_guts *g; - int eflags; - regmatch_t *pmatch; /* [nsub+1] (0 element unused) */ - char *offp; /* offsets work from here */ - char *beginp; /* start of string -- virtual NUL precedes */ - char *endp; /* end of string -- virtual NUL here */ - char *coldp; /* can be no match starting before here */ - char **lastpos; /* [nplus+1] */ - STATEVARS; - states st; /* current states */ - states fresh; /* states for a fresh start */ - states tmp; /* temporary */ - states empty; /* empty set of states */ -}; - -static int matcher(register struct re_guts *g, char *string, size_t nmatch, regmatch_t pmatch[], int eflags); -static char *dissect(register struct match *m, char *start, char *stop, sopno startst, sopno stopst); -static char *backref(register struct match *m, char *start, char *stop, sopno startst, sopno stopst, sopno lev); -static char *fast(register struct match *m, char *start, char *stop, sopno startst, sopno stopst); -static char *slow(register struct match *m, char *start, char *stop, sopno startst, sopno stopst); -static states step(register struct re_guts *g, sopno start, sopno stop, register states bef, int ch, register states aft); -#define BOL (OUT+1) -#define EOL (BOL+1) -#define BOLEOL (BOL+2) -#define NOTHING (BOL+3) -#define BOW (BOL+4) -#define EOW (BOL+5) -#define CODEMAX (BOL+5) /* highest code used */ -#define NONCHAR(c) ((c) > CHAR_MAX) -#define NNONCHAR (CODEMAX-CHAR_MAX) -#define SP(t, s, c) /* nothing */ -#define AT(t, p1, p2, s1, s2) /* nothing */ -#define NOTE(s) /* nothing */ - -/* - - matcher - the actual matching engine - == static int matcher(register struct re_guts *g, char *string, \ - == size_t nmatch, regmatch_t pmatch[], int eflags); - */ -static int /* 0 success, REG_NOMATCH failure */ -matcher(g, string, nmatch, pmatch, eflags) -register struct re_guts *g; -char *string; -size_t nmatch; -regmatch_t pmatch[]; -int eflags; -{ - register char *endp; - register int i; - struct match mv; - register struct match *m = &mv; - register char *dp; - const register sopno gf = g->firststate+1; /* +1 for OEND */ - const register sopno gl = g->laststate; - char *start; - char *stop; - - /* simplify the situation where possible */ - if (g->cflags®_NOSUB) - nmatch = 0; - if (eflags®_STARTEND) { - start = string + pmatch[0].rm_so; - stop = string + pmatch[0].rm_eo; - } else { - start = string; - stop = start + strlen(start); - } - if (stop < start) - return(REG_INVARG); - - /* prescreening; this does wonders for this rather slow code */ - if (g->must != NULL) { - for (dp = start; dp < stop; dp++) - if (*dp == g->must[0] && stop - dp >= g->mlen && - memcmp(dp, g->must, (size_t)g->mlen) == 0) - break; - if (dp == stop) /* we didn't find g->must */ - return(REG_NOMATCH); - } - - /* match struct setup */ - m->g = g; - m->eflags = eflags; - m->pmatch = NULL; - m->lastpos = NULL; - m->offp = string; - m->beginp = start; - m->endp = stop; - STATESETUP(m, 4); - SETUP(m->st); - SETUP(m->fresh); - SETUP(m->tmp); - SETUP(m->empty); - CLEAR(m->empty); - - /* this loop does only one repetition except for backrefs */ - for (;;) { - endp = fast(m, start, stop, gf, gl); - if (endp == NULL) { /* a miss */ - STATETEARDOWN(m); - return(REG_NOMATCH); - } - if (nmatch == 0 && !g->backrefs) - break; /* no further info needed */ - - /* where? */ - assert(m->coldp != NULL); - for (;;) { - NOTE("finding start"); - endp = slow(m, m->coldp, stop, gf, gl); - if (endp != NULL) - break; - assert(m->coldp < m->endp); - m->coldp++; - } - if (nmatch == 1 && !g->backrefs) - break; /* no further info needed */ - - /* oh my, he wants the subexpressions... */ - if (m->pmatch == NULL) - m->pmatch = (regmatch_t *)malloc((m->g->nsub + 1) * - sizeof(regmatch_t)); - if (m->pmatch == NULL) { - STATETEARDOWN(m); - return(REG_ESPACE); - } - for (i = 1; i <= m->g->nsub; i++) - m->pmatch[i].rm_so = m->pmatch[i].rm_eo = -1; - if (!g->backrefs && !