/*
* call-seq:
* typemap.typecast_copy_get( field_str, fieldno, format, encoding )
*
* Cast a field string received by PG::Connection#get_copy_data.
*
* This method implementation uses the #default_type_map to cast field_str.
* It can be derived to change this behaviour.
*
* Parameters:
* * +field_str+ : The String received from the server.
* * +fieldno+ : The field number from left to right.
* * +format+ : The format code (0 = text, 1 = binary)
* * +encoding+ : The encoding of the connection and encoding the returned
* value should get.
*
*/
static VALUE
pg_tmir_typecast_copy_get( VALUE self, VALUE field_str, VALUE fieldno, VALUE format, VALUE enc )
{
t_tmir *this = DATA_PTR( self );
t_typemap *default_tm = DATA_PTR( this->typemap.default_typemap );
int enc_idx = rb_to_encoding_index( enc );
return default_tm->funcs.typecast_copy_get( default_tm, field_str, NUM2INT(fieldno), NUM2INT(format), enc_idx );
}
/*
* call-seq:
* enc.names -> array
*
* Returns the list of name and aliases of the encoding.
*
* Encoding::WINDOWS_31J.names #=> ["Windows-31J", "CP932", "csWindows31J"]
*/
static VALUE
enc_names(VALUE self)
{
VALUE args[2];
args[0] = (VALUE)rb_to_encoding_index(self);
args[1] = rb_ary_new2(0);
st_foreach(enc_table.names, enc_names_i, (st_data_t)args);
return args[1];
}
/*
* call-seq:
* coder.encode( value [, encoding] )
*
* Encodes the given Ruby object into string representation, without
* sending data to/from the database server.
*
* A nil value is passed through.
*
*/
static VALUE
pg_coder_encode(int argc, VALUE *argv, VALUE self)
{
VALUE res;
VALUE intermediate;
VALUE value;
int len, len2;
int enc_idx;
t_pg_coder *this = DATA_PTR(self);
if(argc < 1 || argc > 2){
rb_raise(rb_eArgError, "wrong number of arguments (%i for 1..2)", argc);
}else if(argc == 1){
enc_idx = rb_ascii8bit_encindex();
}else{
enc_idx = rb_to_encoding_index(argv[1]);
}
value = argv[0];
if( NIL_P(value) )
return Qnil;
if( !this->enc_func ){
rb_raise(rb_eRuntimeError, "no encoder function defined");
}
len = this->enc_func( this, value, NULL, &intermediate, enc_idx );
if( len == -1 ){
/* The intermediate value is a String that can be used directly. */
OBJ_INFECT(intermediate, value);
return intermediate;
}
res = rb_str_new(NULL, len);
PG_ENCODING_SET_NOCHECK(res, enc_idx);
len2 = this->enc_func( this, value, RSTRING_PTR(res), &intermediate, enc_idx );
if( len < len2 ){
rb_bug("%s: result length of first encoder run (%i) is less than second run (%i)",
rb_obj_classname( self ), len, len2 );
}
rb_str_set_len( res, len2 );
OBJ_INFECT(res, value);
RB_GC_GUARD(intermediate);
return res;
}
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 VALUE encoding_spec_rb_to_encoding_index(VALUE self, VALUE obj) {
return INT2NUM(rb_to_encoding_index(obj));
}
static VALUE
optimized_unescape(VALUE str, VALUE encoding)
{
long i, len, beg = 0;
VALUE dest = 0;
const char *cstr;
int cr, origenc, encidx = rb_to_encoding_index(encoding);
len = RSTRING_LEN(str);
cstr = RSTRING_PTR(str);
for (i = 0; i < len; ++i) {
char buf[1];
const char c = cstr[i];
int clen = 0;
if (c == '%') {
if (i + 3 > len) break;
if (!ISXDIGIT(cstr[i+1])) continue;
if (!ISXDIGIT(cstr[i+2])) continue;
buf[0] = ((char_to_number(cstr[i+1]) << 4)
| char_to_number(cstr[i+2]));
clen = 2;
}
else if (c == '+') {
buf[0] = ' ';
}
else {
continue;
}
if (!dest) {
dest = rb_str_buf_new(len);
}
rb_str_cat(dest, cstr + beg, i - beg);
i += clen;
beg = i + 1;
rb_str_cat(dest, buf, 1);
}
if (dest) {
rb_str_cat(dest, cstr + beg, len - beg);
preserve_original_state(str, dest);
cr = ENC_CODERANGE_UNKNOWN;
}
else {
dest = rb_str_dup(str);
cr = ENC_CODERANGE(str);
}
origenc = rb_enc_get_index(str);
if (origenc != encidx) {
rb_enc_associate_index(dest, encidx);
if (!ENC_CODERANGE_CLEAN_P(rb_enc_str_coderange(dest))) {
rb_enc_associate_index(dest, origenc);
if (cr != ENC_CODERANGE_UNKNOWN)
ENC_CODERANGE_SET(dest, cr);
}
}
return dest;
}
static VALUE step(VALUE self)
{
sqlite3StmtRubyPtr ctx;
sqlite3_stmt *stmt;
int value, length;
VALUE list;
#ifdef HAVE_RUBY_ENCODING_H
rb_encoding * internal_encoding;
int enc_index;
#endif
Data_Get_Struct(self, sqlite3StmtRuby, ctx);
REQUIRE_OPEN_STMT(ctx);
if(ctx->done_p) return Qnil;
#ifdef HAVE_RUBY_ENCODING_H
{
VALUE db = rb_iv_get(self, "@connection");
