/*
* call-seq:
* parser.parse(yaml)
*
* Parse the YAML document contained in +yaml+. Events will be called on
* the handler set on the parser instance.
*
* See Psych::Parser and Psych::Parser#handler
*/
static VALUE parse(int argc, VALUE *argv, VALUE self)
{
VALUE yaml, path;
yaml_parser_t * parser;
yaml_event_t event;
int done = 0;
int tainted = 0;
#ifdef HAVE_RUBY_ENCODING_H
int encoding = rb_utf8_encindex();
rb_encoding * internal_enc = rb_default_internal_encoding();
#endif
VALUE handler = rb_iv_get(self, "@handler");
if (rb_scan_args(argc, argv, "11", &yaml, &path) == 1) {
if(rb_respond_to(yaml, id_path))
path = rb_funcall(yaml, id_path, 0);
else
path = rb_str_new2("<unknown>");
}
Data_Get_Struct(self, yaml_parser_t, parser);
if (OBJ_TAINTED(yaml)) tainted = 1;
if(rb_respond_to(yaml, id_read)) {
yaml_parser_set_input(parser, io_reader, (void *)yaml);
if (RTEST(rb_obj_is_kind_of(yaml, rb_cIO))) tainted = 1;
} else {
StringValue(yaml);
yaml_parser_set_input_string(
parser,
(const unsigned char *)RSTRING_PTR(yaml),
(size_t)RSTRING_LEN(yaml)
);
}
while(!done) {
if(!yaml_parser_parse(parser, &event)) {
VALUE exception;
exception = make_exception(parser, path);
yaml_parser_delete(parser);
yaml_parser_initialize(parser);
rb_exc_raise(exception);
}
switch(event.type) {
case YAML_STREAM_START_EVENT:
rb_funcall(handler, id_start_stream, 1,
INT2NUM((long)event.data.stream_start.encoding)
);
break;
case YAML_DOCUMENT_START_EVENT:
{
/* Get a list of tag directives (if any) */
VALUE tag_directives = rb_ary_new();
/* Grab the document version */
VALUE version = event.data.document_start.version_directive ?
rb_ary_new3(
(long)2,
INT2NUM((long)event.data.document_start.version_directive->major),
INT2NUM((long)event.data.document_start.version_directive->minor)
) : rb_ary_new();
if(event.data.document_start.tag_directives.start) {
yaml_tag_directive_t *start =
event.data.document_start.tag_directives.start;
yaml_tag_directive_t *end =
event.data.document_start.tag_directives.end;
for(; start != end; start++) {
VALUE handle = Qnil;
VALUE prefix = Qnil;
if(start->handle) {
handle = rb_str_new2((const char *)start->handle);
if (tainted) OBJ_TAINT(handle);
#ifdef HAVE_RUBY_ENCODING_H
PSYCH_TRANSCODE(handle, encoding, internal_enc);
#endif
}
if(start->prefix) {
prefix = rb_str_new2((const char *)start->prefix);
if (tainted) OBJ_TAINT(prefix);
#ifdef HAVE_RUBY_ENCODING_H
PSYCH_TRANSCODE(prefix, encoding, internal_enc);
#endif
}
rb_ary_push(tag_directives, rb_ary_new3((long)2, handle, prefix));
}
}
rb_funcall(handler, id_start_document, 3,
version, tag_directives,
event.data.document_start.implicit == 1 ? Qtrue : Qfalse
);
}
break;
case YAML_DOCUMENT_END_EVENT:
rb_funcall(handler, id_end_document, 1,
event.data.document_end.implicit == 1 ? Qtrue : Qfalse
);
break;
case YAML_ALIAS_EVENT:
{
VALUE alias = Qnil;
if(event.data.alias.anchor) {
alias = rb_str_new2((const char *)event.data.alias.anchor);
if (tainted) OBJ_TAINT(alias);
#ifdef HAVE_RUBY_ENCODING_H
PSYCH_TRANSCODE(alias, encoding, internal_enc);
#endif
}
