version 1.7, 2015/11/19 19:43:40 |
version 1.8, 2015/11/19 22:55:13 |
|
|
|
|
/* The United States Government has rights in this work pursuant */ |
/* The United States Government has rights in this work pursuant */ |
/* to contract no. DE-AC03-76SF00098 between the United States */ |
/* to contract no. DE-AC03-76SF00098 between the United States */ |
/* Department of Energy and the University of California. */ |
/* Department of Energy and the University of California. */ |
|
|
/* This file is part of flex. */ |
/* This file is part of flex. */ |
|
|
|
|
|
|
/* return true if the chr is in the ccl. Takes negation into account. */ |
/* return true if the chr is in the ccl. Takes negation into account. */ |
static bool |
static bool |
ccl_contains (const int cclp, const int ch) |
ccl_contains(const int cclp, const int ch) |
{ |
{ |
int ind, len, i; |
int ind, len, i; |
|
|
len = ccllen[cclp]; |
len = ccllen[cclp]; |
ind = cclmap[cclp]; |
ind = cclmap[cclp]; |
|
|
if (ccltbl[ind + i] == ch) |
if (ccltbl[ind + i] == ch) |
return !cclng[cclp]; |
return !cclng[cclp]; |
|
|
return cclng[cclp]; |
return cclng[cclp]; |
} |
} |
|
|
|
|
/* ccladd - add a single character to a ccl */ |
/* ccladd - add a single character to a ccl */ |
|
|
void ccladd (cclp, ch) |
void |
int cclp; |
ccladd(cclp, ch) |
int ch; |
int cclp; |
|
int ch; |
{ |
{ |
int ind, len, newpos, i; |
int ind, len, newpos, i; |
|
|
check_char (ch); |
check_char(ch); |
|
|
len = ccllen[cclp]; |
len = ccllen[cclp]; |
ind = cclmap[cclp]; |
ind = cclmap[cclp]; |
|
|
|
|
++num_reallocs; |
++num_reallocs; |
|
|
ccltbl = reallocate_Character_array (ccltbl, |
ccltbl = reallocate_Character_array(ccltbl, |
current_max_ccl_tbl_size); |
current_max_ccl_tbl_size); |
} |
} |
|
|
ccllen[cclp] = len + 1; |
ccllen[cclp] = len + 1; |
ccltbl[newpos] = ch; |
ccltbl[newpos] = ch; |
} |
} |
|
|
/* dump_cclp - same thing as list_character_set, but for cclps. */ |
/* dump_cclp - same thing as list_character_set, but for cclps. */ |
|
|
static void dump_cclp (FILE* file, int cclp) |
static void |
|
dump_cclp(FILE * file, int cclp) |
{ |
{ |
int i; |
int i; |
|
|
putc ('[', file); |
putc('[', file); |
|
|
for (i = 0; i < csize; ++i) { |
for (i = 0; i < csize; ++i) { |
if (ccl_contains(cclp, i)){ |
if (ccl_contains(cclp, i)) { |
int start_char = i; |
int start_char = i; |
|
|
putc (' ', file); |
putc(' ', file); |
|
|
fputs (readable_form (i), file); |
fputs(readable_form(i), file); |
|
|
while (++i < csize && ccl_contains(cclp,i)) ; |
while (++i < csize && ccl_contains(cclp, i)); |
|
|
if (i - 1 > start_char) |
if (i - 1 > start_char) |
/* this was a run */ |
/* this was a run */ |
fprintf (file, "-%s", |
fprintf(file, "-%s", |
readable_form (i - 1)); |
readable_form(i - 1)); |
|
|
putc (' ', file); |
putc(' ', file); |
} |
} |
} |
} |
|
|
putc (']', file); |
putc(']', file); |
} |
} |
|
|
|
|
|
|
/* ccl_set_diff - create a new ccl as the set difference of the two given ccls. */ |
/* ccl_set_diff - create a new ccl as the set difference of the two given ccls. */ |
int |
int |
ccl_set_diff (int a, int b) |
ccl_set_diff(int a, int b) |
{ |
{ |
int d, ch; |
int d, ch; |
|
|
/* create new class */ |
/* create new class */ |
d = cclinit(); |
d = cclinit(); |
|
|
/* In order to handle negation, we spin through all possible chars, |
/* |
* addding each char in a that is not in b. |