VALUE encoding = rb_funcall(db, rb_intern("encoding"), 0);
enc_index = NIL_P(encoding) ? rb_utf8_encindex() : rb_to_encoding_index(encoding);
internal_encoding = rb_default_internal_encoding();
}
#endif
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)
);
#ifdef HAVE_RUBY_ENCODING_H
rb_enc_associate_index(str, enc_index);
if(internal_encoding)
str = rb_str_export_to_enc(str, internal_encoding);
#endif
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:
CHECK(sqlite3_db_handle(ctx->st), value);
}
return list;
}
static int
str_transcode(int argc, VALUE *argv, VALUE *self)
{
VALUE dest;
VALUE str = *self;
long blen, slen;
unsigned char *buf, *bp, *sp, *fromp;
rb_encoding *from_enc, *to_enc;
const char *from_e, *to_e;
int from_encidx, to_encidx;
VALUE from_encval, to_encval;
const rb_transcoder *my_transcoder;
rb_transcoding my_transcoding;
int final_encoding = 0;
VALUE opt;
int options = 0;
opt = rb_check_convert_type(argv[argc-1], T_HASH, "Hash", "to_hash");
if (!NIL_P(opt)) {
VALUE v;
argc--;
v = rb_hash_aref(opt, sym_invalid);
if (NIL_P(v)) {
rb_raise(rb_eArgError, "unknown value for invalid: setting");
}
else if (v==sym_ignore) {
options |= INVALID_IGNORE;
}
}
if (argc < 1 || argc > 2) {
rb_raise(rb_eArgError, "wrong number of arguments (%d for 1..2)", argc);
}
if ((to_encidx = rb_to_encoding_index(to_encval = argv[0])) < 0) {
to_enc = 0;
to_encidx = 0;
to_e = StringValueCStr(to_encval);
}
else {
to_enc = rb_enc_from_index(to_encidx);
to_e = rb_enc_name(to_enc);
}
if (argc==1) {
from_encidx = rb_enc_get_index(str);
from_enc = rb_enc_from_index(from_encidx);
from_e = rb_enc_name(from_enc);
}
else if ((from_encidx = rb_to_encoding_index(from_encval = argv[1])) < 0) {
from_enc = 0;
from_e = StringValueCStr(from_encval);
}
else {
from_enc = rb_enc_from_index(from_encidx);
from_e = rb_enc_name(from_enc);
}
if (from_enc && from_enc == to_enc) {
return -1;
}
if (from_enc && to_enc && rb_enc_asciicompat(from_enc) && rb_enc_asciicompat(to_enc)) {
if (ENC_CODERANGE(str) == ENC_CODERANGE_7BIT) {
return to_encidx;
}
}
if (encoding_equal(from_e, to_e)) {
return -1;
}
do { /* loop for multistep transcoding */
/* later, maybe use smaller intermediate strings for very long strings */
if (!(my_transcoder = transcode_dispatch(from_e, to_e))) {
rb_raise(rb_eArgError, "transcoding not supported (from %s to %s)", from_e, to_e);
}
my_transcoding.transcoder = my_transcoder;
if (my_transcoder->preprocessor) {
fromp = sp = (unsigned char *)RSTRING_PTR(str);
slen = RSTRING_LEN(str);
blen = slen + 30; /* len + margin */
dest = rb_str_tmp_new(blen);
bp = (unsigned char *)RSTRING_PTR(dest);
my_transcoding.ruby_string_dest = dest;
(*my_transcoder->preprocessor)(&fromp, &bp, (sp+slen), (bp+blen), &my_transcoding);
if (fromp != sp+slen) {
rb_raise(rb_eArgError, "not fully converted, %td bytes left", sp+slen-fromp);
}
buf = (unsigned char *)RSTRING_PTR(dest);
*bp = '\0';
rb_str_set_len(dest, bp - buf);
str = dest;
}
fromp = sp = (unsigned char *)RSTRING_PTR(str);
slen = RSTRING_LEN(str);
blen = slen + 30; /* len + margin */
dest = rb_str_tmp_new(blen);
bp = (unsigned char *)RSTRING_PTR(dest);
my_transcoding.ruby_string_dest = dest;
my_transcoding.flush_func = str_transcoding_resize;
transcode_loop(&fromp, &bp, (sp+slen), (bp+blen), my_transcoder, &my_transcoding, options);
if (fromp != sp+slen) {
rb_raise(rb_eArgError, "not fully converted, %td bytes left", sp+slen-fromp);
}
buf = (unsigned char *)RSTRING_PTR(dest);
*bp = '\0';
rb_str_set_len(dest, bp - buf);
if (my_transcoder->postprocessor) {
str = dest;
fromp = sp = (unsigned char *)RSTRING_PTR(str);
slen = RSTRING_LEN(str);
blen = slen + 30; /* len + margin */
dest = rb_str_tmp_new(blen);
bp = (unsigned char *)RSTRING_PTR(dest);
my_transcoding.ruby_string_dest = dest;
(*my_transcoder->postprocessor)(&fromp, &bp, (sp+slen), (bp+blen), &my_transcoding);
if (fromp != sp+slen) {
rb_raise(rb_eArgError, "not fully converted, %td bytes left", sp+slen-fromp);
}
buf = (unsigned char *)RSTRING_PTR(dest);
*bp = '\0';
rb_str_set_len(dest, bp - buf);
}
if (encoding_equal(my_transcoder->to_encoding, to_e)) {
final_encoding = 1;
}
else {
from_e = my_transcoder->to_encoding;
str = dest;
}
} while (!final_encoding);
/* set encoding */
if (!to_enc) {
to_encidx = rb_define_dummy_encoding(to_e);
}
*self = dest;
return to_encidx;
}
