|
|
|
|
@ -186,10 +186,10 @@ 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; |
|
|
|
|
struct re_guts *g; |
|
|
|
|
struct parse *p = &pa; |
|
|
|
|
int i; |
|
|
|
|
size_t len; |
|
|
|
|
|
|
|
|
|
#ifdef REDEBUG |
|
|
|
|
#define GOODFLAGS(f) (f) |
|
|
|
|
@ -287,18 +287,18 @@ int cflags; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- p_ere - ERE parser top level, concatenation and alternation |
|
|
|
|
== static void p_ere(register struct parse *p, int stop); |
|
|
|
|
== static void p_ere(struct parse *p, int stop); |
|
|
|
|
*/ |
|
|
|
|
static void |
|
|
|
|
p_ere(p, stop) |
|
|
|
|
register struct parse *p; |
|
|
|
|
struct parse *p; |
|
|
|
|
int stop; /* character this ERE should end at */ |
|
|
|
|
{ |
|
|
|
|
register char c; |
|
|
|
|
register sopno prevback = 0; |
|
|
|
|
register sopno prevfwd = 0; |
|
|
|
|
register sopno conc; |
|
|
|
|
register int first = 1; /* is this the first alternative? */ |
|
|
|
|
char c; |
|
|
|
|
sopno prevback = 0; |
|
|
|
|
sopno prevfwd = 0; |
|
|
|
|
sopno conc; |
|
|
|
|
int first = 1; /* is this the first alternative? */ |
|
|
|
|
|
|
|
|
|
for (;;) |
|
|
|
|
{ |
|
|
|
|
@ -336,17 +336,17 @@ int stop; /* character this ERE should end at */ |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- 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(struct parse *p); |
|
|
|
|
*/ |
|
|
|
|
static void |
|
|
|
|
p_ere_exp(p) |
|
|
|
|
register struct parse *p; |
|
|
|
|
struct parse *p; |
|
|
|
|
{ |
|
|
|
|
register char c; |
|
|
|
|
register sopno pos; |
|
|
|
|
register int count; |
|
|
|
|
register int count2; |
|
|
|
|
register sopno subno; |
|
|
|
|
char c; |
|
|
|
|
sopno pos; |
|
|
|
|
int count; |
|
|
|
|
int count2; |
|
|
|
|
sopno subno; |
|
|
|
|
int wascaret = 0; |
|
|
|
|
|
|
|
|
|
assert(MORE()); /* caller should have ensured this */ |
|
|
|
|
@ -497,11 +497,11 @@ register struct parse *p; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- p_str - string (no metacharacters) "parser" |
|
|
|
|
== static void p_str(register struct parse *p); |
|
|
|
|
== static void p_str(struct parse *p); |
|
|
|
|
*/ |
|
|
|
|
static void |
|
|
|
|
p_str(p) |
|
|
|
|
register struct parse *p; |
|
|
|
|
struct parse *p; |
|
|
|
|
{ |
|
|
|
|
REQUIRE(MORE(), REG_EMPTY); |
|
|
|
|
while (MORE()) |
|
|
|
|
@ -510,8 +510,8 @@ register struct parse *p; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- p_bre - BRE parser top level, anchoring and concatenation |
|
|
|
|
== static void p_bre(register struct parse *p, register int end1, \
|
|
|
|
|
== register int end2); |
|
|
|
|
== static void p_bre(struct parse *p, int end1, \
|
|
|
|
|
== 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 |
|
|
|
|
@ -522,13 +522,13 @@ register struct parse *p; |
|
|
|
|
*/ |
|
|
|
|
static void |
|
|
|
|
p_bre(p, end1, end2) |
|
|
|
|
register struct parse *p; |
|
|
|
|
register int end1; /* first terminating character */ |
|
|
|
|
register int end2; /* second terminating character */ |
|
|
|
|
struct parse *p; |
|
|
|
|
int end1; /* first terminating character */ |
|
|
|
|
int end2; /* second terminating character */ |
|
|
|
|
{ |
|
|
|
|
register sopno start = HERE(); |
|
|
|
|
register int first = 1; /* first subexpression? */ |
|
|
|
|
register int wasdollar = 0; |
|
|
|
|
sopno start = HERE(); |
|
|
|
|
int first = 1; /* first subexpression? */ |
