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

[parser/cli] Using NFC normalization on the output in the parser client (closes #30). Optional command-line arg for parser output dir, useful for spot-checking different experiments

Browse Source
This commit is contained in:
Al
2016-06-22 11:56:35 -04:00
parent e19bc86c5a
commit 106dfa80c3
1 changed files with 14 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
src/address_parser_cli.c
View File

@@ -18,7 +18,6 @@
#include "linenoise/linenoise.h" #include "linenoise/linenoise.h"
#include "log/log.h" #include "log/log.h"
bool load_address_parser_dependencies(void) { bool load_address_parser_dependencies(void) {
if (!address_dictionary_module_setup(NULL)) { if (!address_dictionary_module_setup(NULL)) {
log_error("Could not load address dictionaries\n"); log_error("Could not load address dictionaries\n");
@@ -47,7 +46,7 @@ int main(int argc, char **argv) {
printf("Loading models...\n"); printf("Loading models...\n");
if (!libpostal_setup() || !libpostal_setup_parser()) { if (!libpostal_setup() || !geodb_module_setup(NULL) || !address_parser_module_setup(address_parser_dir)) {
exit(EXIT_FAILURE); exit(EXIT_FAILURE);
} }
@@ -116,15 +115,26 @@ int main(int argc, char **argv) {
printf("Result:\n\n"); printf("Result:\n\n");
printf("{\n"); printf("{\n");
for (int i = 0; i < parsed->num_components; i++) { for (int i = 0; i < parsed->num_components; i++) {
char *json_string = json_encode_string(parsed->components[i]); char *component = parsed->components[i];
utf8proc_uint8_t *normalized = NULL;
utf8proc_map((utf8proc_uint8_t *)component, 0, &normalized, UTF8PROC_NULLTERM | UTF8PROC_COMPOSE);
if (normalized == NULL) {
log_error("Error parsing address\n");
exit(EXIT_FAILURE);
}
char *json_string = json_encode_string((char *)normalized);
printf(" \"%s\": %s%s\n", parsed->labels[i], json_string, i < parsed->num_components - 1 ? "," : ""); printf(" \"%s\": %s%s\n", parsed->labels[i], json_string, i < parsed->num_components - 1 ? "," : "");
free(normalized);
free(json_string);
} }
printf("}\n"); printf("}\n");
printf("\n"); printf("\n");
address_parser_response_destroy(parsed); address_parser_response_destroy(parsed);
} else { } else {
printf("Error parsing address\n"); log_error("Error parsing address\n");
exit(EXIT_FAILURE);
} }
next_input: next_input: