int rb_comparator_func(void * ctx, int a_len, const void * a, int b_len, const void * b)
{
VALUE comparator;
VALUE a_str;
VALUE b_str;
VALUE comparison;
#ifdef HAVE_RUBY_ENCODING_H
rb_encoding * internal_encoding;
internal_encoding = rb_default_internal_encoding();
#endif
comparator = (VALUE)ctx;
a_str = rb_str_new((const char *)a, a_len);
b_str = rb_str_new((const char *)b, b_len);
#ifdef HAVE_RUBY_ENCODING_H
rb_enc_associate_index(a_str, rb_utf8_encindex());
rb_enc_associate_index(b_str, rb_utf8_encindex());
if(internal_encoding) {
a_str = rb_str_export_to_enc(a_str, internal_encoding);
b_str = rb_str_export_to_enc(b_str, internal_encoding);
}
#endif
comparison = rb_funcall(comparator, rb_intern("compare"), 2, a_str, b_str);
return NUM2INT(comparison);
}
static VALUE stringForceUTF8(VALUE arg)
{
if (rb_type(arg) == RUBY_T_STRING && ENCODING_IS_ASCII8BIT(arg))
rb_enc_associate_index(arg, rb_utf8_encindex());
return arg;
}
static VALUE stringForceUTF8(VALUE arg)
{
if (RB_TYPE_P(arg, RUBY_T_STRING) && ENCODING_IS_ASCII8BIT(arg))
rb_enc_associate_index(arg, rb_utf8_encindex());
return arg;
}
static VALUE transcode_string(VALUE src, int * parser_encoding)
{
int utf8 = rb_utf8_encindex();
int utf16le = rb_enc_find_index("UTF16_LE");
int utf16be = rb_enc_find_index("UTF16_BE");
int source_encoding = rb_enc_get_index(src);
if (source_encoding == utf8) {
*parser_encoding = YAML_UTF8_ENCODING;
return src;
}
if (source_encoding == utf16le) {
*parser_encoding = YAML_UTF16LE_ENCODING;
return src;
}
if (source_encoding == utf16be) {
*parser_encoding = YAML_UTF16BE_ENCODING;
return src;
}
src = rb_str_export_to_enc(src, rb_utf8_encoding());
RB_GC_GUARD(src);
*parser_encoding = YAML_UTF8_ENCODING;
return src;
}
/**
* Document-module: MessagePack
*
* MessagePack is a binary-based efficient object serialization library.
* It enables to exchange structured objects between many languages like JSON.
* But unlike JSON, it is very fast and small.
*
* You can install MessagePack with rubygems.
*
* gem install msgpack
*
* Simple usage is as follows:
*
* require 'msgpack'
* msg = [1,2,3].to_msgpack #=> "\x93\x01\x02\x03"
* MessagePack.unpack(msg) #=> [1,2,3]
*
* Use Unpacker class for streaming deserialization.
*
*/
void Init_msgpack(void)
{
mMessagePack = rb_define_module("MessagePack");
rb_define_const(mMessagePack, "VERSION", rb_str_new2(MESSAGEPACK_VERSION));
#ifdef COMPAT_HAVE_ENCODING
s_enc_ascii8bit = rb_ascii8bit_encindex();
s_enc_utf8 = rb_utf8_encindex();
s_enc_usascii = rb_usascii_encindex();
s_enc_utf8_value = rb_enc_from_encoding(rb_utf8_encoding());
#endif
Init_msgpack_unpack(mMessagePack);
Init_msgpack_pack(mMessagePack);
}
static VALUE transcode_io(VALUE src, int * parser_encoding)
{
VALUE io_external_encoding;
int io_external_enc_index;
io_external_encoding = rb_funcall(src, rb_intern("external_encoding"), 0);