|
|
|
|
int wasdollar = 0; |
|
|
|
|
|
|
|
|
|
if (EAT('^')) |
|
|
|
|
{ |
|
|
|
|
@ -554,19 +554,19 @@ register int end2; /* second terminating character */ |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- 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 p_simp_re(struct parse *p, int starordinary); |
|
|
|
|
*/ |
|
|
|
|
static int /* was the simple RE an unbackslashed $? */ |
|
|
|
|
p_simp_re(p, starordinary) |
|
|
|
|
register struct parse *p; |
|
|
|
|
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; |
|
|
|
|
int c; |
|
|
|
|
int count; |
|
|
|
|
int count2; |
|
|
|
|
sopno pos; |
|
|
|
|
int i; |
|
|
|
|
sopno subno; |
|
|
|
|
|
|
|
|
|
#define BACKSL (1<<CHAR_BIT) |
|
|
|
|
|
|
|
|
|
@ -689,14 +689,14 @@ int starordinary; /* is a leading * an ordinary character? */ |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- p_count - parse a repetition count |
|
|
|
|
== static int p_count(register struct parse *p); |
|
|
|
|
== static int p_count(struct parse *p); |
|
|
|
|
*/ |
|
|
|
|
static int /* the value */ |
|
|
|
|
p_count(p) |
|
|
|
|
register struct parse *p; |
|
|
|
|
struct parse *p; |
|
|
|
|
{ |
|
|
|
|
register int count = 0; |
|
|
|
|
register int ndigits = 0; |
|
|
|
|
int count = 0; |
|
|
|
|
int ndigits = 0; |
|
|
|
|
|
|
|
|
|
while (MORE() && isdigit(PEEK()) && count <= DUPMAX) |
|
|
|
|
{ |
|
|
|
|
@ -710,17 +710,17 @@ register struct parse *p; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- p_bracket - parse a bracketed character list |
|
|
|
|
== static void p_bracket(register struct parse *p); |
|
|
|
|
== static void p_bracket(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; |
|
|
|
|
struct parse *p; |
|
|
|
|
{ |
|
|
|
|
register cset *cs = allocset(p); |
|
|
|
|
register int invert = 0; |
|
|
|
|
cset *cs = allocset(p); |
|
|
|
|
int invert = 0; |
|
|
|
|
|
|
|
|
|
/* Dept of Truly Sickening Special-Case Kludges */ |
|
|
|
|
if (p->next + 5 < p->end && strncmp(p->next, "[:<:]]", 6) == 0) |
|
|
|
|
@ -753,8 +753,8 @@ register struct parse *p; |
|
|
|
|
|
|
|
|
|
if (p->g->cflags & REG_ICASE) |
|
|
|
|
{ |
|
|
|
|
register int i; |
|
|
|
|
register int ci; |
|
|
|
|
int i; |
|
|
|
|
int ci; |
|
|
|
|
|
|
|
|
|
for (i = p->g->csetsize - 1; i >= 0; i--) |
|
|
|
|
if (CHIN(cs, i) && isalpha(i)) |
|
|
|
|
@ -768,7 +768,7 @@ register struct parse *p; |
|
|
|
|
} |
|
|
|
|
if (invert) |
|
|
|
|
{ |
|
|
|
|
register int i; |
|
|
|
|
int i; |
|
|
|
|
|
|
|
|
|
for (i = p->g->csetsize - 1; i >= 0; i--) |
|
|
|
|
if (CHIN(cs, i)) |
|
|
|
|
@ -794,17 +794,17 @@ register struct parse *p; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- 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(struct parse *p, cset *cs); |
|
|
|
|
*/ |
|
|
|
|
static void |
|
|
|
|
p_b_term(p, cs) |
|
|
|
|
register struct parse *p; |
|
|
|
|
register cset *cs; |
|
|
|
|
struct parse *p; |
|
|
|
|
cset *cs; |
|
|
|
|
{ |
|
|
|
|
register char c; |
|
|
|
|
register char start, |
|
|
|
|
char c; |
|
|
|
|
char start, |
|
|
|
|
finish; |
|
|
|
|
register int i; |
|
|
|
|
int i; |
|
|
|
|
|
|
|
|
|
/* classify what we've got */ |
|
|
|
|
switch ((MORE()) ? PEEK() : '\0') |
|
|
|
|
@ -865,18 +865,18 @@ register cset *cs; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- 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(struct parse *p, cset *cs); |