(m->eflags®_BACKR)) { - NOTE("dissecting"); - dp = dissect(m, m->coldp, endp, gf, gl); - } else { - if (g->nplus > 0 && m->lastpos == NULL) - m->lastpos = (char **)malloc((g->nplus+1) * - sizeof(char *)); - if (g->nplus > 0 && m->lastpos == NULL) { - free(m->pmatch); - STATETEARDOWN(m); - return(REG_ESPACE); - } - NOTE("backref dissect"); - dp = backref(m, m->coldp, endp, gf, gl, (sopno)0); - } - if (dp != NULL) - break; - - /* uh-oh... we couldn't find a subexpression-level match */ - assert(g->backrefs); /* must be back references doing it */ - assert(g->nplus == 0 || m->lastpos != NULL); - for (;;) { - if (dp != NULL || endp <= m->coldp) - break; /* defeat */ - NOTE("backoff"); - endp = slow(m, m->coldp, endp-1, gf, gl); - if (endp == NULL) - break; /* defeat */ - /* try it on a shorter possibility */ - NOTE("backoff dissect"); - dp = backref(m, m->coldp, endp, gf, gl, (sopno)0); - } - assert(dp == NULL || dp == endp); - if (dp != NULL) /* found a shorter one */ - break; - - /* despite initial appearances, there is no match here */ - NOTE("false alarm"); - start = m->coldp + 1; /* recycle starting later */ - assert(start <= stop); - } - - /* fill in the details if requested */ - if (nmatch > 0) { - pmatch[0].rm_so = m->coldp - m->offp; - pmatch[0].rm_eo = endp - m->offp; - } - if (nmatch > 1) { - assert(m->pmatch != NULL); - for (i = 1; i < nmatch; i++) - if (i <= m->g->nsub) - pmatch[i] = m->pmatch[i]; - else { - pmatch[i].rm_so = -1; - pmatch[i].rm_eo = -1; - } - } - - if (m->pmatch != NULL) - free((char *)m->pmatch); - if (m->lastpos != NULL) - free((char *)m->lastpos); - STATETEARDOWN(m); - return(0); -} - -/* - - dissect - figure out what matched what, no back references - == static char *dissect(register struct match *m, char *start, \ - == char *stop, sopno startst, sopno stopst); - */ -static char * /* == stop (success) always */ -dissect(m, start, stop, startst, stopst) -register struct match *m; -char *start; -char *stop; -sopno startst; -sopno stopst; -{ - register int i; - register sopno ss; /* start sop of current subRE */ - register sopno es; /* end sop of current subRE */ - register char *sp; /* start of string matched by it */ - register char *stp; /* string matched by it cannot pass here */ - register char *rest; /* start of rest of string */ - register char *tail; /* string unmatched by rest of RE */ - register sopno ssub; /* start sop of subsubRE */ - register sopno esub; /* end sop of subsubRE */ - register char *ssp; /* start of string matched by subsubRE */ - register char *sep; /* end of string matched by subsubRE */ - register char *oldssp; /* previous ssp */ - register char *dp; - - AT("diss", start, stop, startst, stopst); - sp = start; - for (ss = startst; ss < stopst; ss = es) { - /* identify end of subRE */ - es = ss; - switch (OP(m->g->strip[es])) { - case OPLUS_: - case OQUEST_: - es += OPND(m->g->strip[es]); - break; - case OCH_: - while (OP(m->g->strip[es]) != O_CH) - es += OPND(m->g->strip[es]); - break; - } - es++; - - /* figure out what it matched */ - switch (OP(m->g->strip[ss])) { - case OEND: - assert(nope); - break; - case OCHAR: - sp++; - break; - case OBOL: - case OEOL: - case OBOW: - case OEOW: - break; - case OANY: - case OANYOF: - sp++; - break; - case OBACK_: - case O_BACK: - assert(nope); - break; - /* cases where length of match is hard to find */ - case OQUEST_: - stp = stop; - for (;;) { - /* how long could this one be? */ - rest = slow(m, sp, stp, ss, es); - assert(rest != NULL); /* it did match */ - /* could the rest match the rest? */ - tail = slow(m, rest, stop, es, stopst); - if (tail == stop) - break; /* yes! */ - /* no -- try a shorter match for this one */ - stp = rest - 1; - assert(stp >= sp); /* it did work */ - } - ssub = ss + 1; - esub = es - 1; - /* did innards match? */ - if (slow(m, sp, rest, ssub, esub) != NULL) { - dp = dissect(m, sp, rest, ssub, esub); - assert(dp == rest); - } else /* no */ - assert(sp == rest); - sp = rest; - break; - case OPLUS_: - stp = stop; - for (;;) { - /* how long could this one be? */ - rest = slow(m, sp, stp, ss, es); - assert(rest != NULL); /* it did match */ - /* could the rest match the rest? */ - tail = slow(m, rest, stop, es, stopst); - if (tail == stop) - break; /* yes! */ - /* no -- try a shorter match for this one */ - stp = rest - 1; - assert(stp >= sp); /* it did work */ - } - ssub = ss + 1; - esub = es - 1; - ssp = sp; - oldssp = ssp; - for (;;) { /* find last match of innards */ - sep = slow(m, ssp, rest, ssub, esub); - if (sep == NULL || sep == ssp) - break; /* failed or matched null */ - oldssp = ssp; /* on to next try */ - ssp = sep; - } - if (sep == NULL) { - /* last successful match */ - sep = ssp; - ssp = oldssp; - } - assert(sep == rest); /* must exhaust substring */ - assert(slow(m, ssp, sep, ssub, esub) == rest); - dp = dissect(m, ssp, sep, ssub, esub); - assert(dp == sep); - sp = rest; - break; - case OCH_: - stp = stop; - for (;;) { - /* how long could this one be? */ - rest = slow(m, sp, stp, ss, es); - assert(rest != NULL); /* it did match */ - /* could the rest match the rest? */ - tail = slow(m, rest, stop, es, stopst); - if (tail == stop) - break; /* yes! */ - /* no -- try a shorter match for this one */ - stp = rest - 1; - assert(stp >= sp); /* it did work */ - } - ssub = ss + 1; - esub = ss + OPND(m->g->strip[ss]) - 1; - assert(OP(m->g->strip[esub]) == OOR1); - for (;;) { /* find first matching branch */ - if (slow(m, sp, rest, ssub, esub) == rest) - break; /* it matched all of it */ - /* that one missed, try next one */ - assert(OP(m->g->strip[esub]) == OOR1); - esub++; - assert(OP(m->g->strip[esub]) == OOR2); - ssub = esub + 1; - esub += OPND(m->g->strip[esub]); - if (OP(m->g->strip[esub]) == OOR2) - esub--; - else - assert(OP(m->g->strip[esub]) == O_CH); - } - dp = dissect(m, sp, rest, ssub, esub); - assert(dp == rest); - sp = rest; - break; - case O_PLUS: - case O_QUEST: - case OOR1: - case OOR2: - case O_CH: - assert(nope); - break; - case OLPAREN: - i = OPND(m->g->strip[ss]); - assert(0 < i && i <= m->g->nsub); - m->pmatch[i].rm_so = sp - m->offp; - break; - case ORPAREN: - i = OPND(m->g->strip[ss]); - assert(0 < i && i <= m->g->nsub); - m->pmatch[i].rm_eo = sp - m->offp; - break; - default: /* uh oh */ - assert(nope); - break; - } - } - - assert(sp == stop); - return(sp); -} - -/* - - backref - figure out what matched what, figuring in back references - == static char *backref(register struct match *m, char *start, \ - == char *stop, sopno startst, sopno stopst, sopno lev); - */ -static char * /* == stop (success) or NULL (failure) */ -backref(m, start, stop, startst, stopst, lev) -register struct match *m; -char *start; -char *stop; -sopno startst; -sopno stopst; -sopno lev; /* PLUS nesting level */ -{ - register int i; - register sopno ss; /* start sop of current subRE */ - register char *sp; /* start of string matched by it */ - register sopno ssub; /* start sop of subsubRE */ - register sopno esub; /* end sop of subsubRE */ - register char *ssp; /* start of string matched by subsubRE */ - register char *dp; - register size_t len; - register int hard; - register sop s; - register regoff_t offsave; - register cset *cs; - - AT("back", start, stop, startst, stopst); - sp = start; - - /* get as far as we can with easy stuff */ - hard = 0; - for (ss = startst; !hard && ss < stopst; ss++) - switch (OP(s = m->g->strip[ss])) { - case OCHAR: - if (sp == stop || *sp++ != (char)OPND(s)) - return(NULL); - break; - case OANY: - if (sp == stop) - return(NULL); - sp++; - break; - case OANYOF: - cs = &m->g->sets[OPND(s)]; - if (sp == stop || !CHIN(cs, *sp++)) - return(NULL); - break; - case OBOL: - if ( (sp == m->beginp && !(m->eflags®_NOTBOL)) || - (sp < m->endp && *(sp-1) == '\n' && - (m->g->cflags®_NEWLINE)) ) - { /* yes */ } - else - return(NULL); - break; - case OEOL: - if ( (sp == m->endp && !(m->eflags®_NOTEOL)) || - (sp < m->endp && *sp == '\n' && - (m->g->cflags®_NEWLINE)) ) - { /* yes */ } - else - return(NULL); - break; - case OBOW: - if (( (sp == m->beginp && !(m->eflags®_NOTBOL)) || - (sp < m->endp && *(sp-1) == '\n' && - (m->g->cflags®_NEWLINE)) || - (sp > m->beginp && - !ISWORD(*(sp-1))) ) && - (sp < m->endp && ISWORD(*sp)) ) - { /* yes */ } - else - return(NULL); - break; - case OEOW: - if (( (sp == m->endp && !