1045 lines
30 KiB
C++
1045 lines
30 KiB
C++
#include "wrap-string.hpp"
|
|
#include "wrap-vector.hpp"
|
|
#include "util.hpp"
|
|
#include "fast-float.hpp"
|
|
#include "luastack.hpp"
|
|
|
|
#include "../../ext/unicode-stuff.hpp"
|
|
|
|
#include <algorithm>
|
|
#include <sys/types.h>
|
|
#include <sys/stat.h>
|
|
#include <iomanip>
|
|
#include <cassert>
|
|
#include <cstdlib>
|
|
#include <cmath>
|
|
#include <charconv>
|
|
|
|
namespace sv {
|
|
|
|
bool case_insensitive_eq(string_view s1, string_view s2) {
|
|
if (s1.size() != s2.size()) return false;
|
|
for (int i = 0; i < int(s1.size()); i++) {
|
|
char c1 = s1[i];
|
|
char c2 = s2[i];
|
|
if (ascii_isupper(c1)) c1 += 'a'-'A';
|
|
if (ascii_isupper(c2)) c2 += 'a'-'A';
|
|
if (c1 != c2) return false;
|
|
}
|
|
return true;
|
|
}
|
|
|
|
bool valid_int64(string_view value) {
|
|
int64_t result;
|
|
const char *last = value.data() + value.size();
|
|
auto r = std::from_chars(value.data(), last, result, 10);
|
|
if (r.ec != std::errc()) return false;
|
|
if (r.ptr != last) return false;
|
|
return true;
|
|
}
|
|
|
|
bool valid_hex64(string_view value) {
|
|
int64_t result;
|
|
const char *last = value.data() + value.size();
|
|
auto r = std::from_chars(value.data(), last, result, 16);
|
|
if (r.ec != std::errc()) return false;
|
|
if (r.ptr != last) return false;
|
|
return true;
|
|
}
|
|
|
|
bool valid_double(string_view value) {
|
|
double result;
|
|
const char *last = value.data() + value.size();
|
|
auto r = fast_float::from_chars(value.data(), last, result);
|
|
if (r.ec != std::errc()) return false;
|
|
if (r.ptr != last) return false;
|
|
return true;
|
|
}
|
|
|
|
bool valid_hostname(string_view value) {
|
|
bool start_word = true;
|
|
bool last_dash = false;
|
|
for (char c : value) {
|
|
if (c == '.') {
|
|
if (start_word) return false;
|
|
if (last_dash) return false;
|
|
start_word = true;
|
|
last_dash = false;
|
|
} else if (c == '-') {
|
|
if (start_word) return false;
|
|
start_word = false;
|
|
last_dash = true;
|
|
} else if (ascii_isalnum(c)) {
|
|
start_word = false;
|
|
last_dash = false;
|
|
} else {
|
|
return false;
|
|
}
|
|
}
|
|
if (start_word || last_dash) return false;
|
|
return true;
|
|
}
|
|
|
|
int64_t to_int64(string_view value, int64_t errval) {
|
|
int64_t result;
|
|
const char *p = value.data();
|
|
const char *last = p + value.size();
|
|
if ((p < last) && (*p == '+')) p++;
|
|
auto r = std::from_chars(p, last, result, 10);
|
|
if (r.ec != std::errc()) return errval;
|
|
if (r.ptr != last) return errval;
|
|
return result;
|
|
}
|
|
|
|
uint64_t to_hex64(string_view value, uint64_t errval) {
|
|
uint64_t result;
|
|
if (sv::zfront(value) == '-') return errval;
|
|
const char *last = value.data() + value.size();
|
|
auto r = std::from_chars(value.data(), last, result, 16);
|
|
if (r.ec != std::errc()) return errval;
|
|
if (r.ptr != last) return errval;
|
|
return result;
|
|
}
|
|
|
|
double to_double(string_view value, double errval) {
|
|
double result;
|
|
const char *last = value.data() + value.size();
|
|
auto r = fast_float::from_chars(value.data(), last, result);
|
|
if (r.ec != std::errc()) return errval;
|
|
if (r.ptr != last) return errval;
|
|
return result;
|
|
}
|
|
|
|
string_view ltrim(string_view v) {
|
|
while ((!v.empty()) && (ascii_isspace(v.front()))) {
|
|
v.remove_prefix(1);
|
|
}
|
|
return v;
|
|
}
|
|
|
|
string_view rtrim(string_view v) {
|
|
while ((!v.empty()) && (ascii_isspace(v.back()))) {
|
|
v.remove_suffix(1);
|
|
}
|
|
return v;
|
|
}
|
|
|
|
string_view trim(string_view v) {