|
|
|
|
*/ |
|
|
|
|
static void |
|
|
|
|
p_b_cclass(p, cs) |
|
|
|
|
register struct parse *p; |
|
|
|
|
register cset *cs; |
|
|
|
|
struct parse *p; |
|
|
|
|
cset *cs; |
|
|
|
|
{ |
|
|
|
|
register char *sp = p->next; |
|
|
|
|
register struct cclass *cp; |
|
|
|
|
register size_t len; |
|
|
|
|
register char *u; |
|
|
|
|
register char c; |
|
|
|
|
char *sp = p->next; |
|
|
|
|
struct cclass *cp; |
|
|
|
|
size_t len; |
|
|
|
|
char *u; |
|
|
|
|
char c; |
|
|
|
|
|
|
|
|
|
while (MORE() && isalpha(PEEK())) |
|
|
|
|
NEXT(); |
|
|
|
|
@ -900,16 +900,16 @@ register cset *cs; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- p_b_eclass - parse an equivalence-class name and deal with it |
|
|
|
|
== static void p_b_eclass(register struct parse *p, register cset *cs); |
|
|
|
|
== static void p_b_eclass(struct parse *p, cset *cs); |
|
|
|
|
* |
|
|
|
|
* This implementation is incomplete. xxx |
|
|
|
|
*/ |
|
|
|
|
static void |
|
|
|
|
p_b_eclass(p, cs) |
|
|
|
|
register struct parse *p; |
|
|
|
|
register cset *cs; |
|
|
|
|
struct parse *p; |
|
|
|
|
cset *cs; |
|
|
|
|
{ |
|
|
|
|
register char c; |
|
|
|
|
char c; |
|
|
|
|
|
|
|
|
|
c = p_b_coll_elem(p, '='); |
|
|
|
|
CHadd(cs, c); |
|
|
|
|
@ -917,13 +917,13 @@ register cset *cs; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- p_b_symbol - parse a character or [..]ed multicharacter collating symbol |
|
|
|
|
== static char p_b_symbol(register struct parse *p); |
|
|
|
|
== static char p_b_symbol(struct parse *p); |
|
|
|
|
*/ |
|
|
|
|
static char /* value of symbol */ |
|
|
|
|
p_b_symbol(p) |
|
|
|
|
register struct parse *p; |
|
|
|
|
struct parse *p; |
|
|
|
|
{ |
|
|
|
|
register char value; |
|
|
|
|
char value; |
|
|
|
|
|
|
|
|
|
REQUIRE(MORE(), REG_EBRACK); |
|
|
|
|
if (!EATTWO('[', '.')) |
|
|
|
|
@ -937,16 +937,16 @@ register struct parse *p; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- 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 p_b_coll_elem(struct parse *p, int endc); |
|
|
|
|
*/ |
|
|
|
|
static char /* value of collating element */ |
|
|
|
|
p_b_coll_elem(p, endc) |
|
|
|
|
register struct parse *p; |
|
|
|
|
struct parse *p; |
|
|
|
|
int endc; /* name ended by endc,']' */ |
|
|
|
|
{ |
|
|
|
|
register char *sp = p->next; |
|
|
|
|
register struct cname *cp; |
|
|
|
|
register int len; |
|
|
|
|
char *sp = p->next; |
|
|
|
|
struct cname *cp; |
|
|
|
|
int len; |
|
|
|
|
|
|
|
|
|
while (MORE() && !SEETWO(endc, ']')) |
|
|
|
|
NEXT(); |
|
|
|
|
@ -985,17 +985,17 @@ int ch; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- bothcases - emit a dualcase version of a two-case character |
|
|
|
|
== static void bothcases(register struct parse *p, int ch); |
|
|
|
|
== static void bothcases(struct parse *p, int ch); |
|
|
|
|
* |
|
|
|
|
* Boy, is this implementation ever a kludge... |
|
|
|
|
*/ |
|
|
|
|
static void |
|
|
|
|
bothcases(p, ch) |
|
|
|
|
register struct parse *p; |
|
|
|
|
struct parse *p; |
|
|
|
|
int ch; |
|
|
|
|
{ |
|
|
|
|
register char *oldnext = p->next; |
|
|
|
|
register char *oldend = p->end; |
|
|
|
|
char *oldnext = p->next; |
|
|
|
|
char *oldend = p->end; |
|
|
|
|
char bracket[3]; |
|
|
|
|
|
|
|
|
|
assert(othercase(ch) != ch);/* p_bracket() would recurse */ |
|
|
|
|
@ -1012,14 +1012,14 @@ int ch; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- ordinary - emit an ordinary character |