rb_funcall(handler, id_alias, 1, alias);
}
break;
case YAML_SCALAR_EVENT:
{
VALUE anchor = Qnil;
VALUE tag = Qnil;
VALUE plain_implicit, quoted_implicit, style;
VALUE val = rb_str_new(
(const char *)event.data.scalar.value,
(long)event.data.scalar.length
);
if (tainted) OBJ_TAINT(val);
#ifdef HAVE_RUBY_ENCODING_H
PSYCH_TRANSCODE(val, encoding, internal_enc);
#endif
if(event.data.scalar.anchor) {
anchor = rb_str_new2((const char *)event.data.scalar.anchor);
if (tainted) OBJ_TAINT(anchor);
#ifdef HAVE_RUBY_ENCODING_H
PSYCH_TRANSCODE(anchor, encoding, internal_enc);
#endif
}
if(event.data.scalar.tag) {
tag = rb_str_new2((const char *)event.data.scalar.tag);
if (tainted) OBJ_TAINT(tag);
#ifdef HAVE_RUBY_ENCODING_H
PSYCH_TRANSCODE(tag, encoding, internal_enc);
#endif
}
plain_implicit =
event.data.scalar.plain_implicit == 0 ? Qfalse : Qtrue;
quoted_implicit =
event.data.scalar.quoted_implicit == 0 ? Qfalse : Qtrue;
style = INT2NUM((long)event.data.scalar.style);
rb_funcall(handler, id_scalar, 6,
val, anchor, tag, plain_implicit, quoted_implicit, style);
}
break;
case YAML_SEQUENCE_START_EVENT:
{
VALUE anchor = Qnil;
VALUE tag = Qnil;
VALUE implicit, style;
if(event.data.sequence_start.anchor) {
anchor = rb_str_new2((const char *)event.data.sequence_start.anchor);
if (tainted) OBJ_TAINT(anchor);
#ifdef HAVE_RUBY_ENCODING_H
PSYCH_TRANSCODE(anchor, encoding, internal_enc);
#endif
}
tag = Qnil;
if(event.data.sequence_start.tag) {
tag = rb_str_new2((const char *)event.data.sequence_start.tag);
if (tainted) OBJ_TAINT(tag);
#ifdef HAVE_RUBY_ENCODING_H
PSYCH_TRANSCODE(tag, encoding, internal_enc);
#endif
}
implicit =
event.data.sequence_start.implicit == 0 ? Qfalse : Qtrue;
style = INT2NUM((long)event.data.sequence_start.style);
rb_funcall(handler, id_start_sequence, 4,
anchor, tag, implicit, style);
}
break;
case YAML_SEQUENCE_END_EVENT:
rb_funcall(handler, id_end_sequence, 0);
break;
case YAML_MAPPING_START_EVENT:
{
VALUE anchor = Qnil;
VALUE tag = Qnil;
VALUE implicit, style;
if(event.data.mapping_start.anchor) {
anchor = rb_str_new2((const char *)event.data.mapping_start.anchor);
if (tainted) OBJ_TAINT(anchor);
#ifdef HAVE_RUBY_ENCODING_H
PSYCH_TRANSCODE(anchor, encoding, internal_enc);
#endif
}
if(event.data.mapping_start.tag) {
tag = rb_str_new2((const char *)event.data.mapping_start.tag);
if (tainted) OBJ_TAINT(tag);
#ifdef HAVE_RUBY_ENCODING_H
PSYCH_TRANSCODE(tag, encoding, internal_enc);
#endif
}
implicit =
event.data.mapping_start.implicit == 0 ? Qfalse : Qtrue;
style = INT2NUM((long)event.data.mapping_start.style);
rb_funcall(handler, id_start_mapping, 4,
anchor, tag, implicit, style);
}
break;
case YAML_MAPPING_END_EVENT:
rb_funcall(handler, id_end_mapping, 0);
break;
case YAML_NO_EVENT:
rb_funcall(handler, id_empty, 0);
break;
case YAML_STREAM_END_EVENT:
rb_funcall(handler, id_end_stream, 0);
done = 1;
break;
}
yaml_event_delete(&event);
}
return self;
}
/*
* call-seq:
* parser.parse(yaml)
*
* Parse the YAML document contained in +yaml+. Events will be called on
* the handler set on the parser instance.