|
|
while ((!v.empty()) && (ascii_isspace(v.front()))) {
|
|
v.remove_prefix(1);
|
|
}
|
|
while ((!v.empty()) && (ascii_isspace(v.back()))) {
|
|
v.remove_suffix(1);
|
|
}
|
|
return v;
|
|
}
|
|
|
|
string_view ltrim(string_view v, char c) {
|
|
while ((!v.empty()) && (v.front() == c)) {
|
|
v.remove_prefix(1);
|
|
}
|
|
return v;
|
|
}
|
|
|
|
string_view rtrim(string_view v, char c) {
|
|
while ((!v.empty()) && (v.back() == c)) {
|
|
v.remove_suffix(1);
|
|
}
|
|
return v;
|
|
}
|
|
|
|
string_view trim(string_view v, char c) {
|
|
while ((!v.empty()) && (v.front() == c)) {
|
|
v.remove_prefix(1);
|
|
}
|
|
while ((!v.empty()) && (v.back() == c)) {
|
|
v.remove_suffix(1);
|
|
}
|
|
return v;
|
|
}
|
|
|
|
bool has_prefix(string_view s, string_view prefix) {
|
|
return 0 == s.compare(0, prefix.size(), prefix);
|
|
}
|
|
|
|
bool has_suffix(string_view s, string_view suffix) {
|
|
if (s.length() >= suffix.length()) {
|
|
return (0 == s.compare (s.length() - suffix.length(), suffix.length(), suffix));
|
|
} else {
|
|
return false;
|
|
}
|
|
}
|
|
|
|
int common_prefix_length(string_view a, string_view b) {
|
|
int minlen = std::min(a.size(), b.size());
|
|
for (int i = 0; i < minlen; i++) {
|
|
if (a[i] != b[i]) return i;
|
|
}
|
|
return minlen;
|
|
}
|
|
|
|
bool is_lua_id(string_view str) {
|
|
if (str.size() == 0) return false;
|
|
char c=str[0];
|
|
if (!ascii_isualpha(c)) return false;
|
|
for (int i = 1; i < int(str.size()); i++) {
|
|
char c = str[i];
|
|
if (!ascii_isualnum(c)) return false;
|
|
}
|
|
return true;
|
|
}
|
|
|
|
bool is_lua_classname(string_view s) {
|
|
if (s == "_G") return false;
|
|
return is_lua_id(s);
|
|
}
|
|
|
|
bool is_lua_comment(string_view s) {
|
|
int start = 0;
|
|
while ((start < int(s.size())) && ((s[start]==' ') || (s[start]=='\t'))) start++;
|
|
return s.substr(start, 2) == "--";
|
|
}
|
|
|
|
bool is_possible_long_lua_expression(string_view s) {
|
|
read_space(s);
|
|
string_view id = read_lua_identifier(s);
|
|
if (id.empty()) return false;
|
|
if ((id == "function") || (id == "if") || (id == "while") || (id == "for") || (id == "repeat") || (id == "do")) return true;
|
|
if (id == "local")
|
|
{
|
|
read_space(s);
|
|
id = read_lua_identifier(s);
|
|
}
|
|
read_space(s);
|
|
read_prefix(s, "="); // If not present, returns false but we continue anyway.
|
|
read_space(s);
|
|
if (has_prefix(s, "[")) return true;
|
|
if (has_prefix(s, "(")) return true;
|
|
return false;
|
|
}
|
|
|
|
bool is_whitespace(string_view s) {
|
|
for (int i = 0; i < int(s.size()); i++) {
|
|
if (!ascii_isspace(s[i])) {
|
|
return false;
|
|
}
|
|
}
|
|
return true;
|
|
}
|
|
|
|
string_view lua_function_proto_name(string_view s) {
|
|
// Skip any leading whitespace.
|
|
read_space(s);
|
|
|
|
// Skip over the word 'function'. If it's not there, fail.
|
|
if (!read_prefix(s, "function")) return string_view();
|
|
|
|
// Skip whitespace. If there isn't any, fail.
|
|
string_view w = read_space(s);
|
|
if (w.empty()) return string_view();
|
|
|
|
// Read the function name
|
|
string_view work = s;
|
|
if (read_lua_identifier(work).empty()) return string_view();
|
|
if (read_prefix(work, ".")) {
|
|
if (read_lua_identifier(work).empty()) return string_view();
|
|
}
|
|
size_t namelen = s.size() - work.size();
|
|
string_view fullname = s.substr(0, namelen);
|
|
s.remove_prefix(namelen);
|
|
|
|
// Skip whitespace
|
|
read_space(s);
|
|
|
|
// Make sure there's an open parentheses.
|
|
if (!read_prefix(s, "(")) return string_view();
|
|
|
|
// This is where we stop parsing.