* In order to handle negation, we spin through all possible chars, |
* (This could be O(n^2), but n is small and bounded.) |
* addding each char in a that is not in b. (This could be O(n^2), |
*/ |
* but n is small and bounded.) |
for ( ch = 0; ch < csize; ++ch ) |
*/ |
if (ccl_contains (a, ch) && !ccl_contains(b, ch)) |
for (ch = 0; ch < csize; ++ch) |
ccladd (d, ch); |
if (ccl_contains(a, ch) && !ccl_contains(b, ch)) |
|
ccladd(d, ch); |
|
|
/* debug */ |
/* debug */ |
if (0){ |
if (0) { |
fprintf(stderr, "ccl_set_diff ("); |
fprintf(stderr, "ccl_set_diff ("); |
fprintf(stderr, "\n "); |
fprintf(stderr, "\n "); |
dump_cclp (stderr, a); |
dump_cclp(stderr, a); |
fprintf(stderr, "\n "); |
fprintf(stderr, "\n "); |
dump_cclp (stderr, b); |
dump_cclp(stderr, b); |
fprintf(stderr, "\n "); |
fprintf(stderr, "\n "); |
dump_cclp (stderr, d); |
dump_cclp(stderr, d); |
fprintf(stderr, "\n)\n"); |
fprintf(stderr, "\n)\n"); |
} |
} |
return d; |
return d; |
} |
} |
|
|
/* ccl_set_union - create a new ccl as the set union of the two given ccls. */ |
/* ccl_set_union - create a new ccl as the set union of the two given ccls. */ |
int |
int |
ccl_set_union (int a, int b) |
ccl_set_union(int a, int b) |
{ |
{ |
int d, i; |
int d, i; |
|
|
/* create new class */ |
/* create new class */ |
d = cclinit(); |
d = cclinit(); |
|
|
/* Add all of a */ |
/* Add all of a */ |
for (i = 0; i < ccllen[a]; ++i) |
for (i = 0; i < ccllen[a]; ++i) |
ccladd (d, ccltbl[cclmap[a] + i]); |
ccladd(d, ccltbl[cclmap[a] + i]); |
|
|
/* Add all of b */ |
/* Add all of b */ |
for (i = 0; i < ccllen[b]; ++i) |
for (i = 0; i < ccllen[b]; ++i) |
ccladd (d, ccltbl[cclmap[b] + i]); |
ccladd(d, ccltbl[cclmap[b] + i]); |
|
|
/* debug */ |
/* debug */ |
if (0){ |
if (0) { |
fprintf(stderr, "ccl_set_union (%d + %d = %d", a, b, d); |
fprintf(stderr, "ccl_set_union (%d + %d = %d", a, b, d); |
fprintf(stderr, "\n "); |
fprintf(stderr, "\n "); |
dump_cclp (stderr, a); |
dump_cclp(stderr, a); |
fprintf(stderr, "\n "); |
fprintf(stderr, "\n "); |
dump_cclp (stderr, b); |
dump_cclp(stderr, b); |
fprintf(stderr, "\n "); |
fprintf(stderr, "\n "); |
dump_cclp (stderr, d); |
dump_cclp(stderr, d); |
fprintf(stderr, "\n)\n"); |
fprintf(stderr, "\n)\n"); |
} |
} |
return d; |
return d; |
} |
} |
|
|
|
|
/* cclinit - return an empty ccl */ |
/* cclinit - return an empty ccl */ |
|
|
int cclinit () |
int |
|
cclinit() |
{ |
{ |
if (++lastccl >= current_maxccls) { |
if (++lastccl >= current_maxccls) { |
current_maxccls += MAX_CCLS_INCREMENT; |
current_maxccls += MAX_CCLS_INCREMENT; |
|
|
++num_reallocs; |
++num_reallocs; |
|
|
cclmap = |
cclmap = |
reallocate_integer_array (cclmap, current_maxccls); |
reallocate_integer_array(cclmap, current_maxccls); |
ccllen = |
ccllen = |
reallocate_integer_array (ccllen, current_maxccls); |
reallocate_integer_array(ccllen, current_maxccls); |
cclng = reallocate_integer_array (cclng, current_maxccls); |
cclng = reallocate_integer_array(cclng, current_maxccls); |
ccl_has_nl = |
ccl_has_nl = |
reallocate_bool_array (ccl_has_nl, |
reallocate_bool_array(ccl_has_nl, |
current_maxccls); |
current_maxccls); |
} |
} |
|
|
if (lastccl == 1) |
if (lastccl == 1) |
/* we're making the first ccl */ |
/* we're making the first ccl */ |
cclmap[lastccl] = 0; |
cclmap[lastccl] = 0; |
|
|
else |
else |
/* The new pointer is just past the end of the last ccl. |
/* |
* Since the cclmap points to the \first/ character of a |
* The new pointer is just past the end of the last ccl. |
* ccl, adding the length of the ccl to the cclmap pointer |
* Since the cclmap points to the \first/ character of a ccl, |
* will produce a cursor to the first free space. |
* adding the length of the ccl to the cclmap pointer will |
|
* produce a cursor to the first free space. |
*/ |
*/ |
cclmap[lastccl] = |
cclmap[lastccl] = |
cclmap[lastccl - 1] + ccllen[lastccl - 1]; |
cclmap[lastccl - 1] + ccllen[lastccl - 1]; |
|
|
ccllen[lastccl] = 0; |
ccllen[lastccl] = 0; |
cclng[lastccl] = 0; /* ccl's start out life un-negated */ |
cclng[lastccl] = 0; /* ccl's start out life un-negated */ |
|
|
|
|
/* cclnegate - negate the given ccl */ |
/* cclnegate - negate the given ccl */ |
|
|
void cclnegate (cclp) |
void |
int cclp; |
cclnegate(cclp) |
|
int cclp; |
{ |
{ |
cclng[cclp] = 1; |
cclng[cclp] = 1; |
ccl_has_nl[cclp] = !ccl_has_nl[cclp]; |
ccl_has_nl[cclp] = !ccl_has_nl[cclp]; |
|
|
* has a non-zero value in the cset array. |
* has a non-zero value in the cset array. |
*/ |
*/ |
|
|
void list_character_set (file, cset) |
void |
FILE *file; |
list_character_set(file, cset) |
int cset[]; |
FILE *file; |
|
int cset[]; |
{ |
{ |
int i; |
int i; |
|
|
putc ('[', file); |
putc('[', file); |
|
|
for (i = 0; i < csize; ++i) { |
for (i = 0; i < csize; ++i) { |
if (cset[i]) { |
if (cset[i]) { |
int start_char = i; |
int start_char = i; |
|
|
putc (' ', file); |
putc(' ', file); |
|
|
fputs (readable_form (i), file); |
fputs(readable_form(i), file); |
|
|
while (++i < csize && cset[i]) ; |
while (++i < csize && cset[i]); |
|
|
if (i - 1 > start_char) |
if (i - 1 > start_char) |
/* this was a run */ |
/* this was a run */ |
fprintf (file, "-%s", |
fprintf(file, "-%s", |
readable_form (i - 1)); |
readable_form(i - 1)); |
|
|
putc (' ', file); |
putc(' ', file); |
} |
} |
} |
} |
|
|
putc (']', file); |
putc(']', file); |
} |
} |
|
|
/** Determines if the range [c1-c2] is unambiguous in a case-insensitive |
/** Determines if the range [c1-c2] is unambiguous in a case-insensitive |
|
|
* @param c2 the upper end of the range |
* @param c2 the upper end of the range |
* @return true if [c1-c2] is not ambiguous for a caseless scanner. |
* @return true if [c1-c2] is not ambiguous for a caseless scanner. |
*/ |
*/ |
bool range_covers_case (int c1, int c2) |
bool |
|
range_covers_case(int c1, int c2) |
{ |
{ |
int i, o; |
int i, o; |
|
|
for (i = c1; i <= c2; i++) { |
for (i = c1; i <= c2; i++) { |
if (has_case (i)) { |
if (has_case(i)) { |
o = reverse_case (i); |
o = reverse_case(i); |
if (o < c1 || c2 < o) |
if (o < c1 || c2 < o) |
return false; |
return false; |
} |
} |
|
|
/** Reverse the case of a character, if possible. |
/** Reverse the case of a character, if possible. |
* @return c if case-reversal does not apply. |
* @return c if case-reversal does not apply. |
*/ |
*/ |
int reverse_case (int c) |
int |
|
reverse_case(int c) |
{ |
{ |
return isupper (c) ? tolower (c) : (islower (c) ? toupper (c) : c); |
return isupper(c) ? tolower(c) : (islower(c) ? toupper(c) : c); |
} |
} |
|
|
/** Return true if c is uppercase or lowercase. */ |
/** Return true if c is uppercase or lowercase. */ |
bool has_case (int c) |
bool |
|
has_case(int c) |
{ |
{ |
return (isupper (c) || islower (c)) ? true : false; |
return (isupper(c) || islower(c)) ? true : false; |
} |
} |