*
* See Psych::Parser and Psych::Parser#handler
*/
static VALUE parse(VALUE self, VALUE yaml)
{
yaml_parser_t parser;
yaml_event_t event;
int done = 0;
#ifdef HAVE_RUBY_ENCODING_H
int encoding = rb_enc_find_index("ASCII-8BIT");
rb_encoding * internal_enc;
#endif
VALUE handler = rb_iv_get(self, "@handler");
yaml_parser_initialize(&parser);
if(rb_respond_to(yaml, id_read)) {
yaml_parser_set_input(&parser, io_reader, (void *)yaml);
} else {
StringValue(yaml);
yaml_parser_set_input_string(
&parser,
(const unsigned char *)RSTRING_PTR(yaml),
(size_t)RSTRING_LEN(yaml)
);
}
while(!done) {
if(!yaml_parser_parse(&parser, &event)) {
size_t line = parser.mark.line;
size_t column = parser.mark.column;
yaml_parser_delete(&parser);
rb_raise(ePsychSyntaxError, "couldn't parse YAML at line %d column %d",
(int)line, (int)column);
}
switch(event.type) {
case YAML_STREAM_START_EVENT:
#ifdef HAVE_RUBY_ENCODING_H
switch(event.data.stream_start.encoding) {
case YAML_ANY_ENCODING:
break;
case YAML_UTF8_ENCODING:
encoding = rb_enc_find_index("UTF-8");
break;
case YAML_UTF16LE_ENCODING:
encoding = rb_enc_find_index("UTF-16LE");
break;
case YAML_UTF16BE_ENCODING:
encoding = rb_enc_find_index("UTF-16BE");
break;
default:
break;
}
internal_enc = rb_default_internal_encoding();
#endif
rb_funcall(handler, id_start_stream, 1,
INT2NUM((long)event.data.stream_start.encoding)
);
break;
case YAML_DOCUMENT_START_EVENT:
{
/* Get a list of tag directives (if any) */
VALUE tag_directives = rb_ary_new();
/* Grab the document version */
VALUE version = event.data.document_start.version_directive ?
rb_ary_new3(
(long)2,
INT2NUM((long)event.data.document_start.version_directive->major),
INT2NUM((long)event.data.document_start.version_directive->minor)
) : rb_ary_new();
if(event.data.document_start.tag_directives.start) {
yaml_tag_directive_t *start =
event.data.document_start.tag_directives.start;
yaml_tag_directive_t *end =
event.data.document_start.tag_directives.end;
for(; start != end; start++) {
VALUE handle = Qnil;
VALUE prefix = Qnil;
if(start->handle) {
handle = rb_str_new2((const char *)start->handle);
#ifdef HAVE_RUBY_ENCODING_H
PSYCH_TRANSCODE(handle, encoding, internal_enc);
#endif
}
if(start->prefix) {
prefix = rb_str_new2((const char *)start->prefix);
#ifdef HAVE_RUBY_ENCODING_H
PSYCH_TRANSCODE(prefix, encoding, internal_enc);
#endif
}
rb_ary_push(tag_directives, rb_ary_new3((long)2, handle, prefix));
}
}
rb_funcall(handler, id_start_document, 3,
version, tag_directives,
event.data.document_start.implicit == 1 ? Qtrue : Qfalse
);
}
break;
case YAML_DOCUMENT_END_EVENT:
rb_funcall(handler, id_end_document, 1,
event.data.document_end.implicit == 1 ? Qtrue : Qfalse
);
break;
case YAML_ALIAS_EVENT:
{
VALUE alias = Qnil;
if(event.data.alias.anchor) {
alias = rb_str_new2((const char *)event.data.alias.anchor);
#ifdef HAVE_RUBY_ENCODING_H
