tommy/libpostal
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[parser] phrase search with address dictionaries should not use the language given at training time since it's not currently available at runtime (without pulling in the language classifier, which may be warranted at some point, especially if the model can be made smaller/sparser)

Browse Source
This commit is contained in:
Al
2017-02-15 22:32:30 -05:00
parent a3e51db32d
commit c380b3e91b
1 changed files with 3 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
src/address_parser.c
View File

@@ -685,16 +685,16 @@ void address_parser_context_fill(address_parser_context_t *context, address_pars
size_t num_tokens = tokens->n;
bool have_address_phrases = search_address_dictionaries_tokens_with_phrases(normalized_str, normalized_tokens, context->language, &address_dictionary_phrases);
bool have_address_phrases = search_address_dictionaries_tokens_with_phrases(normalized_str, normalized_tokens, NULL, &address_dictionary_phrases);
token_phrase_memberships(address_dictionary_phrases, address_phrase_memberships, num_tokens);
for (size_t i = 0; i < num_tokens; i++) {
token_t token = tokens->a[i];
phrase_t prefix_phrase = search_address_dictionaries_prefix(str + token.offset, token.len, language);
phrase_t prefix_phrase = search_address_dictionaries_prefix(str + token.offset, token.len, NULL);
phrase_array_push(context->prefix_phrases, prefix_phrase);
phrase_t suffix_phrase = search_address_dictionaries_suffix(str + token.offset, token.len, language);
phrase_t suffix_phrase = search_address_dictionaries_suffix(str + token.offset, token.len, NULL);
phrase_array_push(context->suffix_phrases, suffix_phrase);
}