/* if no encoding is returned, assume ascii8bit. */
if (NIL_P(io_external_encoding)) {
io_external_enc_index = rb_ascii8bit_encindex();
} else {
io_external_enc_index = rb_to_encoding_index(io_external_encoding);
}
/* Treat US-ASCII as utf_8 */
if (io_external_enc_index == rb_usascii_encindex()) {
*parser_encoding = YAML_UTF8_ENCODING;
return src;
}
if (io_external_enc_index == rb_utf8_encindex()) {
*parser_encoding = YAML_UTF8_ENCODING;
return src;
}
if (io_external_enc_index == rb_enc_find_index("UTF-16LE")) {
*parser_encoding = YAML_UTF16LE_ENCODING;
return src;
}
if (io_external_enc_index == rb_enc_find_index("UTF-16BE")) {
*parser_encoding = YAML_UTF16BE_ENCODING;
return src;
}
/* Just guess on ASCII-8BIT */
if (io_external_enc_index == rb_ascii8bit_encindex()) {
*parser_encoding = YAML_ANY_ENCODING;
return src;
}
/* If the external encoding is something we don't know how to handle,
* fall back to YAML_ANY_ENCODING. */
*parser_encoding = YAML_ANY_ENCODING;
return src;
}
static VALUE transcode_io(VALUE src, int * parser_encoding)
{
VALUE io_external_encoding;
int io_external_enc_index;
io_external_encoding = rb_funcall(src, rb_intern("external_encoding"), 0);
/* if no encoding is returned, assume ascii8bit. */
if (NIL_P(io_external_encoding)) {
io_external_enc_index = rb_ascii8bit_encindex();
} else {
io_external_enc_index = rb_to_encoding_index(io_external_encoding);
}
/* Treat US-ASCII as utf_8 */
if (io_external_enc_index == rb_usascii_encindex()) {
*parser_encoding = YAML_UTF8_ENCODING;
return src;
}
if (io_external_enc_index == rb_utf8_encindex()) {
*parser_encoding = YAML_UTF8_ENCODING;
return src;
}
if (io_external_enc_index == rb_enc_find_index("UTF-16LE")) {
*parser_encoding = YAML_UTF16LE_ENCODING;
return src;
}
if (io_external_enc_index == rb_enc_find_index("UTF-16BE")) {
*parser_encoding = YAML_UTF16BE_ENCODING;
return src;
}
/* Just guess on ASCII-8BIT */
if (io_external_enc_index == rb_ascii8bit_encindex()) {
*parser_encoding = YAML_ANY_ENCODING;
return src;
}
rb_raise(rb_eArgError, "YAML file must be UTF-8, UTF-16LE, or UTF-16BE, not %s",
rb_enc_name(rb_enc_from_index(io_external_enc_index)));
return Qnil;
}
static int
rb_filesystem_encindex(void)
{
int idx;
#if defined NO_LOCALE_CHARMAP
idx = rb_enc_to_index(rb_default_external_encoding());
#elif defined _WIN32 || defined __CYGWIN__
char cp[sizeof(int) * 8 / 3 + 4];
snprintf(cp, sizeof cp, "CP%d", AreFileApisANSI() ? GetACP() : GetOEMCP());
idx = rb_enc_find_index(cp);
if (idx < 0) idx = rb_ascii8bit_encindex();
#elif defined __APPLE__
idx = rb_utf8_encindex();
#else
idx = rb_locale_encindex();
#endif
if (rb_enc_registered("filesystem") < 0) enc_alias_internal("filesystem", idx);
return idx;
}
/*
Return code page number of the encoding.
Cache code page into a hash for performance since finding the code page in
Encoding#names is slow.