PSYCH_TRANSCODE(alias, encoding, internal_enc);
#endif
}
rb_funcall(handler, id_alias, 1, alias);
}
break;
case YAML_SCALAR_EVENT:
{
VALUE anchor = Qnil;
VALUE tag = Qnil;
VALUE plain_implicit, quoted_implicit, style;
VALUE val = rb_str_new(
(const char *)event.data.scalar.value,
(long)event.data.scalar.length
);
#ifdef HAVE_RUBY_ENCODING_H
PSYCH_TRANSCODE(val, encoding, internal_enc);
#endif
if(event.data.scalar.anchor) {
anchor = rb_str_new2((const char *)event.data.scalar.anchor);
#ifdef HAVE_RUBY_ENCODING_H
PSYCH_TRANSCODE(anchor, encoding, internal_enc);
#endif
}
if(event.data.scalar.tag) {
tag = rb_str_new2((const char *)event.data.scalar.tag);
#ifdef HAVE_RUBY_ENCODING_H
PSYCH_TRANSCODE(tag, encoding, internal_enc);
#endif
}
plain_implicit =
event.data.scalar.plain_implicit == 0 ? Qfalse : Qtrue;
quoted_implicit =
event.data.scalar.quoted_implicit == 0 ? Qfalse : Qtrue;
style = INT2NUM((long)event.data.scalar.style);
rb_funcall(handler, id_scalar, 6,
val, anchor, tag, plain_implicit, quoted_implicit, style);
}
break;
case YAML_SEQUENCE_START_EVENT:
{
VALUE anchor = Qnil;
VALUE tag = Qnil;
VALUE implicit, style;
if(event.data.sequence_start.anchor) {
anchor = rb_str_new2((const char *)event.data.sequence_start.anchor);
#ifdef HAVE_RUBY_ENCODING_H
PSYCH_TRANSCODE(anchor, encoding, internal_enc);
#endif
}
tag = Qnil;
if(event.data.sequence_start.tag) {
tag = rb_str_new2((const char *)event.data.sequence_start.tag);
#ifdef HAVE_RUBY_ENCODING_H
PSYCH_TRANSCODE(tag, encoding, internal_enc);
#endif
}
implicit =
event.data.sequence_start.implicit == 0 ? Qfalse : Qtrue;
style = INT2NUM((long)event.data.sequence_start.style);
rb_funcall(handler, id_start_sequence, 4,
anchor, tag, implicit, style);
}
break;
case YAML_SEQUENCE_END_EVENT:
rb_funcall(handler, id_end_sequence, 0);
break;
case YAML_MAPPING_START_EVENT:
{
VALUE anchor = Qnil;
VALUE tag = Qnil;
VALUE implicit, style;
if(event.data.mapping_start.anchor) {
anchor = rb_str_new2((const char *)event.data.mapping_start.anchor);
#ifdef HAVE_RUBY_ENCODING_H
PSYCH_TRANSCODE(anchor, encoding, internal_enc);
#endif
}
if(event.data.mapping_start.tag) {
tag = rb_str_new2((const char *)event.data.mapping_start.tag);
#ifdef HAVE_RUBY_ENCODING_H
PSYCH_TRANSCODE(tag, encoding, internal_enc);
#endif
}
implicit =
event.data.mapping_start.implicit == 0 ? Qfalse : Qtrue;
style = INT2NUM((long)event.data.mapping_start.style);
rb_funcall(handler, id_start_mapping, 4,
anchor, tag, implicit, style);
}
break;
case YAML_MAPPING_END_EVENT:
rb_funcall(handler, id_end_mapping, 0);
break;
case YAML_NO_EVENT:
rb_funcall(handler, id_empty, 0);
break;
case YAML_STREAM_END_EVENT:
rb_funcall(handler, id_end_stream, 0);
done = 1;
break;
}
}
return self;
}
/*
* call-seq:
* parser.parse(yaml)
*
* Parse the YAML document contained in +yaml+. Events will be called on
* the handler set on the parser instance.
*
* See Psych::Parser and Psych::Parser#handler
*/
static VALUE parse(int argc, VALUE *argv, VALUE self)
{
VALUE yaml, path;
yaml_parser_t * parser;
yaml_event_t event;
int done = 0;
int tainted = 0;
int state = 0;
int parser_encoding = YAML_ANY_ENCODING;
int encoding = rb_utf8_encindex();
rb_encoding * internal_enc = rb_default_internal_encoding();
VALUE handler = rb_iv_get(self, "@handler");
if (rb_scan_args(argc, argv, "11", &yaml, &path) == 1) {
if(rb_respond_to(yaml, id_path))
path = rb_funcall(yaml, id_path, 0);
else
path = rb_str_new2("<unknown>");
}
TypedData_Get_Struct(self, yaml_parser_t, &psych_parser_type, parser);
yaml_parser_delete(parser);
yaml_parser_initialize(parser);
if (OBJ_TAINTED(yaml)) tainted = 1;
if (rb_respond_to(yaml, id_read)) {
yaml = transcode_io(yaml, &parser_encoding);
yaml_parser_set_encoding(parser, parser_encoding);
yaml_parser_set_input(parser, io_reader, (void *)yaml);
if (RTEST(rb_obj_is_kind_of(yaml, rb_cIO))) tainted = 1;
} else {
StringValue(yaml);
yaml = transcode_string(yaml, &parser_encoding);
yaml_parser_set_encoding(parser, parser_encoding);
yaml_parser_set_input_string(
parser,
(const unsigned char *)RSTRING_PTR(yaml),
(size_t)RSTRING_LEN(yaml)
);
}
while(!done) {
VALUE event_args[5];
VALUE start_line, start_column, end_line, end_column;
if(!yaml_parser_parse(parser, &event)) {
VALUE exception;
exception = make_exception(parser, path);
yaml_parser_delete(parser);
yaml_parser_initialize(parser);
rb_exc_raise(exception);
}
start_line = INT2NUM((long)event.start_mark.line);
start_column = INT2NUM((long)event.start_mark.column);
end_line = INT2NUM((long)event.end_mark.line);
end_column = INT2NUM((long)event.end_mark.column);
event_args[0] = handler;
event_args[1] = start_line;
event_args[2] = start_column;
event_args[3] = end_line;
event_args[4] = end_column;
rb_protect(protected_event_location, (VALUE)event_args, &state);
switch(event.type) {
case YAML_STREAM_START_EVENT:
{
VALUE args[2];
args[0] = handler;
args[1] = INT2NUM((long)event.data.stream_start.encoding);
rb_protect(protected_start_stream, (VALUE)args, &state);
}
break;
case YAML_DOCUMENT_START_EVENT:
{
VALUE args[4];
/* Get a list of tag directives (if any) */
VALUE tag_directives = rb_ary_new();
/* Grab the document version */
VALUE version = event.data.document_start.version_directive ?
rb_ary_new3(
(long)2,
INT2NUM((long)event.data.document_start.version_directive->major),
INT2NUM((long)event.data.document_start.version_directive->minor)
) : rb_ary_new();
if(event.data.document_start.tag_directives.start) {
yaml_tag_directive_t *start =
event.data.document_start.tag_directives.start;
yaml_tag_directive_t *end =
event.data.document_start.tag_directives.end;
for(; start != end; start++) {
VALUE handle = Qnil;
VALUE prefix = Qnil;
if(start->handle) {
handle = rb_str_new2((const char *)start->handle);
if (tainted) OBJ_TAINT(handle);
PSYCH_TRANSCODE(handle, encoding, internal_enc);
}
if(start->prefix) {
prefix = rb_str_new2((const char *)start->prefix);
if (tainted) OBJ_TAINT(prefix);
PSYCH_TRANSCODE(prefix, encoding, internal_enc);
}
rb_ary_push(tag_directives, rb_ary_new3((long)2, handle, prefix));
}
}
args[0] = handler;
args[1] = version;
args[2] = tag_directives;
args[3] = event.data.document_start.implicit == 1 ? Qtrue : Qfalse;
rb_protect(protected_start_document, (VALUE)args, &state);
}
break;
case YAML_DOCUMENT_END_EVENT:
{
VALUE args[2];
args[0] = handler;