|
|
return fullname;
|
|
}
|
|
|
|
string_view read_space(string_view &source)
|
|
{
|
|
size_t pos = 0;
|
|
while ((pos < source.size()) && ascii_isspace(source[pos])) pos++;
|
|
string_view result = source.substr(0, pos);
|
|
source.remove_prefix(pos);
|
|
return result;
|
|
}
|
|
|
|
string_view read_to_sep(string_view &source, char sep) {
|
|
size_t pos = source.find(sep);
|
|
string_view result;
|
|
if (pos == string_view::npos) {
|
|
result = source;
|
|
source = string_view();
|
|
} else {
|
|
result = source.substr(0, pos);
|
|
source = source.substr(pos + 1);
|
|
}
|
|
return result;
|
|
}
|
|
|
|
string_view read_to_line(string_view &source) {
|
|
size_t pos = source.find('\n');
|
|
string_view result;
|
|
if (pos == string_view::npos) {
|
|
result = source;
|
|
source = string_view();
|
|
} else {
|
|
result = source.substr(0, pos);
|
|
source = source.substr(pos + 1);
|
|
}
|
|
if ((!result.empty()) && (result.back() == '\r')) {
|
|
result.remove_suffix(1);
|
|
}
|
|
return result;
|
|
}
|
|
|
|
bool read_prefix(string_view &source, string_view prefix) {
|
|
if (0 == source.compare(0, prefix.size(), prefix)) {
|
|
source.remove_prefix(prefix.size());
|
|
return true;
|
|
} else {
|
|
return false;
|
|
}
|
|
}
|
|
|
|
string_view read_to_space(string_view &source) {
|
|
size_t pos1 = 0;
|
|
while ((pos1 < source.size()) && (!ascii_isspace(source[pos1]))) {
|
|
pos1 += 1;
|
|
}
|
|
string_view result = source.substr(0, pos1);
|
|
if (pos1 == source.size()) {
|
|
source = string_view();
|
|
return result;
|
|
}
|
|
size_t pos2 = pos1 + 1;
|
|
while ((pos2 < source.size()) && (ascii_isspace(source[pos2]))) {
|
|
pos2 += 1;
|
|
}
|
|
source = source.substr(pos2);
|
|
return result;
|
|
}
|
|
|
|
string_view read_nbytes(string_view &source, int nbytes) {
|
|
if (nbytes < 0) nbytes = 0;
|
|
if (nbytes > int(source.size())) nbytes = source.size();
|
|
string_view result = source.substr(0, nbytes);
|
|
source = source.substr(nbytes);
|
|
return result;
|
|
}
|
|
|
|
string_view read_simple_identifier(string_view &source) {
|
|
size_t len = 0;
|
|
if ((len < source.size()) && (sv::ascii_isalpha(source[len]))) {
|
|
len += 1;
|
|
while ((len < source.size()) && (sv::ascii_isalnum(source[len]))) {
|
|
len += 1;
|
|
}
|
|
}
|
|
string_view result = source.substr(0, len);
|
|
source.remove_prefix(len);
|
|
return result;
|
|
}
|
|
|
|
string_view read_lua_identifier(string_view &source) {
|
|
size_t len = 0;
|
|
if ((len < source.size()) && (sv::ascii_isualpha(source[len]))) {
|
|
len += 1;
|
|
while ((len < source.size()) && (sv::ascii_isualnum(source[len]))) {
|
|
len += 1;
|
|
}
|
|
}
|
|
string_view result = source.substr(0, len);
|
|
source.remove_prefix(len);
|
|
return result;
|
|
}
|
|
|
|
std::string_view read_number(string_view &source, bool plus, bool minus, bool dec, bool exp) {
|
|
const char *p = source.data();
|
|
const char *l = p + source.size();
|
|
if (p == l) return source.substr(0, 0);
|
|
char sign = *p;
|
|
if (sign == '+') {
|
|
if (!plus) return source.substr(0, 0);
|
|
p++;
|
|
}
|
|
if (sign == '-') {
|
|
if (!minus) return source.substr(0, 0);
|
|
p++;
|
|
}
|
|
if (p == l) return source.substr(0, 0);
|
|
bool have_digits = false;
|
|
while ((p < l) && (ascii_isdigit(*p))) {
|
|
have_digits = true;
|
|
p++;
|
|
}
|
|
if ((p < l) && dec && (*p == '.')) {
|
|
p++;
|
|
while ((p < l) && (ascii_isdigit(*p))) {
|
|
have_digits = true;
|
|
p++;
|
|
}
|
|
}
|
|
if (!have_digits) return source.substr(0, 0);
|
|
if ((p < l) && exp && ((*p == 'e')||(*p == 'E'))) {
|
|
p++;
|
|
if ((p < l) && ((*p == '+') || (*p == '-'))) {
|
|
p++;
|
|
}
|
|
bool have_exp = false;
|
|
while ((p < l) && (ascii_isdigit(*p))) {
|
|
have_exp = true;
|
|
p++;
|
|
}
|
|
if (!have_exp) return source.substr(0, 0);
|
|
}
|
|
string_view result = source.substr(0, p - source.data());
|
|
source.remove_prefix(result.size());
|
|
return result;
|
|
}
|
|
|
|
int32_t read_ascii_char(string_view &source) {
|
|
if (source.empty()) return -1;
|
|
int32_t result = source.front();
|
|
source.remove_prefix(1);
|
|
return result;
|
|
}
|
|
|
|
bool valid_number(string_view s, bool plus, bool minus, bool dec, bool exp) {
|
|
read_number(s, plus, minus, dec, exp);
|
|
return s.empty();
|
|
}
|
|
|
|
bool contains_substring_utf8(string_view haystack, string_view needle)
|
|
{
|
|
// Case sensitive is easy. Case insensitive is hard.