|
|
|
|
== static void ordinary(register struct parse *p, register int ch); |
|
|
|
|
== static void ordinary(struct parse *p, int ch); |
|
|
|
|
*/ |
|
|
|
|
static void |
|
|
|
|
ordinary(p, ch) |
|
|
|
|
register struct parse *p; |
|
|
|
|
register int ch; |
|
|
|
|
struct parse *p; |
|
|
|
|
int ch; |
|
|
|
|
{ |
|
|
|
|
register cat_t *cap = p->g->categories; |
|
|
|
|
cat_t *cap = p->g->categories; |
|
|
|
|
|
|
|
|
|
if ((p->g->cflags & REG_ICASE) && isalpha(ch) && othercase(ch) != ch) |
|
|
|
|
bothcases(p, ch); |
|
|
|
|
@ -1033,16 +1033,16 @@ register int ch; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- nonnewline - emit REG_NEWLINE version of OANY |
|
|
|
|
== static void nonnewline(register struct parse *p); |
|
|
|
|
== static void nonnewline(struct parse *p); |
|
|
|
|
* |
|
|
|
|
* Boy, is this implementation ever a kludge... |
|
|
|
|
*/ |
|
|
|
|
static void |
|
|
|
|
nonnewline(p) |
|
|
|
|
register struct parse *p; |
|
|
|
|
struct parse *p; |
|
|
|
|
{ |
|
|
|
|
register char *oldnext = p->next; |
|
|
|
|
register char *oldend = p->end; |
|
|
|
|
char *oldnext = p->next; |
|
|
|
|
char *oldend = p->end; |
|
|
|
|
char bracket[4]; |
|
|
|
|
|
|
|
|
|
p->next = bracket; |
|
|
|
|
@ -1059,23 +1059,23 @@ register struct parse *p; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- 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(struct parse *p, sopno start, int from, int to); |
|
|
|
|
*/ |
|
|
|
|
static void |
|
|
|
|
repeat(p, start, from, to) |
|
|
|
|
register struct parse *p; |
|
|
|
|
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(); |
|
|
|
|
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; |
|
|
|
|
sopno copy; |
|
|
|
|
|
|
|
|
|
if (p->error != 0) /* head off possible runaway recursion */ |
|
|
|
|
return; |
|
|
|
|
@ -1134,11 +1134,11 @@ int to; /* to this number of times (maybe |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- seterr - set an error condition |
|
|
|
|
== static int seterr(register struct parse *p, int e); |
|
|
|
|
== static int seterr(struct parse *p, int e); |
|
|
|
|
*/ |
|
|
|
|
static int /* useless but makes type checking happy */ |
|
|
|
|
seterr(p, e) |
|
|
|
|
register struct parse *p; |
|
|
|
|
struct parse *p; |
|
|
|
|
int e; |
|
|
|
|
{ |
|
|
|
|
if (p->error == 0) /* keep earliest error condition */ |
|
|
|
|
@ -1150,18 +1150,18 @@ int e; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- allocset - allocate a set of characters for [] |
|
|
|
|
== static cset *allocset(register struct parse *p); |
|
|
|
|
== static cset *allocset(struct parse *p); |
|
|
|
|
*/ |
|
|
|
|
static cset * |
|
|
|
|
allocset(p) |
|
|
|
|
register struct parse *p; |
|
|
|
|
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; |
|
|
|
|
int no = p->g->ncsets++; |
|
|
|
|
size_t nc; |
|
|
|
|
size_t nbytes; |
|
|
|
|
cset *cs; |
|
|
|
|
size_t css = (size_t) p->g->csetsize; |
|
|
|
|
int i; |
|
|
|
|
|
|
|
|
|
if (no >= p->ncsalloc) |
|
|
|
|
{ /* need another column of space */ |
|
|
|
|
@ -1208,16 +1208,16 @@ register struct parse *p; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- freeset - free a now-unused set |
|
|
|
|
== static void freeset(register struct parse *p, register cset *cs); |
|
|
|
|
== static void freeset(struct parse *p, cset *cs); |
|
|
|
|
*/ |
|
|