*/
static UINT
code_page(rb_encoding *enc)
{
int enc_idx;
if (!enc)
return system_code_page();
enc_idx = rb_enc_to_index(enc);
/* map US-ASCII and ASCII-8bit as code page 1252 (us-ascii) */
if (enc_idx == rb_usascii_encindex() || enc_idx == rb_ascii8bit_encindex()) {
return 1252;
}
if (enc_idx == rb_utf8_encindex()) {
return CP_UTF8;
}
if (0 <= enc_idx && (unsigned int)enc_idx < rb_code_page.count)
return rb_code_page.table[enc_idx];
return INVALID_CODE_PAGE;
}
static VALUE encoding_spec_rb_utf8_encindex(VALUE self) {
return INT2NUM(rb_utf8_encindex());
}
static VALUE step(VALUE self)
{
sqlite3StmtRubyPtr ctx;
sqlite3_stmt *stmt;
int value, length;
VALUE list;
rb_encoding * internal_encoding;
Data_Get_Struct(self, sqlite3StmtRuby, ctx);
REQUIRE_OPEN_STMT(ctx);
if(ctx->done_p) return Qnil;
{
VALUE db = rb_iv_get(self, "@connection");
rb_funcall(db, rb_intern("encoding"), 0);
internal_encoding = rb_default_internal_encoding();
}
stmt = ctx->st;
value = sqlite3_step(stmt);
length = sqlite3_column_count(stmt);
list = rb_ary_new2((long)length);
switch(value) {
case SQLITE_ROW:
{
int i;
for(i = 0; i < length; i++) {
switch(sqlite3_column_type(stmt, i)) {
case SQLITE_INTEGER:
rb_ary_push(list, LL2NUM(sqlite3_column_int64(stmt, i)));
break;
case SQLITE_FLOAT:
rb_ary_push(list, rb_float_new(sqlite3_column_double(stmt, i)));
break;
case SQLITE_TEXT:
{
VALUE str = rb_tainted_str_new(
(const char *)sqlite3_column_text(stmt, i),
(long)sqlite3_column_bytes(stmt, i)
);
rb_enc_associate_index(str, rb_utf8_encindex());
if(internal_encoding)
str = rb_str_export_to_enc(str, internal_encoding);
rb_ary_push(list, str);
}
break;
case SQLITE_BLOB:
{
VALUE str = rb_tainted_str_new(
(const char *)sqlite3_column_blob(stmt, i),
(long)sqlite3_column_bytes(stmt, i)
);
rb_ary_push(list, str);
}
break;
case SQLITE_NULL:
rb_ary_push(list, Qnil);
break;
default:
rb_raise(rb_eRuntimeError, "bad type");
}
}
}
break;
case SQLITE_DONE:
ctx->done_p = 1;
return Qnil;
break;
default:
sqlite3_reset(stmt);
ctx->done_p = 0;
CHECK(sqlite3_db_handle(ctx->st), value);
}
return list;
}
VALUE require_compiled(VALUE fname, VALUE* result, int bLoad)
{
VALUE path;
char* szName1 = 0;
VALUE retval = Qtrue;
if (TYPE(fname) != T_STRING)
rb_raise(rb_eLoadError, "can not load non-string");
szName1 = RSTRING_PTR(fname);
if ( String_endsWith(szName1,".rb") )
{
rb_str_chop_bang(fname); rb_str_chop_bang(fname); rb_str_chop_bang(fname);
}
//rb_funcall(fname, rb_intern("sub!"), 2, rb_str_new2(".rb"), rb_str_new2("") );
szName1 = RSTRING_PTR(fname);
if ( strcmp("strscan",szName1)==0 || strcmp("enumerator",szName1)==0 ||
strcmp("stringio",szName1)==0 || strcmp("socket",szName1)==0 )
return Qtrue;
RHO_LOCK(require_lock);
if ( !bLoad && isAlreadyLoaded(fname) == Qtrue )
goto RCompExit;
path = find_file(fname);
if ( path != 0 )
{
VALUE seq;
RAWLOG_INFO1("require_compiled: %s", szName1);
//optimize require
//rb_ary_push(GET_VM()->loaded_features, path);
rb_ary_push(GET_VM()->loaded_features, fname);
#ifdef RHODES_EMULATOR
if ( strstr( RSTRING_PTR(path), ".rb") == 0 )
rb_str_cat(path,".rb",3);
GET_VM()->src_encoding_index = rb_utf8_encindex();
rb_load(path, 0);
if( rho_simconf_getBool("reload_app_changes") )
{
if ( strncmp( RSTRING_PTR(path), rho_native_rhopath(), strlen(rho_native_rhopath()) ) == 0 )
rb_ary_delete(GET_VM()->loaded_features, fname);
}
#else
//rb_gc_disable();
seq = loadISeqFromFile(path);
//*result = rb_funcall(seq, rb_intern("eval"), 0 );
*result = rb_iseq_eval(seq);
//rb_gc_enable();
#endif
goto RCompExit;
}
RAWLOG_ERROR1("require_compiled: error: can not find %s", RSTRING_PTR(fname));
retval = Qnil;
RCompExit:
RHO_UNLOCK(require_lock);
return retval;
}
/*
* 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(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;
}
/**
* @param str the string to be scrubbed
* @param repl the replacement character
* @return If given string is invalid, returns a new string. Otherwise, returns Qnil.
