now hakurei can tokenize any japanese phrase
This commit is contained in:
parent
2fce3c4a48
commit
c402bffc2a
|
@ -6,7 +6,7 @@
|
|||
#include <cctype>
|
||||
#include <xcdat.hpp>
|
||||
|
||||
// as this is in test phase, i recommend testing with the input being "特別協力組合理事"
|
||||
|
||||
|
||||
struct Entry {
|
||||
std::string_view decoded_view;
|
||||
|
|
109
src/main.cc~
Normal file
109
src/main.cc~
Normal file
|
@ -0,0 +1,109 @@
|
|||
#include <iostream>
|
||||
#include <vector>
|
||||
#include <algorithm>
|
||||
#include <limits>
|
||||
#include <cstring>
|
||||
#include <cctype>
|
||||
#include <xcdat.hpp>
|
||||
|
||||
// as this is in test phase, i recommend testing with the input being "特別協力組合理事"
|
||||
|
||||
struct Entry {
|
||||
std::string_view decoded_view;
|
||||
uint64_t id;
|
||||
Entry(std::string_view decoded_view, uint64_t id) : decoded_view(decoded_view), id(id) {}
|
||||
};
|
||||
|
||||
int main(int argc, char* argv[]) {
|
||||
if (argc < 2) {
|
||||
std::cerr << "Usage: " << argv[0] << " <search_string> [--debug] [--dict <path_to_dictionary>]" << std::endl;
|
||||
return 1;
|
||||
}
|
||||
|
||||
std::string search_string;
|
||||
bool debug_mode = false;
|
||||
std::string dict_file = "dict.bin"; // Default dictionary file path
|
||||
|
||||
// Parse command line arguments
|
||||
for (int i = 1; i < argc; ++i) {
|
||||
if (std::strcmp(argv[i], "--debug") == 0) {
|
||||
debug_mode = true;
|
||||
} else if (std::strcmp(argv[i], "--dict") == 0 && i + 1 < argc) {
|
||||
dict_file = argv[i + 1];
|
||||
i++; // Skip the next argument as it's the dictionary file path
|
||||
} else {
|
||||
search_string = argv[i];
|
||||
}
|
||||
}
|
||||
|
||||
if (search_string.empty()) {
|
||||
std::cerr << "Search string not provided." << std::endl;
|
||||
return 1;
|
||||
}
|
||||
|
||||
std::string raw_output;
|
||||
const auto trie = xcdat::load<xcdat::trie_15_type>(dict_file);
|
||||
|
||||
std::vector<std::string> substrings;
|
||||
std::vector<Entry> results;
|
||||
while (!search_string.empty()) {
|
||||
auto itr = trie.make_prefix_iterator(search_string);
|
||||
|
||||
while (itr.next()) {
|
||||
results.emplace_back(itr.decoded_view(), itr.id());
|
||||
}
|
||||
|
||||
size_t min_length = std::numeric_limits<size_t>::max();
|
||||
std::string_view smallest_prefix;
|
||||
|
||||
for (const auto& entry : results) {
|
||||
if (entry.decoded_view.size() < min_length) {
|
||||
min_length = entry.decoded_view.size();
|
||||
smallest_prefix = entry.decoded_view;
|
||||
}
|
||||
}
|
||||
|
||||
if (min_length > 0) {
|
||||
std::string substring = search_string.substr(0, search_string.find(smallest_prefix));
|
||||
if (!substring.empty()) {
|
||||
substrings.push_back(substring);
|
||||
}
|
||||
std::cout << search_string << " - " << smallest_prefix << " = ";
|
||||
|
||||
if (debug_mode) {
|
||||
raw_output += search_string + " - " + std::string(smallest_prefix) + " = ";
|
||||
}
|
||||
|
||||
size_t pos = search_string.find(smallest_prefix);
|
||||
if (pos != std::string::npos) {
|
||||
search_string.erase(0, pos + smallest_prefix.length());
|
||||
}
|
||||
else break;
|
||||
|
||||
std::cout << search_string << std::endl;
|
||||
|
||||
if (smallest_prefix.length() == 0)
|
||||
break;
|
||||
|
||||
if (debug_mode) {
|
||||
raw_output += search_string + '\n';
|
||||
}
|
||||
|
||||
while (!search_string.empty() && std::isspace(search_string.front())) {
|
||||
search_string.erase(0, 1);
|
||||
}
|
||||
} else {
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
if (debug_mode) {
|
||||
std::cout << "Stored raw output:" << std::endl << raw_output << std::endl;
|
||||
}
|
||||
|
||||
for (const auto& sub : substrings) {
|
||||
std::cout << sub << std::endl;
|
||||
}
|
||||
|
||||
return 0;
|
||||
}
|
Loading…
Reference in a new issue