|
|
static void |
|
|
|
|
freeset(p, cs) |
|
|
|
|
register struct parse *p; |
|
|
|
|
register cset *cs; |
|
|
|
|
struct parse *p; |
|
|
|
|
cset *cs; |
|
|
|
|
{ |
|
|
|
|
register int i; |
|
|
|
|
register cset *top = &p->g->sets[p->g->ncsets]; |
|
|
|
|
register size_t css = (size_t) p->g->csetsize; |
|
|
|
|
int i; |
|
|
|
|
cset *top = &p->g->sets[p->g->ncsets]; |
|
|
|
|
size_t css = (size_t) p->g->csetsize; |
|
|
|
|
|
|
|
|
|
for (i = 0; i < css; i++) |
|
|
|
|
CHsub(cs, i); |
|
|
|
|
@ -1227,7 +1227,7 @@ register cset *cs; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- freezeset - final processing on a set of characters |
|
|
|
|
== static int freezeset(register struct parse *p, register cset *cs); |
|
|
|
|
== static int freezeset(struct parse *p, 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 |
|
|
|
|
@ -1237,14 +1237,14 @@ register cset *cs; |
|
|
|
|
*/ |
|
|
|
|
static int /* set number */ |
|
|
|
|
freezeset(p, cs) |
|
|
|
|
register struct parse *p; |
|
|
|
|
register cset *cs; |
|
|
|
|
struct parse *p; |
|
|
|
|
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; |
|
|
|
|
uch h = cs->hash; |
|
|
|
|
int i; |
|
|
|
|
cset *top = &p->g->sets[p->g->ncsets]; |
|
|
|
|
cset *cs2; |
|
|
|
|
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++) |
|
|
|
|
@ -1269,15 +1269,15 @@ register cset *cs; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- 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 firstch(struct parse *p, cset *cs); |
|
|
|
|
*/ |
|
|
|
|
static int /* character; there is no "none" value */ |
|
|
|
|
firstch(p, cs) |
|
|
|
|
register struct parse *p; |
|
|
|
|
register cset *cs; |
|
|
|
|
struct parse *p; |
|
|
|
|
cset *cs; |
|
|
|
|
{ |
|
|
|
|
register int i; |
|
|
|
|
register size_t css = (size_t) p->g->csetsize; |
|
|
|
|
int i; |
|
|
|
|
size_t css = (size_t) p->g->csetsize; |
|
|
|
|
|
|
|
|
|
for (i = 0; i < css; i++) |
|
|
|
|
if (CHIN(cs, i)) |
|
|
|
|
@ -1288,16 +1288,16 @@ register cset *cs; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- nch - number of characters in a set |
|
|
|
|
== static int nch(register struct parse *p, register cset *cs); |
|
|
|
|
== static int nch(struct parse *p, cset *cs); |
|
|
|
|
*/ |
|
|
|
|
static int |
|
|
|
|
nch(p, cs) |
|
|
|
|
register struct parse *p; |
|
|
|
|
register cset *cs; |
|
|
|
|
struct parse *p; |
|
|
|
|
cset *cs; |
|
|
|
|
{ |
|
|
|
|
register int i; |
|
|
|
|
register size_t css = (size_t) p->g->csetsize; |
|
|
|
|
register int n = 0; |
|
|
|
|
int i; |
|
|
|
|
size_t css = (size_t) p->g->csetsize; |
|
|
|
|
int n = 0; |
|
|
|
|
|
|
|
|
|
for (i = 0; i < css; i++) |
|
|
|
|
if (CHIN(cs, i)) |
|
|
|
|
@ -1307,16 +1307,16 @@ register cset *cs; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- mcadd - add a collating element to a cset |
|
|
|
|
== static void mcadd(register struct parse *p, register cset *cs, \
|
|
|
|
|
== register char *cp); |
|
|
|
|
== static void mcadd(struct parse *p, cset *cs, \
|
|
|
|
|
== char *cp); |
|
|
|
|
*/ |
|
|
|
|
static void |
|
|
|
|
mcadd(p, cs, cp) |
|
|
|
|
register struct parse *p; |
|
|
|
|
register cset *cs; |
|
|
|
|
register char *cp; |
|
|
|
|
struct parse *p; |
|
|
|
|
cset *cs; |
|
|
|
|
char *cp; |
|
|
|
|
{ |
|
|
|
|
register size_t oldend = cs->smultis; |
|
|
|
|
size_t oldend = cs->smultis; |
|
|
|
|
|
|
|
|
|