args[1] = event.data.document_end.implicit == 1 ? Qtrue : Qfalse;
rb_protect(protected_end_document, (VALUE)args, &state);
}
break;
case YAML_ALIAS_EVENT:
{
VALUE args[2];
VALUE alias = Qnil;
if(event.data.alias.anchor) {
alias = rb_str_new2((const char *)event.data.alias.anchor);
if (tainted) OBJ_TAINT(alias);
PSYCH_TRANSCODE(alias, encoding, internal_enc);
}
args[0] = handler;
args[1] = alias;
rb_protect(protected_alias, (VALUE)args, &state);
}
break;
case YAML_SCALAR_EVENT:
{
VALUE args[7];
VALUE anchor = Qnil;
VALUE tag = Qnil;
VALUE plain_implicit, quoted_implicit, style;
VALUE val = rb_str_new(
(const char *)event.data.scalar.value,
(long)event.data.scalar.length
);
if (tainted) OBJ_TAINT(val);
PSYCH_TRANSCODE(val, encoding, internal_enc);
if(event.data.scalar.anchor) {
anchor = rb_str_new2((const char *)event.data.scalar.anchor);
if (tainted) OBJ_TAINT(anchor);
PSYCH_TRANSCODE(anchor, encoding, internal_enc);
}
if(event.data.scalar.tag) {
tag = rb_str_new2((const char *)event.data.scalar.tag);
if (tainted) OBJ_TAINT(tag);
PSYCH_TRANSCODE(tag, encoding, internal_enc);
}
plain_implicit =
event.data.scalar.plain_implicit == 0 ? Qfalse : Qtrue;
quoted_implicit =
event.data.scalar.quoted_implicit == 0 ? Qfalse : Qtrue;
style = INT2NUM((long)event.data.scalar.style);
args[0] = handler;
args[1] = val;
args[2] = anchor;
args[3] = tag;
args[4] = plain_implicit;
args[5] = quoted_implicit;
args[6] = style;
rb_protect(protected_scalar, (VALUE)args, &state);
}
break;
case YAML_SEQUENCE_START_EVENT:
{
VALUE args[5];
VALUE anchor = Qnil;
VALUE tag = Qnil;
VALUE implicit, style;
if(event.data.sequence_start.anchor) {
anchor = rb_str_new2((const char *)event.data.sequence_start.anchor);
if (tainted) OBJ_TAINT(anchor);
PSYCH_TRANSCODE(anchor, encoding, internal_enc);
}
tag = Qnil;
if(event.data.sequence_start.tag) {
tag = rb_str_new2((const char *)event.data.sequence_start.tag);
if (tainted) OBJ_TAINT(tag);
PSYCH_TRANSCODE(tag, encoding, internal_enc);
}
implicit =
event.data.sequence_start.implicit == 0 ? Qfalse : Qtrue;
style = INT2NUM((long)event.data.sequence_start.style);
args[0] = handler;
args[1] = anchor;
args[2] = tag;
args[3] = implicit;
args[4] = style;
rb_protect(protected_start_sequence, (VALUE)args, &state);
}
break;
case YAML_SEQUENCE_END_EVENT:
rb_protect(protected_end_sequence, handler, &state);
break;
case YAML_MAPPING_START_EVENT:
{
VALUE args[5];
VALUE anchor = Qnil;
VALUE tag = Qnil;
VALUE implicit, style;
if(event.data.mapping_start.anchor) {
anchor = rb_str_new2((const char *)event.data.mapping_start.anchor);
if (tainted) OBJ_TAINT(anchor);
PSYCH_TRANSCODE(anchor, encoding, internal_enc);
}
if(event.data.mapping_start.tag) {
tag = rb_str_new2((const char *)event.data.mapping_start.tag);
if (tainted) OBJ_TAINT(tag);
PSYCH_TRANSCODE(tag, encoding, internal_enc);
}
implicit =
event.data.mapping_start.implicit == 0 ? Qfalse : Qtrue;
style = INT2NUM((long)event.data.mapping_start.style);
args[0] = handler;
args[1] = anchor;
args[2] = tag;
args[3] = implicit;
args[4] = style;
rb_protect(protected_start_mapping, (VALUE)args, &state);
}
break;
case YAML_MAPPING_END_EVENT:
rb_protect(protected_end_mapping, handler, &state);
break;
case YAML_NO_EVENT:
rb_protect(protected_empty, handler, &state);
break;
case YAML_STREAM_END_EVENT:
rb_protect(protected_end_stream, handler, &state);
done = 1;
break;
}
yaml_event_delete(&event);
if (state) rb_jump_tag(state);
}
return self;
}