|
|
return haystack.find(needle) != std::string::npos;
|
|
}
|
|
|
|
using UC = UnicodeStuff<eng::string, eng::u16string, eng::u32string>;
|
|
|
|
int32_t read_codepoint_utf8(string_view &source) { return UC::read_codepoint_utf8(source); }
|
|
bool valid_utf8(string_view s) { return UC::valid_utf8(s); }
|
|
|
|
} // namespace sv
|
|
|
|
|
|
namespace util {
|
|
|
|
eng::string ascii_tolower(std::string_view s) {
|
|
eng::string mod(s);
|
|
for (int i = 0; i < int(mod.size()); i++) {
|
|
if (sv::ascii_isupper(mod[i])) {
|
|
mod[i] += 'a' - 'A';
|
|
}
|
|
}
|
|
return mod;
|
|
}
|
|
|
|
eng::string ascii_toupper(std::string_view s) {
|
|
eng::string mod(s);
|
|
for (int i = 0; i < int(mod.size()); i++) {
|
|
if (sv::ascii_islower(mod[i])) {
|
|
mod[i] += 'A' - 'a';
|
|
}
|
|
}
|
|
return mod;
|
|
}
|
|
|
|
void quote_string(const eng::string &s, std::ostream *os) {
|
|
bool anysq = false;
|
|
bool anydq = false;
|
|
for (char c : s) {
|
|
if (c == '\'') anysq = true;
|
|
if (c == '"') anydq = true;
|
|
}
|
|
bool usesinglequote = (!anysq)||(anydq);
|
|
(*os) << (usesinglequote ? '\'' : '"');
|
|
std::string_view str(s);
|
|
while (!str.empty()) {
|
|
unsigned char c0 = (unsigned char)(str[0]);
|
|
int cp = sv::read_codepoint_utf8(str);
|
|
if (cp < 0) {
|
|
(*os) << "\\" << dec.width(3).fill('0').val(c0);
|
|
str.remove_prefix(1);
|
|
} else if (cp < 32) {
|
|
c0 = ((unsigned char)cp);
|
|
switch (c0) {
|
|
case '\n': (*os) << "\\n"; break;
|
|
case '\t': (*os) << "\\t"; break;
|
|
case '\r': (*os) << "\\r"; break;
|
|
case '\b': (*os) << "\\b"; break;
|
|
default:
|
|
(*os) << "\\" << dec.width(3).fill('0').val(c0);
|
|
break;
|
|
}
|
|
} else if (cp == '"') {
|
|
(*os) << (usesinglequote ? "\"" : "\\\"");
|
|
} else if (cp == '\'') {
|
|
(*os) << (usesinglequote ? "\\'" : "'");
|
|
} else if (cp == '\\') {
|
|
(*os) << "\\\\";
|
|
} else {
|
|
write_codepoint_utf8(cp, os);
|
|
}
|
|
}
|
|
(*os) << (usesinglequote ? '\'' : '"');
|
|
}
|
|
|
|
void base64_encode(std::string_view str, std::ostream *oss) {
|
|
const char *encode_tab =
|
|
"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
|
|
const char *s = str.data();
|
|
size_t size = str.size();
|
|
for (size_t i = 0; i < size; i += 3) {
|
|
uint32_t block = ((unsigned char)(s[i])) << 16;
|
|
if (i + 1 < size) block |= ((unsigned char)(s[i + 1])) << 8;
|
|
if (i + 2 < size) block |= ((unsigned char)(s[i + 2]));
|
|
(*oss) << encode_tab[(block>>18)&0x3F];
|
|
(*oss) << encode_tab[(block>>12)&0x3F];
|
|
(*oss) << ((i + 1 < size) ? encode_tab[(block>>6)&0x3F] : '=');
|
|
(*oss) << ((i + 2 < size) ? encode_tab[(block>>0)&0x3F] : '=');
|
|
}
|
|
}
|
|
|
|
bool base64_decode(std::string_view str, std::ostream *oss) {
|
|
uint32_t chunk = 0;
|
|
int fill = 0;
|
|
int skip = 0;
|
|
bool clean = true;
|
|
for (int i = 0; i < int(str.size()); i++) {
|
|
char c = str[i];
|
|
uint32_t value;
|
|
|
|
if ((c >= 'A') && (c <= 'Z')) value = c - 'A';
|
|
else if ((c >= 'a') && (c <= 'z')) value = c - 'a' + 26;
|
|
else if ((c >= '0') && (c <= '9')) value = c - '0' + 52;
|
|
else if (c == '+') value = 62;
|
|
else if (c == '/') value = 63;
|
|
else if (c == '=') { value = 0; skip ++; }
|
|
else { clean=false; continue; }
|
|
|
|
chunk = (chunk << 6) | value;
|
|
fill ++;
|
|
if (fill == 4) {
|
|
oss->put((chunk>>16) & 0xFF);
|
|
if (skip < 2) oss->put((chunk>>8) & 0xFF);
|
|
if (skip < 1) oss->put(chunk & 0xFF);
|
|
chunk = 0; fill = 0; skip = 0;
|
|
}
|
|
}
|
|
if (fill != 0) clean = false;
|
|
return clean;
|
|
}
|
|
|
|
IdVector id_vector_create(int64_t id1, int64_t id2, int64_t id3, int64_t id4) {