cs->smultis += strlen(cp) + 1; |
|
|
|
|
if (cs->multis == NULL) |
|
|
|
|
@ -1335,16 +1335,16 @@ register char *cp; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- mcsub - subtract a collating element from a cset |
|
|
|
|
== static void mcsub(register cset *cs, register char *cp); |
|
|
|
|
== static void mcsub(cset *cs, char *cp); |
|
|
|
|
*/ |
|
|
|
|
/*
|
|
|
|
|
static void |
|
|
|
|
mcsub(cs, cp) |
|
|
|
|
register cset *cs; |
|
|
|
|
register char *cp; |
|
|
|
|
cset *cs; |
|
|
|
|
char *cp; |
|
|
|
|
{ |
|
|
|
|
register char *fp = mcfind(cs, cp); |
|
|
|
|
register size_t len = strlen(fp); |
|
|
|
|
char *fp = mcfind(cs, cp); |
|
|
|
|
size_t len = strlen(fp); |
|
|
|
|
|
|
|
|
|
assert(fp != NULL); |
|
|
|
|
memmove(fp, fp + len + 1, |
|
|
|
|
@ -1364,13 +1364,13 @@ register char *cp; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- mcin - is a collating element in a cset? |
|
|
|
|
== static int mcin(register cset *cs, register char *cp); |
|
|
|
|
== static int mcin(cset *cs, char *cp); |
|
|
|
|
*/ |
|
|
|
|
/*
|
|
|
|
|
static int |
|
|
|
|
mcin(cs, cp) |
|
|
|
|
register cset *cs; |
|
|
|
|
register char *cp; |
|
|
|
|
cset *cs; |
|
|
|
|
char *cp; |
|
|
|
|
{ |
|
|
|
|
return(mcfind(cs, cp) != NULL); |
|
|
|
|
} |
|
|
|
|
@ -1378,15 +1378,15 @@ register char *cp; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- mcfind - find a collating element in a cset |
|
|
|
|
== static char *mcfind(register cset *cs, register char *cp); |
|
|
|
|
== static char *mcfind(cset *cs, char *cp); |
|
|
|
|
*/ |
|
|
|
|
/*
|
|
|
|
|
static char * |
|
|
|
|
mcfind(cs, cp) |
|
|
|
|
register cset *cs; |
|
|
|
|
register char *cp; |
|
|
|
|
cset *cs; |
|
|
|
|
char *cp; |
|
|
|
|
{ |
|
|
|
|
register char *p; |
|
|
|
|
char *p; |
|
|
|
|
|
|
|
|
|
if (cs->multis == NULL) |
|
|
|
|
return(NULL); |
|
|
|
|
@ -1398,47 +1398,47 @@ register char *cp; |
|
|
|
|
*/ |
|
|
|
|
/*
|
|
|
|
|
- mcinvert - invert the list of collating elements in a cset |
|
|
|
|
== static void mcinvert(register struct parse *p, register cset *cs); |
|
|
|
|
== static void mcinvert(struct parse *p, 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; |
|
|
|
|
struct parse *p; |
|
|
|
|
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); |
|
|
|
|
== static void mccase(struct parse *p, 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; |
|
|
|
|
struct parse *p; |
|
|
|
|
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 isinsets(struct re_guts *g, int c); |
|
|
|
|
*/ |
|
|
|
|
static int /* predicate */ |
|
|
|
|
isinsets(g, c) |
|
|
|
|
register struct re_guts *g; |
|
|
|
|
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; |
|
|
|
|
uch *col; |
|
|
|
|
int i; |
|
|
|
|
int ncols = (g->ncsets + (CHAR_BIT - 1)) / CHAR_BIT; |
|
|
|
|
unsigned uc = (unsigned char) c; |
|
|
|
|
|
|
|
|
|
for (i = 0, col = g->setbits; i < ncols; i++, col += g->csetsize) |
|
|
|
|
if (col[uc] != 0) |
|
|
|
|
@ -1448,19 +1448,19 @@ int c; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- samesets - are these two characters in exactly the same sets? |
|
|
|
|
== static int samesets(register struct re_guts *g, int c1, int c2); |
|
|
|
|
== static int samesets(struct re_guts *g, int c1, int c2); |
|
|
|
|
*/ |
|
|
|
|
static int /* predicate */ |
|
|
|
|
samesets(g, c1, c2) |
|
|
|
|
register struct re_guts *g; |
