Feature #5588 » 5588_negative_lookahead.patch
| include/ruby/oniguruma.h | ||
|---|---|---|
|
#define ONIG_OPTION_NONE 0U
|
||
|
#define ONIG_OPTION_IGNORECASE 1U
|
||
|
#define ONIG_OPTION_EXTEND (ONIG_OPTION_IGNORECASE << 1)
|
||
|
#define ONIG_OPTION_MULTILINE (ONIG_OPTION_EXTEND << 1)
|
||
|
#define ONIG_OPTION_NEGATE (ONIG_OPTION_EXTEND << 1)
|
||
|
#define ONIG_OPTION_MULTILINE (ONIG_OPTION_NEGATE << 1)
|
||
|
#define ONIG_OPTION_SINGLELINE (ONIG_OPTION_MULTILINE << 1)
|
||
|
#define ONIG_OPTION_FIND_LONGEST (ONIG_OPTION_SINGLELINE << 1)
|
||
|
#define ONIG_OPTION_FIND_NOT_EMPTY (ONIG_OPTION_FIND_LONGEST << 1)
|
||
| re.c | ||
|---|---|---|
|
#define KCODE_FIXED FL_USER4
|
||
|
#define ARG_REG_OPTION_MASK \
|
||
|
(ONIG_OPTION_IGNORECASE|ONIG_OPTION_MULTILINE|ONIG_OPTION_EXTEND)
|
||
|
(ONIG_OPTION_IGNORECASE|ONIG_OPTION_MULTILINE|ONIG_OPTION_EXTEND|ONIG_OPTION_NEGATE)
|
||
|
#define ARG_ENCODING_FIXED 16
|
||
|
#define ARG_ENCODING_NONE 32
|
||
| ... | ... | |
|
case 'm':
|
||
|
val = ONIG_OPTION_MULTILINE;
|
||
|
break;
|
||
|
case 'v':
|
||
|
val = ONIG_OPTION_NEGATE;
|
||
|
break;
|
||
|
default:
|
||
|
val = 0;
|
||
|
break;
|
||
| ... | ... | |
|
}
|
||
|
static char *
|
||
|
option_to_str(char str[4], int options)
|
||
|
option_to_str(char str[5], int options)
|
||
|
{
|
||
|
char *p = str;
|
||
|
if (options & ONIG_OPTION_MULTILINE) *p++ = 'm';
|
||
|
if (options & ONIG_OPTION_IGNORECASE) *p++ = 'i';
|
||
|
if (options & ONIG_OPTION_EXTEND) *p++ = 'x';
|
||
|
if (options & ONIG_OPTION_NEGATE) *p++ = 'v';
|
||
|
*p = 0;
|
||
|
return str;
|
||
|
}
|
||
| ... | ... | |
|
rb_reg_expr_str(str, s, len, enc, resenc);
|
||
|
rb_str_buf_cat2(str, "/");
|
||
|
if (re) {
|
||
|
char opts[4];
|
||
|
char opts[5];
|
||
|
rb_reg_check(re);
|
||
|
if (*option_to_str(opts, RREGEXP(re)->ptr->options))
|
||
|
rb_str_buf_cat2(str, opts);
|
||
| ... | ... | |
|
* generally more readable version of <i>rxp</i>.