|
|
IdVector result;
|
|
if (id1 >= 0) result.push_back(id1);
|
|
if (id2 >= 0) result.push_back(id2);
|
|
if (id3 >= 0) result.push_back(id3);
|
|
if (id4 >= 0) result.push_back(id4);
|
|
return result;
|
|
}
|
|
|
|
void print_id_vector(const IdVector &idv, std::ostream *os) {
|
|
bool first = true;
|
|
for (int64_t id : idv) {
|
|
if (!first) (*os) << ",";
|
|
(*os) << id;
|
|
first = false;
|
|
}
|
|
}
|
|
|
|
void print_id_vector(const std::vector<uint64_t> &idv, std::ostream *os) {
|
|
bool first = true;
|
|
for (int64_t id : idv) {
|
|
if (!first) (*os) << ",";
|
|
(*os) << id;
|
|
first = false;
|
|
}
|
|
}
|
|
|
|
eng::string id_vector_debug_string(const IdVector &idv) {
|
|
eng::ostringstream oss;
|
|
print_id_vector(idv, &oss);
|
|
return oss.str();
|
|
}
|
|
|
|
IdVector sort_union_id_vectors(const IdVector &v1, const IdVector &v2) {
|
|
IdVector result(v1.size() + v2.size());
|
|
int next = 0;
|
|
for (int64_t id : v1) result[next++] = id;
|
|
for (int64_t id : v2) result[next++] = id;
|
|
std::sort(result.begin(), result.end());
|
|
int64_t prev = -1;
|
|
int64_t count = 0;
|
|
for (int64_t id : result) {
|
|
if (id != prev) {
|
|
prev = id;
|
|
result[count++] = id;
|
|
}
|
|
}
|
|
result.resize(count);
|
|
return result;
|
|
}
|
|
|
|
HashValue hash_string(std::string_view s) {
|
|
uint64_t hash1 = 0;
|
|
uint64_t hash2 = 0;
|
|
SpookyHash::ChainHash128(s.data(), s.size(), &hash1, &hash2);
|
|
return util::HashValue(hash1, hash2);
|
|
}
|
|
|
|
HashValue hash_string(HashValue prev, std::string_view s) {
|
|
uint64_t hash1 = prev.first;
|
|
uint64_t hash2 = prev.second;
|
|
SpookyHash::ChainHash128(s.data(), s.size(), &hash1, &hash2);
|
|
return util::HashValue(hash1, hash2);
|
|
}
|
|
|
|
HashValue hash_id_vector(const IdVector &idv) {
|
|
uint64_t hash1 = 0;
|
|
uint64_t hash2 = 0;
|
|
SpookyHash::ChainHash128(&idv[0], idv.size() * sizeof(int64_t), &hash1, &hash2);
|
|
return util::HashValue(hash1, hash2);
|
|
}
|
|
|
|
eng::string hash_to_hex(const HashValue &hv) {
|
|
eng::ostringstream oss;
|
|
oss << hex64.val(hv.first) << hex64.val(hv.second);
|
|
return oss.str();
|
|
}
|
|
static inline uint64_t Rot64(uint64_t x, int k)
|
|
{
|
|
return (x << k) | (x >> (64 - k));
|
|
}
|
|
|
|
uint64_t hash_ints(uint64_t a, uint64_t b, uint64_t c, uint64_t d) {
|
|
uint64_t h0 = c ^ 0xc548cebf3714dbb9;
|
|
uint64_t h1 = d ^ 0xd23a7edd44383f8d;
|
|
uint64_t h2 = a ^ 0x7356f92e4b154df7;
|
|
uint64_t h3 = b ^ 0x55ce09295766838d;
|
|
|
|
h3 ^= h2; h2 = Rot64(h2,15); h3 += h2;
|
|
h0 ^= h3; h3 = Rot64(h3,52); h0 += h3;
|
|
h1 ^= h0; h0 = Rot64(h0,26); h1 += h0;
|
|
h2 ^= h1; h1 = Rot64(h1,51); h2 += h1;
|
|
h3 ^= h2; h2 = Rot64(h2,28); h3 += h2;
|
|
h0 ^= h3; h3 = Rot64(h3,9); h0 += h3;
|
|
h1 ^= h0; h0 = Rot64(h0,47); h1 += h0;
|
|
h2 ^= h1; h1 = Rot64(h1,54); h2 += h1;
|
|
h3 ^= h2; h2 = Rot64(h2,32); h3 += h2;
|
|
h0 ^= h3; h3 = Rot64(h3,25); h0 += h3;
|
|
h1 ^= h0; h0 = Rot64(h0,63); h1 += h0;
|
|
|
|
return h1;
|
|
}
|
|
|
|
double hash_to_double(uint64_t hash) {
|
|
return (hash >> (64-53)) * 0x1p-53;
|
|
}
|
|
|
|
StringVec split(const eng::string &s, char sep) {
|
|
StringVec result;
|
|
int start = 0;
|
|
for (int i = 0; i < int(s.size()); i++) {
|
|
if (s[i] == sep) {
|
|
result.push_back(s.substr(start, i-start));
|
|
start = i+1;
|
|
}
|
|
}
|
|
if (start < int(s.size())) {
|
|
result.push_back(s.substr(start));
|
|
}
|
|
return result;
|
|
}
|
|
|
|
static eng::string substr_nocr(const eng::string &s, int start, int len) {
|
|
if ((len > 0) && (s[start + len - 1] == '\r')) {