*/
static VALUE
str_scrub0(int argc, VALUE *argv, VALUE str)
{
int cr = ENC_CODERANGE(str);
rb_encoding *enc;
int encidx;
VALUE repl;
if (cr == ENC_CODERANGE_7BIT || cr == ENC_CODERANGE_VALID)
return Qnil;
enc = STR_ENC_GET(str);
rb_scan_args(argc, argv, "01", &repl);
if (argc != 0) {
repl = str_compat_and_valid(repl, enc);
}
if (rb_enc_dummy_p(enc)) {
return Qnil;
}
encidx = rb_enc_to_index(enc);
#define DEFAULT_REPLACE_CHAR(str) do { \
static const char replace[sizeof(str)-1] = str; \
rep = replace; replen = (int)sizeof(replace); \
} while (0)
if (rb_enc_asciicompat(enc)) {
const char *p = RSTRING_PTR(str);
const char *e = RSTRING_END(str);
const char *p1 = p;
const char *rep;
long replen;
int rep7bit_p;
VALUE buf = Qnil;
if (rb_block_given_p()) {
rep = NULL;
replen = 0;
rep7bit_p = FALSE;
}
else if (!NIL_P(repl)) {
rep = RSTRING_PTR(repl);
replen = RSTRING_LEN(repl);
rep7bit_p = (ENC_CODERANGE(repl) == ENC_CODERANGE_7BIT);
}
else if (encidx == rb_utf8_encindex()) {
DEFAULT_REPLACE_CHAR("\xEF\xBF\xBD");
rep7bit_p = FALSE;
}
else {
DEFAULT_REPLACE_CHAR("?");
rep7bit_p = TRUE;
}
cr = ENC_CODERANGE_7BIT;
p = search_nonascii(p, e);
if (!p) {
p = e;
}
while (p < e) {
int ret = rb_enc_precise_mbclen(p, e, enc);
if (MBCLEN_NEEDMORE_P(ret)) {
break;
}
else if (MBCLEN_CHARFOUND_P(ret)) {
cr = ENC_CODERANGE_VALID;
p += MBCLEN_CHARFOUND_LEN(ret);
}
else if (MBCLEN_INVALID_P(ret)) {
/*
* p1~p: valid ascii/multibyte chars
* p ~e: invalid bytes + unknown bytes
*/
long clen = rb_enc_mbmaxlen(enc);
if (NIL_P(buf)) buf = rb_str_buf_new(RSTRING_LEN(str));
if (p > p1) {
rb_str_buf_cat(buf, p1, p - p1);
}
if (e - p < clen) clen = e - p;
if (clen <= 2) {
clen = 1;
}
else {
const char *q = p;
clen--;
for (; clen > 1; clen--) {
ret = rb_enc_precise_mbclen(q, q + clen, enc);
if (MBCLEN_NEEDMORE_P(ret)) break;
if (MBCLEN_INVALID_P(ret)) continue;
UNREACHABLE;
}
}
if (rep) {
rb_str_buf_cat(buf, rep, replen);
if (!rep7bit_p) cr = ENC_CODERANGE_VALID;
}
else {
repl = rb_yield(rb_enc_str_new(p, clen, enc));
repl = str_compat_and_valid(repl, enc);
rb_str_buf_cat(buf, RSTRING_PTR(repl), RSTRING_LEN(repl));
if (ENC_CODERANGE(repl) == ENC_CODERANGE_VALID)
cr = ENC_CODERANGE_VALID;
}
p += clen;
p1 = p;
p = search_nonascii(p, e);
if (!p) {
p = e;
break;
}
}
else {
UNREACHABLE;
}
}
if (NIL_P(buf)) {
if (p == e) {
ENC_CODERANGE_SET(str, cr);
return Qnil;
}
buf = rb_str_buf_new(RSTRING_LEN(str));
}
if (p1 < p) {
rb_str_buf_cat(buf, p1, p - p1);
}
if (p < e) {
if (rep) {
rb_str_buf_cat(buf, rep, replen);
if (!rep7bit_p) cr = ENC_CODERANGE_VALID;
}
else {
repl = rb_yield(rb_enc_str_new(p, e-p, enc));
repl = str_compat_and_valid(repl, enc);
rb_str_buf_cat(buf, RSTRING_PTR(repl), RSTRING_LEN(repl));