|
||
|
*
|
||
|
* r1 = /ab+c/ix #=> /ab+c/ix
|
||
|
* s1 = r1.to_s #=> "(?ix-m:ab+c)"
|
||
|
* r2 = Regexp.new(s1) #=> /(?ix-m:ab+c)/
|
||
|
* s1 = r1.to_s #=> "(?ix-mv:ab+c)"
|
||
|
* r2 = Regexp.new(s1) #=> /(?ix-mv:ab+c)/
|
||
|
* r1 == r2 #=> false
|
||
|
* r1.source #=> "ab+c"
|
||
|
* r2.source #=> "(?ix-m:ab+c)"
|
||
|
* r2.source #=> "(?ix-mv:ab+c)"
|
||
|
*/
|
||
|
static VALUE
|
||
|
rb_reg_to_s(VALUE re)
|
||
|
{
|
||
|
int options, opt;
|
||
|
const int embeddable = ONIG_OPTION_MULTILINE|ONIG_OPTION_IGNORECASE|ONIG_OPTION_EXTEND;
|
||
|
const int embeddable = ARG_REG_OPTION_MASK;
|
||
|
long len;
|
||
|
const UChar* ptr;
|
||
|
VALUE str = rb_str_buf_new2("(?");
|
||
|
char optbuf[5];
|
||
|
char optbuf[6];
|
||
|
rb_encoding *enc = rb_enc_get(re);
|
||
|
rb_reg_check(re);
|
||
| ... | ... | |
|
static VALUE
|
||
|
rb_enc_reg_error_desc(const char *s, long len, rb_encoding *enc, int options, const char *err)
|
||
|
{
|
||
|
char opts[6];
|
||
|
char opts[7];
|
||
|
VALUE desc = rb_str_buf_new2(err);
|
||
|
rb_encoding *resenc = rb_default_internal_encoding();
|
||
|
if (resenc == NULL) resenc = rb_default_external_encoding();
|
||
| ... | ... | |
|
* options are propagated, and new options may not be specified (a change as of
|
||
|
* Ruby 1.8). If <i>options</i> is a <code>Fixnum</code>, it should be one or
|
||
|
* more of the constants <code>Regexp::EXTENDED</code>,
|
||
|
* <code>Regexp::IGNORECASE</code>, and <code>Regexp::MULTILINE</code>,
|
||
|
* <em>or</em>-ed together. Otherwise, if <i>options</i> is not
|
||
|
* <code>nil</code>, the regexp will be case insensitive.
|
||
|
* When the <i>lang</i> parameter is `n' or `N' sets the regexp no encoding.
|
||
|
* <code>Regexp::IGNORECASE</code>, <code>Regexp::MULTILINE</code>, and
|
||
|
* <code>Regexp::NEGATED</code>, <em>or</em>-ed together. Otherwise, if
|
||
|
* <i>options</i> is not <code>nil</code>, the regexp will be case
|
||
|
* insensitive. When the <i>lang</i> parameter is `n' or `N' sets the regexp
|
||
|
* no encoding.
|
||
|
*
|
||
|
* r1 = Regexp.new('^a-z+:\\s+\w+') #=> /^a-z+:\s+\w+/
|
||
|
* r2 = Regexp.new('cat', true) #=> /cat/i
|
||
| ... | ... | |
|
* Regexp.union("a+b*c") #=> /a\+b\*c/
|
||
|
* Regexp.union("skiing", "sledding") #=> /skiing|sledding/
|
||
|
* Regexp.union(["skiing", "sledding"]) #=> /skiing|sledding/
|
||
|
* Regexp.union(/dogs/, /cats/i) #=> /(?-mix:dogs)|(?i-mx:cats)/
|
||
|
* Regexp.union(/dogs/, /cats/i) #=> /(?-mixv:dogs)|(?i-mxv:cats)/
|
||
|
*/
|
||
|
static VALUE
|
||
|
rb_reg_s_union_m(VALUE self, VALUE args)
|
||
| ... | ... | |
|
/* see Regexp.options and Regexp.new */
|
||
|
rb_define_const(rb_cRegexp, "EXTENDED", INT2FIX(ONIG_OPTION_EXTEND));
|