(m->eflags®_NOTEOL)) || - (sp < m->endp && *sp == '\n' && - (m->g->cflags®_NEWLINE)) || - (sp < m->endp && !ISWORD(*sp)) ) && - (sp > m->beginp && ISWORD(*(sp-1))) ) - { /* yes */ } - else - return(NULL); - break; - case O_QUEST: - break; - case OOR1: /* matches null but needs to skip */ - ss++; - s = m->g->strip[ss]; - do { - assert(OP(s) == OOR2); - ss += OPND(s); - } while (OP(s = m->g->strip[ss]) != O_CH); - /* note that the ss++ gets us past the O_CH */ - break; - default: /* have to make a choice */ - hard = 1; - break; - } - if (!hard) { /* that was it! */ - if (sp != stop) - return(NULL); - return(sp); - } - ss--; /* adjust for the for's final increment */ - - /* the hard stuff */ - AT("hard", sp, stop, ss, stopst); - s = m->g->strip[ss]; - switch (OP(s)) { - case OBACK_: /* the vilest depths */ - i = OPND(s); - assert(0 < i && i <= m->g->nsub); - if (m->pmatch[i].rm_eo == -1) - return(NULL); - assert(m->pmatch[i].rm_so != -1); - len = m->pmatch[i].rm_eo - m->pmatch[i].rm_so; - assert(stop - m->beginp >= len); - if (sp > stop - len) - return(NULL); /* not enough left to match */ - ssp = m->offp + m->pmatch[i].rm_so; - if (memcmp(sp, ssp, len) != 0) - return(NULL); - while (m->g->strip[ss] != SOP(O_BACK, i)) - ss++; - return(backref(m, sp+len, stop, ss+1, stopst, lev)); - break; - case OQUEST_: /* to null or not */ - dp = backref(m, sp, stop, ss+1, stopst, lev); - if (dp != NULL) - return(dp); /* not */ - return(backref(m, sp, stop, ss+OPND(s)+1, stopst, lev)); - break; - case OPLUS_: - assert(m->lastpos != NULL); - assert(lev+1 <= m->g->nplus); - m->lastpos[lev+1] = sp; - return(backref(m, sp, stop, ss+1, stopst, lev+1)); - break; - case O_PLUS: - if (sp == m->lastpos[lev]) /* last pass matched null */ - return(backref(m, sp, stop, ss+1, stopst, lev-1)); - /* try another pass */ - m->lastpos[lev] = sp; - dp = backref(m, sp, stop, ss-OPND(s)+1, stopst, lev); - if (dp == NULL) - return(backref(m, sp, stop, ss+1, stopst, lev-1)); - else - return(dp); - break; - case OCH_: /* find the right one, if any */ - ssub = ss + 1; - esub = ss + OPND(s) - 1; - assert(OP(m->g->strip[esub]) == OOR1); - for (;;) { /* find first matching branch */ - dp = backref(m, sp, stop, ssub, esub, lev); - if (dp != NULL) - return(dp); - /* that one missed, try next one */ - if (OP(m->g->strip[esub]) == O_CH) - return(NULL); /* there is none */ - esub++; - assert(OP(m->g->strip[esub]) == OOR2); - ssub = esub + 1; - esub += OPND(m->g->strip[esub]); - if (OP(m->g->strip[esub]) == OOR2) - esub--; - else - assert(OP(m->g->strip[esub]) == O_CH); - } - break; - case OLPAREN: /* must undo assignment if rest fails */ - i = OPND(s); - assert(0 < i && i <= m->g->nsub); - offsave = m->pmatch[i].rm_so; - m->pmatch[i].rm_so = sp - m->offp; - dp = backref(m, sp, stop, ss+1, stopst, lev); - if (dp != NULL) - return(dp); - m->pmatch[i].rm_so = offsave; - return(NULL); - break; - case ORPAREN: /* must undo assignment if rest fails */ - i = OPND(s); - assert(0 < i && i <= m->g->nsub); - offsave = m->pmatch[i].rm_eo; - m->pmatch[i].rm_eo = sp - m->offp; - dp = backref(m, sp, stop, ss+1, stopst, lev); - if (dp != NULL) - return(dp); - m->pmatch[i].rm_eo = offsave; - return(NULL); - break; - default: /* uh oh */ - assert(nope); - break; - } - - /* "can't happen" */ - assert(nope); - /* NOTREACHED */ - return((char *)NULL); /* dummy */ -} - -/* - - fast - step through the string at top speed - == static char *fast(register struct match *m, char *start, \ - == char *stop, sopno startst, sopno stopst); - */ -static char * /* where tentative match ended, or NULL */ -fast(m, start, stop, startst, stopst) -register struct match *m; -char *start; -char *stop; -sopno startst; -sopno stopst; -{ - register states st = m->st; - register states fresh = m->fresh; - register states tmp = m->tmp; - register char *p = start; - register int c = (start == m->beginp) ? OUT : *(start-1); - register int lastc; /* previous c */ - register int flagch; - register int i; - register char *coldp; /* last p after which no match was underway */ - - CLEAR(st); - SET1(st, startst); - st = step(m->g, startst, stopst, st, NOTHING, st); - ASSIGN(fresh, st); - SP("start", st, *p); - coldp = NULL; - for (;;) { - /* next character */ - lastc = c; - c = (p == m->endp) ? OUT : *p; - if (EQ(st, fresh)) - coldp = p; - - /* is there an EOL and/or BOL between lastc and c? */ - flagch = '\0'; - i = 0; - if ( (lastc == '\n' && m->g->cflags®_NEWLINE) || - (lastc == OUT && !(m->eflags®_NOTBOL)) ) { - flagch = BOL; - i = m->g->nbol; - } - if ( (c == '\n' && m->g->cflags®_NEWLINE) || - (c == OUT && !