|
|
|
|
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; |
|
|
|
|
uch *col; |
|
|
|
|
int i; |
|
|
|
|
int ncols = (g->ncsets + (CHAR_BIT - 1)) / CHAR_BIT; |
|
|
|
|
unsigned uc1 = (unsigned char) c1; |
|
|
|
|
unsigned uc2 = (unsigned char) c2; |
|
|
|
|
|
|
|
|
|
for (i = 0, col = g->setbits; i < ncols; i++, col += g->csetsize) |
|
|
|
|
if (col[uc1] != col[uc2]) |
|
|
|
|
@ -1470,17 +1470,17 @@ int c2; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- categorize - sort out character categories |
|
|
|
|
== static void categorize(struct parse *p, register struct re_guts *g); |
|
|
|
|
== static void categorize(struct parse *p, struct re_guts *g); |
|
|
|
|
*/ |
|
|
|
|
static void |
|
|
|
|
categorize(p, g) |
|
|
|
|
struct parse *p; |
|
|
|
|
register struct re_guts *g; |
|
|
|
|
struct re_guts *g; |
|
|
|
|
{ |
|
|
|
|
register cat_t *cats = g->categories; |
|
|
|
|
register int c; |
|
|
|
|
register int c2; |
|
|
|
|
register cat_t cat; |
|
|
|
|
cat_t *cats = g->categories; |
|
|
|
|
int c; |
|
|
|
|
int c2; |
|
|
|
|
cat_t cat; |
|
|
|
|
|
|
|
|
|
/* avoid making error situations worse */ |
|
|
|
|
if (p->error != 0) |
|
|
|
|
@ -1499,16 +1499,16 @@ register struct re_guts *g; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- dupl - emit a duplicate of a bunch of sops |
|
|
|
|
== static sopno dupl(register struct parse *p, sopno start, sopno finish); |
|
|
|
|
== static sopno dupl(struct parse *p, sopno start, sopno finish); |
|
|
|
|
*/ |
|
|
|
|
static sopno /* start of duplicate */ |
|
|
|
|
dupl(p, start, finish) |
|
|
|
|
register struct parse *p; |
|
|
|
|
struct parse *p; |
|
|
|
|
sopno start; /* from here */ |
|
|
|
|
sopno finish; /* to this less one */ |
|
|
|
|
{ |
|
|
|
|
register sopno ret = HERE(); |
|
|
|
|
register sopno len = finish - start; |
|
|
|
|
sopno ret = HERE(); |
|
|
|
|
sopno len = finish - start; |
|
|
|
|
|
|
|
|
|
assert(finish >= start); |
|
|
|
|
if (len == 0) |
|
|
|
|
@ -1523,7 +1523,7 @@ sopno finish; /* to this less one */ |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- doemit - emit a strip operator |
|
|
|
|
== static void doemit(register struct parse *p, sop op, size_t opnd); |
|
|
|
|
== static void doemit(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 |
|
|
|
|
@ -1531,7 +1531,7 @@ sopno finish; /* to this less one */ |
|
|
|
|
*/ |
|
|
|
|
static void |
|
|
|
|
doemit(p, op, opnd) |
|
|
|
|
register struct parse *p; |
|
|
|
|
struct parse *p; |
|
|
|
|
sop op; |
|
|
|
|
size_t opnd; |
|
|
|
|
{ |
|
|
|
|
@ -1553,18 +1553,18 @@ size_t 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(struct parse *p, sop op, size_t opnd, sopno pos); |
|
|
|
|
*/ |
|
|
|
|
static void |
|
|
|
|
doinsert(p, op, opnd, pos) |
|
|
|
|
register struct parse *p; |
|
|
|
|
struct parse *p; |
|
|
|
|
sop op; |
|
|
|
|
size_t opnd; |
|
|
|
|
sopno pos; |
|
|
|
|
{ |
|
|
|
|
register sopno sn; |
|
|
|
|
register sop s; |
|
|
|
|
register int i; |
|
|
|
|
sopno sn; |
|
|
|
|
sop s; |
|
|
|
|
int i; |
|
|
|
|
|
|
|
|
|
/* avoid making error situations worse */ |
|
|
|
|
if (p->error != 0) |
|
|
|
|
@ -1596,12 +1596,12 @@ sopno pos; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- dofwd - complete a forward reference |
|
|
|
|
== static void dofwd(register struct parse *p, sopno pos, sop value); |
|
|
|
|