||
|
/* see Regexp.options and Regexp.new */
|
||
|
rb_define_const(rb_cRegexp, "NEGATED", INT2FIX(ONIG_OPTION_NEGATE));
|
||
|
/* see Regexp.options and Regexp.new */
|
||
|
rb_define_const(rb_cRegexp, "MULTILINE", INT2FIX(ONIG_OPTION_MULTILINE));
|
||
|
/* see Regexp.options and Regexp.new */
|
||
|
rb_define_const(rb_cRegexp, "FIXEDENCODING", INT2FIX(ARG_ENCODING_FIXED));
|
||
| regcomp.c | ||
|---|---|---|
|
fprintf(f, "<enclose:%"PRIxPTR"> ", (intptr_t)node);
|
||
|
switch (NENCLOSE(node)->type) {
|
||
|
case ENCLOSE_OPTION:
|
||
|
fprintf(f, "option:%d\n", NENCLOSE(node)->option);
|
||
|
print_indent_tree(f, NENCLOSE(node)->target, indent + add);
|
||
|
fprintf(f, "option:%d", NENCLOSE(node)->option);
|
||
|
break;
|
||
|
case ENCLOSE_MEMORY:
|
||
|
fprintf(f, "memory:%d", NENCLOSE(node)->regnum);
|
||
| regexec.c | ||
|---|---|---|
|
if (r != ONIG_MISMATCH)
|
||
|
fprintf(stderr, "onig_search: error %d\n", r);
|
||
|
#endif
|
||
|
return r;
|
||
|
goto negate;
|
||
|
mismatch_no_msa:
|
||
|
r = ONIG_MISMATCH;
|
||
| ... | ... | |
|
if (r != ONIG_MISMATCH)
|
||
|
fprintf(stderr, "onig_search: error %d\n", r);
|
||
|
#endif
|
||
|
return r;
|
||
|
goto negate;
|
||
|
match:
|
||
|
ONIG_STATE_DEC_THREAD(reg);
|
||
|
MATCH_ARG_FREE(msa);
|
||
|
return s - str;
|
||
|
r = s - str;
|
||
|
/* fall */
|
||
|
negate:
|
||
|
if (r >= ONIG_MISMATCH && IS_NEGATE(reg->options))
|
||
|
return r == ONIG_MISMATCH ? ONIG_NORMAL : ONIG_MISMATCH;
|
||
|
return r;
|
||
|
}
|
||
|
extern OnigEncoding
|
||
| regint.h | ||
|---|---|---|
|
#define IS_MULTILINE(option) ((option) & ONIG_OPTION_MULTILINE)
|
||
|
#define IS_IGNORECASE(option) ((option) & ONIG_OPTION_IGNORECASE)
|
||
|
#define IS_EXTEND(option) ((option) & ONIG_OPTION_EXTEND)
|
||
|
#define IS_NEGATE(option) ((option) & ONIG_OPTION_NEGATE)
|
||
|
#define IS_FIND_LONGEST(option) ((option) & ONIG_OPTION_FIND_LONGEST)
|
||
|
#define IS_FIND_NOT_EMPTY(option) ((option) & ONIG_OPTION_FIND_NOT_EMPTY)
|
||
|
#define IS_FIND_CONDITION(option) ((option) & \
|
||
| regparse.c | ||
|---|---|---|
|
#ifdef USE_POSIXLINE_OPTION
|
||
|
case 'p':
|
||
|
#endif
|
||
|
case '-': case 'i': case 'm': case 's': case 'x':
|
||
|
case '-': case 'i': case 'm': case 's': case 'x': case 'v':
|
||
|
{
|
||
|
int neg = 0;
|
||
| ... | ... | |
|
case '-': neg = 1; break;
|
||
|
case 'x': ONOFF(option, ONIG_OPTION_EXTEND, neg); break;
|
||
|
case 'v': ONOFF(option, ONIG_OPTION_NEGATE, neg); break;
|
||
|
case 'i': ONOFF(option, ONIG_OPTION_IGNORECASE, neg); break;
|
||
|
case 's':
|
||
|
if (IS_SYNTAX_OP2(env->syntax, ONIG_SYN_OP2_OPTION_PERL)) {
|
||
| ... | ... | |
|
if (r < 0) return r;
|
||
|
*np = node_new_option(option);
|
||
|