(m->eflags®_NOTEOL)) ) { - flagch = (flagch == BOL) ? BOLEOL : EOL; - i += m->g->neol; - } - if (i != 0) { - for (; i > 0; i--) - st = step(m->g, startst, stopst, st, flagch, st); - SP("boleol", st, c); - } - - /* how about a word boundary? */ - if ( (flagch == BOL || (lastc != OUT && !ISWORD(lastc))) && - (c != OUT && ISWORD(c)) ) { - flagch = BOW; - } - if ( (lastc != OUT && ISWORD(lastc)) && - (flagch == EOL || (c != OUT && !ISWORD(c))) ) { - flagch = EOW; - } - if (flagch == BOW || flagch == EOW) { - st = step(m->g, startst, stopst, st, flagch, st); - SP("boweow", st, c); - } - - /* are we done? */ - if (ISSET(st, stopst) || p == stop) - break; /* NOTE BREAK OUT */ - - /* no, we must deal with this character */ - ASSIGN(tmp, st); - ASSIGN(st, fresh); - assert(c != OUT); - st = step(m->g, startst, stopst, tmp, c, st); - SP("aft", st, c); - assert(EQ(step(m->g, startst, stopst, st, NOTHING, st), st)); - p++; - } - - assert(coldp != NULL); - m->coldp = coldp; - if (ISSET(st, stopst)) - return(p+1); - else - return(NULL); -} - -/* - - slow - step through the string more deliberately - == static char *slow(register struct match *m, char *start, \ - == char *stop, sopno startst, sopno stopst); - */ -static char * /* where it ended */ -slow(m, start, stop, startst, stopst) -register struct match *m; -char *start; -char *stop; -sopno startst; -sopno stopst; -{ - register states st = m->st; - register states empty = m->empty; - register states tmp = m->tmp; - register char *p = start; - register int c = (start == m->beginp) ? OUT : *(start-1); - register int lastc; /* previous c */ - register int flagch; - register int i; - register char *matchp; /* last p at which a match ended */ - - AT("slow", start, stop, startst, stopst); - CLEAR(st); - SET1(st, startst); - SP("sstart", st, *p); - st = step(m->g, startst, stopst, st, NOTHING, st); - matchp = NULL; - for (;;) { - /* next character */ - lastc = c; - c = (p == m->endp) ? OUT : *p; - - /* is there an EOL and/or BOL between lastc and c? */ - flagch = '\0'; - i = 0; - if ( (lastc == '\n' && m->g->cflags®_NEWLINE) || - (lastc == OUT && !(m->eflags®_NOTBOL)) ) { - flagch = BOL; - i = m->g->nbol; - } - if ( (c == '\n' && m->g->cflags®_NEWLINE) || - (c == OUT && !(m->eflags®_NOTEOL)) ) { - flagch = (flagch == BOL) ? BOLEOL : EOL; - i += m->g->neol; - } - if (i != 0) { - for (; i > 0; i--) - st = step(m->g, startst, stopst, st, flagch, st); - SP("sboleol", st, c); - } - - /* how about a word boundary? */ - if ( (flagch == BOL || (lastc != OUT && !ISWORD(lastc))) && - (c != OUT && ISWORD(c)) ) { - flagch = BOW; - } - if ( (lastc != OUT && ISWORD(lastc)) && - (flagch == EOL || (c != OUT && !ISWORD(c))) ) { - flagch = EOW; - } - if (flagch == BOW || flagch == EOW) { - st = step(m->g, startst, stopst, st, flagch, st); - SP("sboweow", st, c); - } - - /* are we done? */ - if (ISSET(st, stopst)) - matchp = p; - if (EQ(st, empty) || p == stop) - break; /* NOTE BREAK OUT */ - - /* no, we must deal with this character */ - ASSIGN(tmp, st); - ASSIGN(st, empty); - assert(c != OUT); - st = step(m->g, startst, stopst, tmp, c, st); - SP("saft", st, c); - assert(EQ(step(m->g, startst, stopst, st, NOTHING, st), st)); - p++; - } - - return(matchp); -} - - -static states -step(g, start, stop, bef, ch, aft) -register struct re_guts *g; -sopno start; /* start state within strip */ -sopno stop; /* state after stop state within strip */ -register states bef; /* states reachable before */ -int ch; /* character or NONCHAR code */ -register states aft; /* states already known reachable after */ -{ - register cset *cs; - register sop s; - register sopno pc; - register onestate here; /* note, macros know this name */ - register sopno look; - register long i; - - for (pc = start, INIT(here, pc); pc != stop; pc++, INC(here)) { - s = g->strip[pc]; - switch (OP(s)) { - case OEND: - assert(pc == stop-1); - break; - case OCHAR: - /* only characters can match */ - assert(!NONCHAR(ch) || ch != (char)OPND(s)); - if (ch == (char)OPND(s)) - FWD(aft, bef, 1); - break; - case OBOL: - if (ch == BOL || ch == BOLEOL) - FWD(aft, bef, 1); - break; - case OEOL: - if (ch == EOL || ch == BOLEOL) - FWD(aft, bef, 1); - break; - case OBOW: - if (ch == BOW) - FWD(aft, bef, 1); - break; - case OEOW: - if (ch == EOW) - FWD(aft, bef, 1); - break; - case OANY: - if (!NONCHAR(ch)) - FWD(aft, bef, 1); - break; - case OANYOF: - cs = &g->sets[OPND(s)]; - if (!NONCHAR(ch) && CHIN(cs, ch)) - FWD(aft, bef, 1); - break; - case OBACK_: /* ignored here */ - case O_BACK: - FWD(aft, aft, 1); - break; - case OPLUS_: /* forward, this is just an empty */ - FWD(aft, aft, 1); - break; - case O_PLUS: /* both forward and back */ - FWD(aft, aft, 1); - i = ISSETBACK(aft, OPND(s)); - BACK(aft, aft, OPND(s)); - if (!i && ISSETBACK(aft, OPND(s))) { - /* oho, must reconsider loop body */ - pc -= OPND(s) + 1; - INIT(here, pc); - } - break; - case OQUEST_: /* two branches, both forward */ - FWD(aft, aft, 1); - FWD(aft, aft, OPND(s)); - break; - case O_QUEST: /* just an empty */ - FWD(aft, aft, 1); - break; - case OLPAREN: /* not significant here */ - case ORPAREN: - FWD(aft, aft, 1); - break; - case OCH_: /* mark the first two branches */ - FWD(aft, aft, 1); - assert(OP(g->strip[pc+OPND(s)]) == OOR2); - FWD(aft, aft, OPND(s)); - break; - case OOR1: /* done a branch, find the O_CH */ - if (ISSTATEIN(aft, here)) { - for (look = 1; - OP(s = g->strip[pc+look]) != O_CH; - look += OPND(s)) - assert(OP(s) == OOR2); - FWD(aft, aft, look); - } - break; - case OOR2: /* propagate OCH_'s marking */ - FWD(aft, aft, 1); - if (OP(g->strip[pc+OPND(s)]) != O_CH) { - assert(OP(g->strip[pc+OPND(s)]) == OOR2); - FWD(aft, aft, OPND(s)); - } - break; - case O_CH: /* just empty */ - FWD(aft, aft, 1); - break; - default: /* ooooops... */ - assert(nope); - break; - } - } - - return(aft); -} - -#undef matcher -#undef fast -#undef slow -#undef dissect -#undef backref -#undef step -#undef print -#undef at -#undef match - -int /* 0 success, REG_NOMATCH failure */ -regexec(preg, string, nmatch, pmatch, eflags) -const regex_t *preg; -const char *string; -size_t nmatch; -regmatch_t pmatch[]; -int eflags; -{ - register struct re_guts *g = preg->re_g; -#define GOODFLAGS(f) ((f)&(REG_NOTBOL|REG_NOTEOL|REG_STARTEND)) - - if (preg->re_magic != MAGIC1 || g->magic != MAGIC2) - return(REG_BADPAT); - assert(!(g->iflags&BAD)); - if (g->iflags&BAD) /* backstop for no-debug case */ - return(REG_BADPAT); - eflags = GOODFLAGS(eflags); - - if (g->nstates <= CHAR_BIT*sizeof(states1) && !(eflags®_LARGE)) - return(smatcher(g, (char *)string, nmatch, pmatch, eflags)); - else - return(lmatcher(g, (char *)string, nmatch, pmatch, eflags)); -#undef GOODFLAGS -} -/* - - regfree - free everything - = extern void regfree(regex_t *); - */ -void -regfree(preg) -regex_t *preg; -{ - register struct re_guts *g; - - if (preg->re_magic != MAGIC1) /* oops */ - return; /* nice to complain, but hard */ - - g = preg->re_g; - if (g == NULL || g->magic != MAGIC2) /* oops again */ - return; - preg->re_magic = 0; /* mark it invalid */ - g->magic = 0; /* mark it invalid */ - - if (g->strip != NULL) - free((char *)g->strip); - if (g->sets != NULL) - free((char *)g->sets); - if (g->setbits != NULL) - free((char *)g->setbits); - if (g->must != NULL) - free(g->must); - free((char *)g); -} - -#ifdef WITH_MAIN -int main(int argc, char* argv[]){ - regex_t preg; - int i, s; - if(argc<2) return 1; - if (regcomp(&preg, argv[1], REG_NOSUB)) { - fprintf(stderr, "Failed to compile regex\n"); - return 2; - } - for (i = 2; i -#include - - -#define uchar unsigned char - -static const uchar perm1[56] = {57, 49, 41, 33, 25, 17, 9, - 1, 58, 50, 42, 34, 26, 18, - 10, 2, 59, 51, 43, 35, 27, - 19, 11, 3, 60, 52, 44, 36, - 63, 55, 47, 39, 31, 23, 15, - 7, 62, 54, 46, 38, 30, 22, - 14, 6, 61, 53, 45, 37, 29, - 21, 13, 5, 28, 20, 12, 4}; - -static const uchar perm2[48] = {14, 17, 11, 24, 1, 5, - 3, 28, 15, 6, 21, 10, - 23, 19, 12, 4, 26, 8, - 16, 7, 27, 20, 13, 2, - 41, 52, 31, 37, 47, 55, - 30, 40, 51, 45, 33, 48, - 44, 49, 39, 56, 34, 53, - 46, 42, 50, 36, 29, 32}; - -static const uchar perm3[64] = {58, 50, 42, 34, 26, 18, 10, 2, - 60, 52, 44, 36, 28, 20, 12, 4, - 62, 54, 46, 38, 30, 22, 14, 6, - 64, 56, 48, 40, 32, 24, 16, 8, - 57, 49, 41, 33, 25, 17, 9, 1, - 59, 51, 43, 35, 27, 19, 11, 3, - 61, 53, 45, 37, 29, 21, 13, 5, - 63, 55, 47, 39, 31, 23, 15, 7}; - -static const uchar perm4[48] = { 32, 1, 2, 3, 4, 5, - 4, 5, 6, 7, 8, 9, - 8, 9, 10, 11, 12, 13, - 12, 13, 14, 15, 16, 17, - 16, 17, 18, 19, 20, 21, - 20, 21, 22, 23, 24, 25, - 24, 25, 26, 27, 28, 29, - 28, 29, 30, 31, 32, 1}; - -static const uchar perm5[32] = { 16, 7, 20, 21, - 29, 12, 28, 17, - 1, 15, 23, 26, - 5, 18, 31, 10, - 2, 8, 24, 14, - 32, 27, 3, 9, - 19, 13, 30, 6, - 22, 11, 4, 25}; - - -static const uchar perm6[64] ={ 40, 8, 48, 16, 56, 24, 64, 32, - 39, 7, 47, 15, 55, 23, 63, 31, - 38, 6, 46, 14, 54, 22, 62, 30, - 37, 5, 45, 13, 53, 21, 61, 29, - 36, 4, 44, 12, 52, 20, 60, 28, - 35, 3, 43, 11, 51, 19, 59, 27, - 34, 2, 42, 10, 50, 18, 58, 26, - 33, 1, 41, 9, 49, 17, 57, 25}; - - -static const uchar sc[16] = {1, 1, 2, 2, 2, 2, 2, 2, 1, 2, 2, 2, 2, 2, 2, 1}; - -static const uchar sbox[8][4][16] = { - {{14, 4, 13, 1, 2, 15, 11, 8, 3, 10, 6, 12, 5, 9, 0, 7}, - {0, 15, 7, 4, 14, 2, 13, 1, 10, 6, 12, 11, 9, 5, 3, 8}, - {4, 1, 14, 8, 13, 6, 2, 11, 15, 12, 9, 7, 3, 10, 5, 0}, - {15, 12, 8, 2, 4, 9, 1, 7, 5, 11, 3, 14, 10, 0, 6, 13}}, - - {{15, 1, 8, 14, 6, 11, 3, 4, 9, 7, 2, 13, 12, 0, 5, 10}, - {3, 13, 4, 7, 15, 2, 8, 14, 12, 0, 1, 10, 6, 9, 11, 5}, - {0, 14, 7, 11, 10, 4, 13, 1, 5, 8, 12, 6, 9, 3, 2, 15}, - {13, 8, 10, 1, 3, 15, 4, 2, 11, 6, 7, 12, 0, 5, 14, 9}}, - - {{10, 0, 9, 14, 6, 3, 15, 5, 1, 13, 12, 7, 11, 4, 2, 8}, - {13, 7, 0, 9, 3, 4, 6, 10, 2, 8, 5, 14, 12, 11, 15, 1}, - {13, 6, 4, 9, 8, 15, 3, 0, 11, 1, 2, 12, 5, 10, 14, 7}, - {1, 10, 13, 0, 6, 9, 8, 7, 4, 15, 14, 3, 11, 5, 2, 12}}, - - {{7, 13, 14, 3, 0, 6, 9, 10, 1, 2, 8, 5, 11, 12, 4, 15}, - {13, 8, 11, 5, 6, 15, 0, 3, 4, 7, 2, 12, 1, 10, 14, 9}, - {10, 6, 9, 0, 12, 11, 7, 13, 15, 1, 3, 14, 5, 2, 8, 4}, - {3, 15, 0, 6, 10, 1, 13, 8, 9, 4, 5, 11, 12, 7, 2, 14}}, - - {{2, 12, 4, 1, 7, 10, 11, 6, 8, 5, 3, 15, 13, 0, 14, 9}, - {14, 11, 2, 12, 4, 7, 13, 1, 5, 0, 15, 10, 3, 9, 8, 6}, - {4, 2, 1, 11, 10, 13, 7, 8, 15, 9, 12, 5, 6, 3, 0, 14}, - {11, 8, 12, 7, 1, 14, 2, 13, 6, 15, 0, 9, 10, 4, 5, 3}}, - - {{12, 1, 10, 15, 9, 2, 6, 8, 0, 13, 3, 4, 14, 7, 5, 11}, - {10, 15, 4, 2, 7, 12, 9, 5, 6, 1, 13, 14, 0, 11, 3, 8}, - {9, 14, 15, 5, 2, 8, 12, 3, 7, 0, 4, 10, 1, 13, 11, 6}, - {4, 3, 2, 12, 9, 5, 15, 10, 11, 14, 1, 7, 6, 0, 8, 13}}, - - {{4, 11, 2, 14, 15, 0, 8, 13, 3, 12, 9, 7, 5, 10, 6, 1}, - {13, 0, 11, 7, 4, 9, 1, 10, 14, 3, 5, 12, 2, 15, 8, 6}, - {1, 4, 11, 13, 12, 3, 7, 14, 10, 15, 6, 8, 0, 5, 9, 2}, - {6, 11, 13, 8, 1, 4, 10, 7, 9, 5, 0, 15, 14, 2, 3, 12}}, - - {{13, 2, 8, 4, 6, 15, 11, 1, 10, 9, 3, 14, 5, 0, 12, 7}, - {1, 15, 13, 8, 10, 3, 7, 4, 12, 5, 6, 11, 0, 14, 9, 2}, - {7, 11, 4, 1, 9, 12, 14, 2, 0, 6, 10, 13, 15, 3, 5, 8}, - {2, 1, 14, 7, 4, 10, 8, 13, 15, 12, 9, 0, 3, 5, 6, 11}}}; - -static void permute(char *out, const char *in, const uchar *p, int n) -{ - int i; - for (i=0;i>1; - key[1] = ((str[0]&0x01)<<6) | (str[1]>>2); - key[2] = ((str[1]&0x03)<<5) | (str[2]>>3); - key[3] = ((str[2]&0x07)<<4) | (str[3]>>4); - key[4] = ((str[3]&0x0F)<<3) | (str[4]>>5); - key[5] = ((str[4]&0x1F)<<2) | (str[5]>>6); - key[6] = ((str[5]&0x3F)<<1) | (str[6]>>7); - key[7] = str[6]&0x7F; - for (i=0;i<8;i++) { - key[i] = (key[i]<<1); - } -} - - -static void smbhash(unsigned char *out, const unsigned char *in, unsigned char *key) -{ - int i; - char outb[64]; - char inb[64]; - char keyb[64]; - unsigned char key2[8]; - - str_to_key(key, key2); - - for (i=0;i<64;i++) { - inb[i] = (in[i/8] & (1<<(7-(i%8)))) ? 1 : 0; - keyb[i] = (key2[i/8] & (1<<(7-(i%8)))) ? 1 : 0; - outb[i] = 0; - } - - dohash(outb, inb, keyb); - - for (i=0;i<8;i++) { - out[i] = 0; - } - - for (i=0;i<64;i++) { - if (outb[i]) - out[i/8] |= (1<<(7-(i%8))); - } -} - -/* - * Converts the password to uppercase, and creates the LM - * password hash. - */ -void lmpwdhash(const unsigned char *password,unsigned char *lmhash) -{ - int i; - unsigned char p14[14]; - static unsigned char sp8[8] = {0x4b, 0x47, 0x53, 0x21, 0x40, 0x23, 0x24, 0x25}; - - memset(p14, 0, sizeof(p14)); - for (i = 0; i < 14 && password[i]; i++) { - p14[i] = toupper((int) password[i]); - } - - smbhash(lmhash, sp8, p14); - smbhash(lmhash+8, sp8, p14+7); -} - -/* - * Take the NT or LM password, and return the MSCHAP response - * - * The win_password MUST be exactly 16 bytes long. - */ -void mschap(const unsigned char *win_password, - const unsigned char *challenge, unsigned char *response) -{ - unsigned char p21[21]; - - memset(p21, 0, sizeof(p21)); - memcpy(p21, win_password, 16); - - smbhash(response, challenge, p21); - smbhash(response+8, challenge, p21+7); - smbhash(response+16, challenge, p21+14); -} diff --git a/src/ntlm.c b/src/ntlm.c deleted file mode 100644 index 8f966d3..0000000 --- a/src/ntlm.c +++ /dev/null @@ -1,88 +0,0 @@ -/* - 3APA3A simpliest proxy server - (c) 2002-2021 by Vladimir Dubrovin <3proxy@3proxy.org> - - please read License Agreement - -*/ - -#include "proxy.h" -struct ntlmchal { - unsigned char sig[8]; - unsigned char messtype[4]; - unsigned char dom_len[2]; - unsigned char dom_max_len[2]; - unsigned char dom_offset[4]; - unsigned char flags[4]; - unsigned char challenge[8]; - unsigned char reserved[8]; - unsigned char addr_len[2]; - unsigned char addr_max_len[2]; - unsigned char addr_offset[4]; - unsigned char data[1]; -}; - -struct ntlmreq { - unsigned char sig[8]; - unsigned char messtype[4]; - unsigned char flags[4]; - unsigned char dom_len[2]; - unsigned char dom_max_len[2]; - unsigned char dom_offset[4]; - unsigned char pad1[2]; - unsigned char host_len[2]; - unsigned char host_max_len[2]; - unsigned char host_offset[4]; - unsigned char pad2[2]; - unsigned char data[1]; -}; - -int text2unicode(const char * text, char * buf, int buflen){ - int count = 0; - buflen = ((buflen>>1)<<1); - if(!text || !buflen) return 0; - do { - buf[count++] = toupper(*text++); - buf[count++] = '\0'; - } while (*text && count < buflen); - return count; -} - -void unicode2text(const char *unicode, char * buf, int len){ - int i; - if(!unicode || !len) return; - for(i=0; isig, "NTLMSSP", 8); - chal->messtype[0] = 2; - gethostname(hostname, 128); - hostname[15] = 0; - len = (((int)strlen(hostname)) << 1); - chal->dom_len[0] = len; - chal->dom_max_len[0] = len; - chal->dom_offset[0] = (unsigned char)((unsigned char *)chal->data - (unsigned char *)chal); - chal->flags[0] = 0x03; - chal->flags[1] = 0x82; - chal->flags[2] = 0x81; - chal->flags[3] = 0xA0; - text2unicode(hostname, (char *)chal->data, 64); - time((time_t *)challenge); - memcpy(challenge+4, SAADDR(¶m->sincr), 4); - challenge[1]^=*SAPORT(¶m->sincr); - for(i = 0; i < 8; i++) challenge[i] ^= myrand(challenge, 8); - memcpy(chal->challenge, challenge, 8); - en64((unsigned char *)tmpbuf, (unsigned char *)buf, (int)((unsigned char *)chal->data - (unsigned char *)chal) + len); -} diff --git a/src/proxy.c b/src/proxy.c index a915db0..c282a7a 100644 --- a/src/proxy.c +++ b/src/proxy.c @@ -381,54 +381,6 @@ for(;;){ param->username = (unsigned char *)mystrdup((char *)username); continue; } -#ifndef NOCRYPT - if(param->srv->usentlm && !strncasecmp((char *)sb, "ntlm", 4)){ - sb+=4; - while(isspace(*sb))sb++; - i = de64(sb, username, 1023); - if(i<=16)continue; - username[i] = 0; - if(strncasecmp((char *)username, "NTLMSSP", 8)) continue; - if(username[8] == 1) { - while( (i = sockgetlinebuf(param, CLIENT, buf, BUFSIZE - 1, '\n', conf.timeouts[STRING_S])) > 2){ - if(i> 15 && (!strncasecmp((char *)(buf), "content-length", 14))){ - buf[i]=0; - sscanf((char *)buf + 15, "%"PRINTF_INT64_MODIFIER"u", &contentlength64); - } - } - while( contentlength64 > 0 && (i = sockgetlinebuf(param, CLIENT, buf, (BUFSIZE < contentlength64)? BUFSIZE - 1:(int)contentlength64, '\n', conf.timeouts[STRING_S])) > 0){ - if ((uint64_t)i > contentlength64) break; - contentlength64-=i; - } - contentlength64 = 0; - if(param->password)myfree(param->password); - param->password = myalloc(32); - param->pwtype = 2; - i = (int)strlen(proxy_stringtable[13]); - memcpy(buf, proxy_stringtable[13], i); - genchallenge(param, (char *)param->password, (char *)buf + i); - memcpy(buf + strlen((char *)buf), "\r\n\r\n", 5); - socksend(param, param->clisock, buf, (int)strlen((char *)buf), conf.timeouts[STRING_S]); - ckeepalive = keepalive = 1; - goto REQUESTEND; - } - if(username[8] == 3 && param->pwtype == 2 && i>=80) { - unsigned offset, len; - - len = username[20] + (((unsigned)username[21]) << 8); - offset = username[24] + (((unsigned)username[25]) << 8); - if(len != 24 || len + offset > (unsigned)i) continue; - memcpy(param->password + 8, username + offset, 24); - len = username[36] + (((unsigned)username[37]) << 8); - offset = username[40] + (((unsigned)username[41]) << 8); - if(len> 255 || len + offset > (unsigned)i) continue; - if(param->username) myfree(param->username); - unicode2text((char *)username+offset, (char *)username+offset, (len>>1)); - param->username = (unsigned char *)mystrdup((char *)username+offset); - } - continue; - } -#endif } #endif if(!isconnect && (