|
|
len -= 1;
|
|
}
|
|
return s.substr(start, len);
|
|
}
|
|
|
|
StringVec split_lines(const eng::string &s) {
|
|
StringVec result;
|
|
int start = 0;
|
|
for (int i = 0; i < int(s.size()); i++) {
|
|
if (s[i]=='\n') {
|
|
result.push_back(substr_nocr(s, start, i-start));
|
|
start = i + 1;
|
|
}
|
|
}
|
|
if (start < int(s.size())) {
|
|
result.push_back(substr_nocr(s, start, s.size()-start));
|
|
}
|
|
return result;
|
|
}
|
|
|
|
StringVec split_docstring(const eng::string &s) {
|
|
StringVec result;
|
|
int start = 0;
|
|
for (int i = 0; i < int(s.size()); i++) {
|
|
if (s[i]=='|') {
|
|
int len = i-start;
|
|
if ((len > 0)||(start > 0)) {
|
|
result.push_back(s.substr(start, i-start));
|
|
}
|
|
start = i + 1;
|
|
}
|
|
}
|
|
if (start < int(s.size())) {
|
|
result.push_back(s.substr(start, s.size()-start));
|
|
}
|
|
return result;
|
|
}
|
|
|
|
eng::string join(const StringVec &strs, const eng::string &sep) {
|
|
if (strs.empty()) return "";
|
|
eng::ostringstream oss;
|
|
oss << strs[0];
|
|
for (int i = 1; i < int(strs.size()); i++) {
|
|
oss << sep << strs[i];
|
|
}
|
|
return oss.str();
|
|
}
|
|
|
|
eng::string repeat_string(const eng::string &a, int n) {
|
|
int len = a.size();
|
|
eng::string result(len * n, ' ');
|
|
for (int i = 0; i < n; i++) {
|
|
for (int j = 0; j < len; j++) {
|
|
result[i*len + j] = a[j];
|
|
}
|
|
}
|
|
return result;
|
|
}
|
|
|
|
eng::string tolower(eng::string input) {
|
|
for (int i = 0; i < int(input.size()); i++) {
|
|
input[i] = std::tolower(input[i]);
|
|
}
|
|
return input;
|
|
}
|
|
|
|
eng::string toupper(eng::string input) {
|
|
for (int i = 0; i < int(input.size()); i++) {
|
|
input[i] = std::toupper(input[i]);
|
|
}
|
|
return input;
|
|
}
|
|
|
|
static int buffer_codepoint_utf8(char32_t scp, char *buffer) {
|
|
uint32_t cp = (uint32_t)scp;
|
|
unsigned char *c = (unsigned char *)buffer;
|
|
if (cp < 0) {
|
|
return 0;
|
|
}
|
|
else if (cp <= 0x7F) {
|
|
c[0] = cp;
|
|
return 1;
|
|
}
|
|
else if (cp <= 0x7FF) {
|
|
c[0] = (cp>>6)+192;
|
|
c[1] = (cp&63)+128;
|
|
return 2;
|
|
}
|
|
else if (cp <= 0xFFFF) {
|
|
if ((cp >= 0xD800) && (cp <= 0xDFFF)) {
|
|
return 0;
|
|
}
|
|
c[0] = (cp>>12)+224;
|
|
c[1] = ((cp>>6)&63)+128;
|
|
c[2] = (cp&63)+128;
|
|
return 3;
|
|
}
|
|
else if (cp <= 0x10FFFF) {
|
|
c[0] = (cp>>18)+240;
|
|
c[1] = ((cp>>12)&63)+128;
|
|
c[2] = ((cp>>6)&63)+128;
|
|
c[3] = (cp&63)+128;
|
|
return 4;
|
|
} else {
|
|
return 0;
|
|
}
|
|
}
|
|
|
|
eng::string get_codepoint_utf8(uint32_t cp) {
|
|
char buffer[4];
|
|
int len = buffer_codepoint_utf8(cp, buffer);
|
|
return eng::string(buffer, len);
|
|
}
|
|
|
|
bool write_codepoint_utf8(int32_t cp, std::ostream *s) {
|
|
char buffer[4];
|
|
int len = buffer_codepoint_utf8(cp, buffer);
|
|
(*s) << std::string_view(buffer, len);
|
|
return (len > 0);
|
|
}
|
|
|
|
double distance_squared(double x1, double y1, double x2, double y2) {
|
|
double dx = x1 - x2;
|
|
double dy = y1 - y2;
|
|
return dx*dx + dy*dy;
|
|
}
|
|
|
|
LuaSourcePtr make_lua_source(const eng::string &code) {
|
|
LuaSourcePtr result(new LuaSourceVec);
|
|
eng::string fn = "file.lua";
|
|
result->push_back(std::make_pair(fn, code));
|
|
return result;
|
|
}
|
|
|
|
|
|
|
|
void (*dprint_hook)(const char *oneline, size_t size);
|
|
|
|
void hook_dprint(void (*func)(const char *oneline, size_t size)) {
|
|
dprint_hook = func;
|
|
}
|
|
|
|
void dprintview(std::string_view view) {
|
|
// Drop the final newline, if any. We're going
|
|
// to automatically end with a newline and we don't
|
|
// want to double up.