CHECK_NULL_RETURN_MEMERR(*np);
|
||
|
/* expand "(?v:r)" into "(?:(?!r).)" */
|
||
|
if (IS_NEGATE(option)) {
|
||
|
Node *seq, *nla, *any;
|
||
|
/* build "(?!r)" */
|
||
|
nla = onig_node_new_anchor(ANCHOR_PREC_READ_NOT);
|
||
|
CHECK_NULL_RETURN_MEMERR(nla);
|
||
|
NANCHOR(nla)->target = target;
|
||
|
/* build "." */
|
||
|
any = node_new_anychar();
|
||
|
if (IS_NULL(any)) {
|
||
|
onig_node_free(nla);
|
||
|
return ONIGERR_MEMORY;
|
||
|
}
|
||
|
/* put "(?!r)" and "." in sequence: "(?!r)." */
|
||
|
seq = node_new_list(nla, NULL);
|
||
|
if (IS_NULL(seq)) {
|
||
|
onig_node_free(nla);
|
||
|
onig_node_free(any);
|
||
|
return ONIGERR_MEMORY;
|
||
|
}
|
||
|
NCDR(seq) = node_new_list(any, NULL);
|
||
|
if (IS_NULL(NCDR(seq))) {
|
||
|
onig_node_free(nla);
|
||
|
onig_node_free(any);
|
||
|
onig_node_free(seq);
|
||
|
return ONIGERR_MEMORY;
|
||
|
}
|
||
|
target = seq;
|
||
|
}
|
||
|
NENCLOSE(*np)->target = target;
|
||
|
*src = p;
|
||
|
return 0;
|
||
| test/ruby/test_regexp.rb | ||
|---|---|---|
|
end
|
||
|
def test_to_s
|
||
|
assert_equal '(?-mix:\x00)', Regexp.new("\0").to_s
|
||
|
assert_equal '(?-mixv:\x00)', Regexp.new("\0").to_s
|
||
|
end
|
||
|
def test_union
|
||
| ... | ... | |
|
end
|
||
|
def test_to_s2
|
||
|
assert_equal('(?-mix:foo)', /(?:foo)/.to_s)
|
||
|
assert_equal('(?m-ix:foo)', /(?:foo)/m.to_s)
|
||
|
assert_equal('(?mi-x:foo)', /(?:foo)/mi.to_s)
|
||
|
assert_equal('(?mix:foo)', /(?:foo)/mix.to_s)
|
||
|
assert_equal('(?m-ix:foo)', /(?m-ix:foo)/.to_s)
|
||
|
assert_equal('(?mi-x:foo)', /(?mi-x:foo)/.to_s)
|
||
|
assert_equal('(?mix:foo)', /(?mix:foo)/.to_s)
|
||
|
assert_equal('(?mix:)', /(?mix)/.to_s)
|
||
|
assert_equal('(?-mix:(?mix:foo) )', /(?mix:foo) /.to_s)
|
||
|
assert_equal('(?-mixv:foo)', /(?:foo)/.to_s)
|
||
|
assert_equal('(?m-ixv:foo)', /(?:foo)/m.to_s)
|
||
|
assert_equal('(?mi-xv:foo)', /(?:foo)/mi.to_s)
|
||
|
assert_equal('(?mix-v:foo)', /(?:foo)/mix.to_s)
|
||
|
assert_equal('(?mixv:foo)', /(?:foo)/mixv.to_s)
|
||
|
assert_equal('(?m-ixv:foo)', /(?m-ixv:foo)/.to_s)
|
||
|
assert_equal('(?mi-xv:foo)', /(?mi-xv:foo)/.to_s)
|
||
|
assert_equal('(?mix-v:foo)', /(?mix-v:foo)/.to_s)
|
||
|
assert_equal('(?mixv:foo)', /(?mixv:foo)/.to_s)
|
||
|
assert_equal('(?mixv:)', /(?mixv)/.to_s)
|
||
|
assert_equal('(?-mixv:(?mixv:foo) )', /(?mixv:foo) /.to_s)
|
||
|
end
|
||
|
def test_casefold_p
|
||
| ... | ... | |
|
def test_options
|
||
|
assert_equal(Regexp::IGNORECASE, /a/i.options)
|
||
|
assert_equal(Regexp::EXTENDED, /a/x.options)
|
||
|
assert_equal(Regexp::NEGATED, /a/v.options)
|
||
|
assert_equal(Regexp::MULTILINE, /a/m.options)
|
||
|
end
|
||
| ... | ... | |
|
assert_equal(/foo/, Regexp.union(/foo/))
|
||
|