if (ENC_CODERANGE(repl) == ENC_CODERANGE_VALID)
cr = ENC_CODERANGE_VALID;
}
}
ENCODING_CODERANGE_SET(buf, rb_enc_to_index(enc), cr);
return buf;
}
else {
/* ASCII incompatible */
const char *p = RSTRING_PTR(str);
const char *e = RSTRING_END(str);
const char *p1 = p;
VALUE buf = Qnil;
const char *rep;
long replen;
long mbminlen = rb_enc_mbminlen(enc);
if (!NIL_P(repl)) {
rep = RSTRING_PTR(repl);
replen = RSTRING_LEN(repl);
}
else if (!strcasecmp(rb_enc_name(enc), "UTF-16BE")) {
DEFAULT_REPLACE_CHAR("\xFF\xFD");
}
else if (!strcasecmp(rb_enc_name(enc), "UTF-16LE")) {
DEFAULT_REPLACE_CHAR("\xFD\xFF");
}
else if (!strcasecmp(rb_enc_name(enc), "UTF-32BE")) {
DEFAULT_REPLACE_CHAR("\x00\x00\xFF\xFD");
}
else if (!strcasecmp(rb_enc_name(enc), "UTF-32lE")) {
DEFAULT_REPLACE_CHAR("\xFD\xFF\x00\x00");
}
else {
DEFAULT_REPLACE_CHAR("?");
}
while (p < e) {
int ret = rb_enc_precise_mbclen(p, e, enc);
if (MBCLEN_NEEDMORE_P(ret)) {
break;
}
else if (MBCLEN_CHARFOUND_P(ret)) {
p += MBCLEN_CHARFOUND_LEN(ret);
}
else if (MBCLEN_INVALID_P(ret)) {
const char *q = p;
long clen = rb_enc_mbmaxlen(enc);
if (NIL_P(buf)) buf = rb_str_buf_new(RSTRING_LEN(str));
if (p > p1) rb_str_buf_cat(buf, p1, p - p1);
if (e - p < clen) clen = e - p;
if (clen <= mbminlen * 2) {
clen = mbminlen;
}
else {
clen -= mbminlen;
for (; clen > mbminlen; clen-=mbminlen) {
ret = rb_enc_precise_mbclen(q, q + clen, enc);
if (MBCLEN_NEEDMORE_P(ret)) break;
if (MBCLEN_INVALID_P(ret)) continue;
UNREACHABLE;
}
}
if (rep) {
rb_str_buf_cat(buf, rep, replen);
}
else {
repl = rb_yield(rb_enc_str_new(p, e-p, enc));
repl = str_compat_and_valid(repl, enc);
rb_str_buf_cat(buf, RSTRING_PTR(repl), RSTRING_LEN(repl));
}
p += clen;
p1 = p;
}
else {
UNREACHABLE;
}
}
if (NIL_P(buf)) {
if (p == e) {
ENC_CODERANGE_SET(str, ENC_CODERANGE_VALID);
return Qnil;
}
buf = rb_str_buf_new(RSTRING_LEN(str));
}
if (p1 < p) {
rb_str_buf_cat(buf, p1, p - p1);
}
if (p < e) {
if (rep) {
rb_str_buf_cat(buf, rep, replen);
}
else {
repl = rb_yield(rb_enc_str_new(p, e-p, enc));
repl = str_compat_and_valid(repl, enc);
rb_str_buf_cat(buf, RSTRING_PTR(repl), RSTRING_LEN(repl));
}
}
ENCODING_CODERANGE_SET(buf, rb_enc_to_index(enc), ENC_CODERANGE_VALID);
return buf;
}
}
/*
* 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_utf8_encindex();
rb_encoding * internal_enc = rb_default_internal_encoding();
#endif
VALUE handler = rb_iv_get(self, "@handler");
Data_Get_Struct(self, yaml_parser_t, 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;
rb_raise(ePsychSyntaxError, "couldn't parse YAML at line %d column %d",
(int)line, (int)column);
}
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);
#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;
}