== static void dofwd(struct parse *p, sopno pos, sop value); |
|
|
|
|
*/ |
|
|
|
|
static void |
|
|
|
|
dofwd(p, pos, value) |
|
|
|
|
register struct parse *p; |
|
|
|
|
register sopno pos; |
|
|
|
|
struct parse *p; |
|
|
|
|
sopno pos; |
|
|
|
|
sop value; |
|
|
|
|
{ |
|
|
|
|
/* avoid making error situations worse */ |
|
|
|
|
@ -1614,14 +1614,14 @@ sop value; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- enlarge - enlarge the strip |
|
|
|
|
== static void enlarge(register struct parse *p, sopno size); |
|
|
|
|
== static void enlarge(struct parse *p, sopno size); |
|
|
|
|
*/ |
|
|
|
|
static void |
|
|
|
|
enlarge(p, size) |
|
|
|
|
register struct parse *p; |
|
|
|
|
register sopno size; |
|
|
|
|
struct parse *p; |
|
|
|
|
sopno size; |
|
|
|
|
{ |
|
|
|
|
register sop *sp; |
|
|
|
|
sop *sp; |
|
|
|
|
|
|
|
|
|
if (p->ssize >= size) |
|
|
|
|
return; |
|
|
|
|
@ -1638,12 +1638,12 @@ register sopno size; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- stripsnug - compact the strip |
|
|
|
|
== static void stripsnug(register struct parse *p, register struct re_guts *g); |
|
|
|
|
== static void stripsnug(struct parse *p, struct re_guts *g); |
|
|
|
|
*/ |
|
|
|
|
static void |
|
|
|
|
stripsnug(p, g) |
|
|
|
|
register struct parse *p; |
|
|
|
|
register struct re_guts *g; |
|
|
|
|
struct parse *p; |
|
|
|
|
struct re_guts *g; |
|
|
|
|
{ |
|
|
|
|
g->nstates = p->slen; |
|
|
|
|
g->strip = (sop *) realloc((char *) p->strip, p->slen * sizeof(sop)); |
|
|
|
|
@ -1656,7 +1656,7 @@ register struct re_guts *g; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- findmust - fill in must and mlen with longest mandatory literal string |
|
|
|
|
== static void findmust(register struct parse *p, register struct re_guts *g); |
|
|
|
|
== static void findmust(struct parse *p, 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 |
|
|
|
|
@ -1667,15 +1667,15 @@ register struct re_guts *g; |
|
|
|
|
static void |
|
|
|
|
findmust(p, g) |
|
|
|
|
struct parse *p; |
|
|
|
|
register struct re_guts *g; |
|
|
|
|
struct re_guts *g; |
|
|
|
|
{ |
|
|
|
|
register sop *scan; |
|
|
|
|
sop *scan; |
|
|
|
|
sop *start = 0; |
|
|
|
|
register sop *newstart = 0; |
|
|
|
|
register sopno newlen; |
|
|
|
|
register sop s; |
|
|
|
|
register char *cp; |
|
|
|
|
register sopno i; |
|
|
|
|
sop *newstart = 0; |
|
|
|
|
sopno newlen; |
|
|
|
|
sop s; |
|
|
|
|
char *cp; |
|
|
|
|
sopno i; |
|
|
|
|
|
|
|
|
|
/* avoid making error situations worse */ |
|
|
|
|
if (p->error != 0) |
|
|
|
|
@ -1750,17 +1750,17 @@ register struct re_guts *g; |
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
- pluscount - count + nesting |
|
|
|
|
== static sopno pluscount(register struct parse *p, register struct re_guts *g); |
|
|
|
|
== static sopno pluscount(struct parse *p, struct re_guts *g); |
|
|
|
|
*/ |
|
|
|
|
static sopno /* nesting depth */ |
|
|
|
|
pluscount(p, g) |
|
|
|
|
struct parse *p; |
|
|
|
|
register struct re_guts *g; |
|
|
|
|
struct re_guts *g; |
|
|
|
|
{ |
|
|
|
|
register sop *scan; |
|
|
|
|
register sop s; |
|
|
|
|
register sopno plusnest = 0; |
|
|
|
|
register sopno maxnest = 0; |
|
|
|
|
sop *scan; |
|
|
|
|
sop s; |
|
|
|
|
sopno plusnest = 0; |
|
|
|
|
sopno maxnest = 0; |
|
|
|
|
|
|
|
|
|
if (p->error != 0) |
|
|
|
|
return (0); /* there may not be an OEND */ |
|
|
|
|
|
Bruce Momjian
28 years ago