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

[parser] parser only inserts spaces in the output if there were spaces (or other ignorable tokens) in the normalized input

Browse Source
This commit is contained in:
Al
2017-03-15 03:34:59 -04:00
parent d43989cf1c
commit 1a1f0a44d2
1 changed files with 9 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
src/address_parser.c
View File

@@ -1712,6 +1712,7 @@ address_parser_response_t *address_parser_parse(char *address, char *language, c
// If the whole input string is a single known phrase at the SUBURB level or higher, bypass sequence prediction altogether
phrase_t only_phrase = NULL_PHRASE;
token_t token, prev_token;
bool is_postal = false;
if (context->component_phrases->n == 1) {
only_phrase = context->component_phrases->a[0];
@@ -1784,8 +1785,11 @@ address_parser_response_t *address_parser_parse(char *address, char *language, c
cstring_array *labels = cstring_array_new_size(num_strings);
cstring_array *components = cstring_array_new_size(strlen(address) + num_strings);
token_t *tokens = tokenized_str->tokens->a;
for (size_t i = 0; i < num_strings; i++) {
char *str = tokenized_string_get_token(tokenized_str, i);
char *label = cstring_array_get_string(token_labels, i);
if (prev_label == NULL || strcmp(label, prev_label) != 0) {
@@ -1795,7 +1799,11 @@ address_parser_response_t *address_parser_parse(char *address, char *language, c
}
if (prev_label != NULL && strcmp(label, prev_label) == 0) {
cstring_array_cat_string(components, " ");
token = tokens[i];
prev_token = tokens[i - 1];
if (token.offset > prev_token.offset + prev_token.len) {
cstring_array_cat_string(components, " ");
}
cstring_array_cat_string(components, str);
} else {
cstring_array_append_string(components, str);