|
|
if ((view.size() > 0) && (view.back() == '\n')) {
|
|
view.remove_suffix(1);
|
|
}
|
|
|
|
// Chop it up into lines and call the hook one line
|
|
// at a time.
|
|
const char *buffer = view.data();
|
|
const char *base = buffer;
|
|
for (int i = 0; i < int(view.size()); i++) {
|
|
if ((buffer[i] == '\n') || (buffer[i] == 0)) {
|
|
size_t sz = buffer + i - base;
|
|
if (dprint_hook == nullptr) {
|
|
fwrite(base, 1, sz, stderr);
|
|
fwrite("\n", 1, 1, stderr);
|
|
} else {
|
|
dprint_hook(base, sz);
|
|
}
|
|
base = buffer + i + 1;
|
|
}
|
|
}
|
|
|
|
// Output the final line.
|
|
size_t sz = buffer + view.size() - base;
|
|
if (sz > 0) {
|
|
if (dprint_hook == nullptr) {
|
|
fwrite(base, 1, sz, stderr);
|
|
fwrite("\n", 1, 1, stderr);
|
|
} else {
|
|
dprint_hook(base, sz);
|
|
}
|
|
}
|
|
|
|
// If we're sending to stderr, flush. If not, then
|
|
// the hook routine is responsible for flushing its own
|
|
// output.
|
|
if (dprint_hook == nullptr) {
|
|
fflush(stderr);
|
|
}
|
|
}
|
|
|
|
void dprintf(const char *format, ...) {
|
|
char buffer[256];
|
|
va_list args;
|
|
va_start (args, format);
|
|
int n = vsnprintf(buffer, 256, format, args);
|
|
if (n <= 255) {
|
|
dprintview(std::string_view(buffer, n));
|
|
} else {
|
|
char *lbuffer = (char *)malloc(n + 1);
|
|
vsnprintf(lbuffer, n+1, format, args);
|
|
dprintview(std::string_view(lbuffer, n));
|
|
free(lbuffer);
|
|
}
|
|
}
|
|
|
|
} // namespace util
|
|
|
|
|
|
static std::string_view read_number_x(const char *p, bool plus, bool minus, bool dec, bool exp) {
|
|
std::string_view source = p;
|
|
return sv::read_number(source, plus, minus, dec, exp);
|
|
}
|
|
|
|
eng::string util::decode_token(uint64_t value) {
|
|
static const char encoding[] =
|
|
"\0_0123456789abcdefghijklmnopqrstuvwxyz";
|
|
uint64_t n = value;
|
|
char buffer[13] = {};
|
|
for (int i = 11; i >= 0; i--) {
|
|
buffer[i] = encoding[n % 38];
|
|
n /= 38;
|
|
}
|
|
return eng::string(buffer);
|
|
}
|
|
|
|
LuaDefine(unittests_util, "", "some unit tests") {
|
|
// Test valid_hostname
|
|
LuaAssert(L, sv::valid_hostname("foo123"));
|
|
LuaAssert(L, !sv::valid_hostname("foo%123"));
|
|
LuaAssert(L, sv::valid_hostname("foo-bar"));
|
|
LuaAssert(L, !sv::valid_hostname("-foo"));
|
|
LuaAssert(L, !sv::valid_hostname("foo-"));
|
|
LuaAssert(L, sv::valid_hostname("foo.bar.baz"));
|
|
LuaAssert(L, sv::valid_hostname("foo-bar.baz"));
|
|
LuaAssert(L, !sv::valid_hostname("foo-bar-.baz"));
|
|
LuaAssert(L, !sv::valid_hostname("foo.-bar-baz"));
|
|
|
|
// test str_to_int64, str_to_double
|
|
LuaAssert(L, sv::to_int64("123") == 123);
|
|
LuaAssert(L, sv::to_int64("123.4") == INT64_MAX);
|
|
LuaAssert(L, sv::to_int64("12ab") == INT64_MAX);
|
|
LuaAssert(L, sv::to_int64("") == INT64_MAX);
|
|
LuaAssert(L, sv::to_double("123.5") == 123.5);
|
|
LuaAssert(L, std::isnan(sv::to_double("12ab")));
|
|
LuaAssert(L, std::isnan(sv::to_double("")));
|
|
|
|
// Test trim, ltrim, rtrim
|
|
LuaAssert(L, sv::ltrim(" foo ") == "foo ");
|
|
LuaAssert(L, sv::rtrim(" foo ") == " foo");
|
|
LuaAssert(L, sv::trim(" foo ") == "foo");
|
|
LuaAssert(L, sv::trim("foo") == "foo");
|
|
LuaAssert(L, sv::trim("") == "");
|
|
LuaAssert(L, sv::ltrim("**foo**", '*') == "foo**");
|
|
