From 2106a6cfe4f44ebc57e14ff3e09ad5d98edad76c Mon Sep 17 00:00:00 2001
From: Al <albarrentine@gmail.com>
Date: Sat, 8 Aug 2015 19:44:50 -0400
Subject: [PATCH] [build] Adding command-line test and bench programs

---
 src/Makefile.am |  7 ++++-
 src/bench.c     | 82 +++++++++++++++++++++++++++++++++++++++++++++++++
 src/main.c      | 65 +++++++++++++++++++++++++++++++++++++++
 3 files changed, 153 insertions(+), 1 deletion(-)
 create mode 100644 src/bench.c
 create mode 100644 src/main.c

diff --git a/src/Makefile.am b/src/Makefile.am
index 070601da..5c1d7e13 100644
--- a/src/Makefile.am
+++ b/src/Makefile.am
@@ -19,8 +19,13 @@ noinst_LTLIBRARIES = libscanner.la
 libscanner_la_SOURCES = scanner.c
 libscanner_la_CFLAGS =  $(CFLAGS_O0)
 
+noinst_PROGRAMS = libpostal bench
+libpostal_SOURCES = main.c
+libpostal_LDADD = libpostal.la
+bench_SOURCES = bench.c
+bench_LDADD = libpostal.la libscanner.la
+
 pkginclude_HEADERS = libpostal.h
-pkgdata_DATA = libpostal_data.tar.gz
 
 LIBPOSTAL_S3_BUCKET_NAME = libpostal
 LIBPOSTAL_S3_BUCKET_URL = http://$(LIBPOSTAL_S3_BUCKET_NAME).s3.amazonaws.com
diff --git a/src/bench.c b/src/bench.c
new file mode 100644
index 00000000..323e57ab
--- /dev/null
+++ b/src/bench.c
@@ -0,0 +1,82 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/time.h>
+
+#include "libpostal.h"
+#include "log/log.h"
+#include "scanner.h"
+
+int main(int argc, char **argv) {
+    if (argc < 3) {
+        log_error("Usage: test_libpostal string languages...\n");
+        exit(EXIT_FAILURE);
+    }
+
+    char *str = argv[1];
+    char *languages[argc - 2];
+    for (int i = 0; i < argc - 2; i++) {
+        char *arg = argv[i + 2];
+        if (strlen(arg) >= MAX_LANGUAGE_LEN) {
+            printf("arg %d was longer than a language code (%d chars). Make sure to quote the input string\n", i + 2, MAX_LANGUAGE_LEN - 1);
+        }
+        languages[i] = arg;
+    }
+
+    if (!libpostal_setup()) {
+        exit(EXIT_FAILURE);
+    }
+
+    normalize_options_t options = {
+        .num_languages = 1,
+        .languages = languages,
+        .address_components = ADDRESS_HOUSE_NUMBER | ADDRESS_STREET | ADDRESS_UNIT,
+        .latin_ascii = 1,
+        .transliterate = 1,
+        .strip_accents = 1,
+        .decompose = 1,
+        .lowercase = 1,
+        .trim_string = 1,
+        .replace_word_hyphens = 1,
+        .delete_word_hyphens = 0,
+        .replace_numeric_hyphens = 0,
+        .delete_numeric_hyphens = 0,
+        .split_alpha_from_numeric = 1,
+        .delete_final_periods = 1,
+        .delete_acronym_periods = 1,
+        .drop_english_possessives = 1,
+        .delete_apostrophes = 1,
+        .expand_numex = 1,
+        .roman_numerals = 1
+    };
+
+    uint64_t num_expansions;
+
+    char **strings;
+    char *normalized;
+
+    int num_loops = 100000;
+
+    token_array *tokens = tokenize(str);
+    uint64_t num_tokens = tokens->n;
+    token_array_destroy(tokens);
+
+    clock_t t1 = clock();
+    for (int i = 0; i < num_loops; i++) {
+        strings = expand_address(str, options, &num_expansions);
+        for (uint64_t i = 0; i < num_expansions; i++) {
+            normalized = strings[i];
+            free(normalized);
+        }
+        free(strings);
+    }
+    clock_t t2 = clock();
+
+    double benchmark_time = (double)(t2 - t1) / CLOCKS_PER_SEC;
+    printf("Benchmark time: %f\n", benchmark_time);
+    double addresses_per_second = num_loops / benchmark_time;
+    printf("addresses/s = %f\n", addresses_per_second);
+    double tokens_per_second = (num_loops * num_tokens) / benchmark_time;
+    printf("tokens/s = %f\n", tokens_per_second);
+    libpostal_teardown();
+}
diff --git a/src/main.c b/src/main.c
new file mode 100644
index 00000000..e803a5ef
--- /dev/null
+++ b/src/main.c
@@ -0,0 +1,65 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/time.h>
+
+#include "libpostal.h"
+#include "log/log.h"
+
+int main(int argc, char **argv) {
+    if (argc < 3) {
+        log_error("Usage: test_libpostal string languages...\n");
+        exit(EXIT_FAILURE);
+    }
+    char *str = argv[1];
+    char *languages[argc - 2];
+    for (int i = 0; i < argc - 2; i++) {
+        char *arg = argv[i + 2];
+        if (strlen(arg) >= MAX_LANGUAGE_LEN) {
+            printf("arg %d was longer than a language code (%d chars). Make sure to quote the input string\n", i + 2, MAX_LANGUAGE_LEN - 1);
+        }
+        languages[i] = arg;
+    }
+
+    if (!libpostal_setup()) {
+        exit(EXIT_FAILURE);
+    }
+
+    normalize_options_t options = {
+        .num_languages = 1,
+        .languages = languages,
+        .address_components = ADDRESS_HOUSE_NUMBER | ADDRESS_STREET | ADDRESS_UNIT,
+        .latin_ascii = 1,
+        .transliterate = 1,
+        .strip_accents = 1,
+        .decompose = 1,
+        .lowercase = 1,
+        .trim_string = 1,
+        .replace_word_hyphens = 1,
+        .delete_word_hyphens = 0,
+        .replace_numeric_hyphens = 0,
+        .delete_numeric_hyphens = 0,
+        .split_alpha_from_numeric = 1,
+        .delete_final_periods = 1,
+        .delete_acronym_periods = 1,
+        .drop_english_possessives = 1,
+        .delete_apostrophes = 1,
+        .expand_numex = 1,
+        .roman_numerals = 1
+    };
+
+    uint64_t num_expansions;
+
+    char **strings = expand_address(str, options, &num_expansions);
+
+    char *normalized;
+    for (uint64_t i = 0; i < num_expansions; i++) {
+        normalized = strings[i];
+        printf("%s\n", normalized);        
+        free(normalized);
+    }
+
+    free(strings);
+
+    libpostal_teardown();
+}