assert_equal(/foo/, Regexp.union([/foo/]))
|
||
|
assert_equal(/\t/, Regexp.union("\t"))
|
||
|
assert_equal(/(?-mix:\u3042)|(?-mix:\u3042)/, Regexp.union(/\u3042/, /\u3042/))
|
||
|
assert_equal(/(?-mixv:\u3042)|(?-mixv:\u3042)/, Regexp.union(/\u3042/, /\u3042/))
|
||
|
assert_equal("\u3041", "\u3041"[Regexp.union(/\u3042/, "\u3041")])
|
||
|
end
|
||
| ... | ... | |
|
assert_match(/invalid hex escape/, error.message)
|
||
|
assert_equal(1, error.message.scan(/.*invalid .*escape.*/i).size, bug3539)
|
||
|
end
|
||
|
def test_negated_regexp_creation
|
||
|
assert_nothing_raised { eval("/ruby/v") }
|
||
|
assert_nothing_raised { eval("/(?v:ruby)/") }
|
||
|
assert_nothing_raised { eval("/(?-v:ruby)/") }
|
||
|
negated = Regexp.new("ruby", Regexp::NEGATED)
|
||
|
assert_equal(/ruby/v, negated)
|
||
|
assert_equal(/ruby/v, Regexp.new(negated))
|
||
|
assert_equal(/ruby/v, Regexp.new(Regexp.new(negated)))
|
||
|
assert_equal(/(?v-mix:ruby)/, Regexp.new(negated.to_s))
|
||
|
end
|
||
|
def test_negated_regexp_matching
|
||
|
assert_match(/ruby/, "ruby")
|
||
|
assert_match(/ruby/, "rubyperl")
|
||
|
assert_match(/ruby/, "perlruby")
|
||
|
assert_no_match(/ruby/, "perl")
|
||
|
assert_match(/(?-v:ruby)/, "ruby")
|
||
|
assert_match(/(?-v:ruby)/, "rubyperl")
|
||
|
assert_match(/(?-v:ruby)/, "perlruby")
|
||
|
assert_no_match(/(?-v:ruby)/, "perl")
|
||
|
assert_no_match(/ruby/v, "ruby")
|
||
|
assert_no_match(/ruby/v, "rubyperl")
|
||
|
assert_no_match(/ruby/v, "perlruby")
|
||
|
assert_match(/ruby/v, "perl")
|
||
|
assert_match(/(?v:ruby)/, "ruby")
|
||
|
assert_no_match(/^(?v:ruby)/, "ruby")
|
||
|
assert_match(/(?v:ruby)/, "rubyperl")
|
||
|
assert_no_match(/^(?v:ruby)/, "rubyperl")
|
||
|
assert_match(/(?v:ruby)/, "perlruby")
|
||
|
assert_match(/^(?v:ruby)/, "perlruby")
|
||
|
assert_match(/(?v:ruby)/, "perl")
|
||
|
assert_match(/^(?v:ruby)/, "perl")
|
||
|
assert_no_match(/a(?v:b)c/, "abc")
|
||
|
assert_no_match(/a(?v:b)c/, "ac")
|
||
|
assert_match(/a(?v:b)?c/, "ac")
|
||
|
assert_match(/a(?v:b)c/, "axc")
|
||
|
assert_no_match(/a(?v:b)c/, "axbc")
|
||
|
assert_no_match(/a(?v:b)+c/, "axbc")
|
||
|
assert_match(/a(?v:b)bc/, "axbc")
|
||
|
assert_no_match(/a(?v:b)c/, "ab_c")
|
||
|
assert_no_match(/a(?v:b)c/, "a_bc")
|
||
|
assert_match(/a(?v:b)bc/, "a_bc")
|
||
|
assert_no_match(/"[^<&"]*"/, '"aa<&a"')
|
||
|
assert_no_match(/"(?v:[<&"])*"/, '"aa<&a"')
|
||
|
assert_no_match(/"(?v:<|&|")*"/, '"aa<&a"')
|
||
|
languages = %w[ruby perl python lisp smalltalk]
|
||
|
assert_equal %w[perl lisp smalltalk], languages.grep(/l/)
|
||
|
assert_equal %w[ruby python], languages.grep(/l/v)
|
||
|
assert_equal %w[ruby perl python smalltalk], languages.grep(/^(?v:l)/)
|
||
|
end
|
||
|
end
|
||