/**
* Create a hubbub parser
*
* \param enc Source document encoding, or NULL to autodetect
* \param fix_enc Permit fixing up of encoding if it's frequently misused
* \param alloc Memory (de)allocation function
* \param pw Pointer to client-specific private data (may be NULL)
* \param parser Pointer to location to receive parser instance
* \return HUBBUB_OK on success,
* HUBBUB_BADPARM on bad parameters,
* HUBBUB_NOMEM on memory exhaustion,
* HUBBUB_BADENCODING if ::enc is unsupported
*/
hubbub_error hubbub_parser_create(const char *enc, bool fix_enc,
hubbub_allocator_fn alloc, void *pw, hubbub_parser **parser)
{
parserutils_error perror;
hubbub_error error;
hubbub_parser *p;
if (alloc == NULL || parser == NULL)
return HUBBUB_BADPARM;
p = alloc(NULL, sizeof(hubbub_parser), pw);
if (p == NULL)
return HUBBUB_NOMEM;
/* If we have an encoding and we're permitted to fix up likely broken
* ones, then attempt to do so. */
if (enc != NULL && fix_enc == true) {
uint16_t mibenum = parserutils_charset_mibenum_from_name(enc,
strlen(enc));
if (mibenum != 0) {
hubbub_charset_fix_charset(&mibenum);
enc = parserutils_charset_mibenum_to_name(mibenum);
}
}
perror = parserutils_inputstream_create(enc,
enc != NULL ? HUBBUB_CHARSET_CONFIDENT : HUBBUB_CHARSET_UNKNOWN,
hubbub_charset_extract, alloc, pw, &p->stream);
if (perror != PARSERUTILS_OK) {
alloc(p, 0, pw);
return hubbub_error_from_parserutils_error(perror);
}
error = hubbub_tokeniser_create(p->stream, alloc, pw, &p->tok);
if (error != HUBBUB_OK) {
parserutils_inputstream_destroy(p->stream);
alloc(p, 0, pw);
return error;
}
error = hubbub_treebuilder_create(p->tok, alloc, pw, &p->tb);
if (error != HUBBUB_OK) {
hubbub_tokeniser_destroy(p->tok);
parserutils_inputstream_destroy(p->stream);
alloc(p, 0, pw);
return error;
}
p->alloc = alloc;
p->pw = pw;
*parser = p;
return HUBBUB_OK;
}
int main (int argc, char **argv)
{
parserutils_charset_aliases_canon *c;
UNUSED(argc);
UNUSED(argv);
c = parserutils__charset_alias_canonicalise("moose", 5);
if (c) {
printf("FAIL - found invalid encoding 'moose'\n");
return 1;
}
c = parserutils__charset_alias_canonicalise("csinvariant", 11);
if (c) {
printf("%s %d\n", c->name, c->mib_enum);
} else {
printf("FAIL - failed finding encoding 'csinvariant'\n");
return 1;
}
c = parserutils__charset_alias_canonicalise("csinvariant\"", 12);
if (c) {
printf("%s %d\n", c->name, c->mib_enum);
} else {
printf("FAIL - failed finding encoding 'csinvariant'\n");
return 1;
}
c = parserutils__charset_alias_canonicalise("nats-sefi-add", 13);
if (c) {
printf("%s %d\n", c->name, c->mib_enum);
} else {
printf("FAIL - failed finding encoding 'nats-sefi-add'\n");
return 1;
}
printf("%d\n", parserutils_charset_mibenum_from_name(c->name,
strlen(c->name)));
printf("%s\n", parserutils_charset_mibenum_to_name(c->mib_enum));
c = parserutils__charset_alias_canonicalise("u.t.f.8", 7);
if (c) {
printf("%s %d\n", c->name, c->mib_enum);
} else {
printf("FAIL - failed finding encoding 'u.t.f.8'\n");
return 1;
}
printf("PASS\n");
return 0;
}
void run_test(const uint8_t *data, size_t len, char *expected)
{
uint16_t mibenum = 0;
hubbub_charset_source source = HUBBUB_CHARSET_UNKNOWN;
static int testnum;
assert(hubbub_charset_extract(data, len,
&mibenum, &source) == HUBBUB_OK);
assert(mibenum != 0);
printf("%d: Detected charset %s (%d) Source %d Expected %s (%d)\n",
++testnum, parserutils_charset_mibenum_to_name(mibenum),
mibenum, source, expected,
parserutils_charset_mibenum_from_name(
expected, strlen(expected)));
assert(mibenum == parserutils_charset_mibenum_from_name(
expected, strlen(expected)));
}
/**
* Process a meta tag as if "in head".
*
* \param treebuilder The treebuilder instance
* \param token The token to process
*/
static hubbub_error process_meta_in_head(hubbub_treebuilder *treebuilder,
const hubbub_token *token)
{
static uint16_t utf16, utf16be, utf16le;
uint16_t charset_enc = 0;
uint16_t content_type_enc = 0;
size_t i;
hubbub_error err = HUBBUB_OK;
err = insert_element(treebuilder, &token->data.tag, false);
if (err != HUBBUB_OK)
return err;
/** \todo ack sc flag */
if (treebuilder->tree_handler->encoding_change == NULL)
return err;
/* Grab UTF-16 MIBenums */
if (utf16 == 0) {
utf16 = parserutils_charset_mibenum_from_name(
"utf-16", SLEN("utf-16"));
utf16be = parserutils_charset_mibenum_from_name(
"utf-16be", SLEN("utf-16be"));
utf16le = parserutils_charset_mibenum_from_name(
"utf-16le", SLEN("utf-16le"));
assert(utf16 != 0 && utf16be != 0 && utf16le != 0);
}
for (i = 0; i < token->data.tag.n_attributes; i++) {
hubbub_attribute *attr = &token->data.tag.attributes[i];
if (hubbub_string_match(attr->name.ptr, attr->name.len,
(const uint8_t *) "charset",
SLEN("charset")) == true) {
/* Extract charset */
charset_enc = parserutils_charset_mibenum_from_name(
(const char *) attr->value.ptr,
attr->value.len);
} else if (hubbub_string_match(attr->name.ptr, attr->name.len,
(const uint8_t *) "content",
SLEN("content")) == true) {
/* Extract charset from Content-Type */
content_type_enc = hubbub_charset_parse_content(
attr->value.ptr, attr->value.len);
}
}
/* Fall back, if necessary */
if (charset_enc == 0 && content_type_enc != 0)
charset_enc = content_type_enc;
if (charset_enc != 0) {
const char *name;
hubbub_charset_fix_charset(&charset_enc);
/* Change UTF-16 to UTF-8 */
if (charset_enc == utf16le || charset_enc == utf16be ||
charset_enc == utf16) {
charset_enc = parserutils_charset_mibenum_from_name(
"UTF-8", SLEN("UTF-8"));
}
name = parserutils_charset_mibenum_to_name(charset_enc);
err = treebuilder->tree_handler->encoding_change(
treebuilder->tree_handler->ctx, name);
}
return err;
}