LuaAssert(L, sv::rtrim("**foo**", '*') == "**foo");
|
|
LuaAssert(L, sv::trim("**foo**", '*') == "foo");
|
|
LuaAssert(L, sv::trim("foo", '*') == "foo");
|
|
LuaAssert(L, sv::trim("", '*') == "");
|
|
|
|
// Test read_to_line
|
|
std::string_view v = "foo\nbar\r\nbaz";
|
|
|
|
std::string_view v1 = sv::read_to_line(v);
|
|
LuaAssertStrEq(L, v1, "foo");
|
|
LuaAssertStrEq(L, v, "bar\r\nbaz");
|
|
|
|
std::string_view v2 = sv::read_to_line(v);
|
|
LuaAssertStrEq(L, v2, "bar");
|
|
LuaAssertStrEq(L, v, "baz");
|
|
|
|
std::string_view v3 = sv::read_to_line(v);
|
|
LuaAssertStrEq(L, v3, "baz");
|
|
LuaAssert(L, sv::isnull(v));
|
|
|
|
// Test read_to_space
|
|
v = "foo bar baz";
|
|
|
|
std::string_view s1 = sv::read_to_space(v);
|
|
LuaAssertStrEq(L, s1, "foo");
|
|
LuaAssertStrEq(L, v, "bar baz");
|
|
|
|
std::string_view s2 = sv::read_to_space(v);
|
|
LuaAssertStrEq(L, s2, "bar");
|
|
LuaAssertStrEq(L, v, "baz");
|
|
|
|
std::string_view s3 = sv::read_to_space(v);
|
|
LuaAssertStrEq(L, s3, "baz");
|
|
LuaAssert(L, sv::isnull(v));
|
|
|
|
// Test the unioning of ID vectors.
|
|
util::IdVector idv1,idv2;
|
|
idv1.push_back(1);
|
|
idv1.push_back(6);
|
|
idv1.push_back(4);
|
|
idv2.push_back(5);
|
|
idv2.push_back(1);
|
|
idv2.push_back(6);
|
|
util::IdVector joined = util::sort_union_id_vectors(idv1, idv2);
|
|
LuaAssert(L, joined.size() == 4);
|
|
LuaAssert(L, joined[0] == 1);
|
|
LuaAssert(L, joined[1] == 4);
|
|
LuaAssert(L, joined[2] == 5);
|
|
LuaAssert(L, joined[3] == 6);
|
|
|
|
// Test the string split routine.
|
|
util::StringVec sv1 = util::split("foo,bar,baz", ',');
|
|
LuaAssert(L, sv1.size() == 3);
|
|
LuaAssert(L, sv1[0] == "foo");
|
|
LuaAssert(L, sv1[1] == "bar");
|
|
LuaAssert(L, sv1[2] == "baz");
|
|
util::StringVec sv2 = util::split(",foo,,bar", ',');
|
|
LuaAssert(L, sv2.size() == 4);
|
|
LuaAssert(L, sv2[0]=="");
|
|
LuaAssert(L, sv2[1]=="foo");
|
|
LuaAssert(L, sv2[2]=="");
|
|
LuaAssert(L, sv2[3]=="bar");
|
|
|
|
// Test the split_lines routine.
|
|
util::StringVec sv3 = util::split_lines("foo\n\nbar\r\nbaz\r\n\r\n");
|
|
LuaAssert(L, sv3.size() == 5);
|
|
LuaAssert(L, sv3[0] == "foo");
|
|
LuaAssert(L, sv3[1] == "");
|
|
LuaAssert(L, sv3[2] == "bar");
|
|
LuaAssert(L, sv3[3] == "baz");
|
|
LuaAssert(L, sv3[4] == "");
|
|
|
|
// Test the repeat string routine.
|
|
LuaAssertStrEq(L, util::repeat_string("abc", 3), "abcabcabc");
|
|
|
|
// test toupper and tolower
|
|
LuaAssert(L, util::toupper("fooBar") == "FOOBAR");
|
|
LuaAssert(L, util::tolower("fooBar") == "foobar");
|
|
|
|
// Test distance_squared
|
|
LuaAssert(L, util::distance_squared(1, 1, 5, 4) == 25.0);
|
|
LuaAssert(L, util::distance_squared(5, 4, 1, 1) == 25.0);
|
|
|
|
// Test XYZ.
|
|
util::XYZ xyza(3,4,5), xyzb(3,4,5), xyzc(3,4,6);
|
|
LuaAssert(L, xyza.x == 3);
|
|
LuaAssert(L, xyza.y == 4);
|
|
LuaAssert(L, xyza.z == 5);
|
|
LuaAssert(L, xyza == xyzb);
|
|
LuaAssert(L, xyza != xyzc);
|
|
LuaAssert(L, xyza.debug_string() == "(3,4,5)");
|
|
|
|
// Test hash_to_string
|
|
LuaAssertStrEq(L, util::hash_to_hex(util::HashValue(0x1234,0x789a)),
|
|
"0000000000001234000000000000789a");
|
|
|
|
// Test hash_to_double
|
|
LuaAssert(L, util::hash_to_double(0x1000000000000000) == 1.0/16.0);
|
|
LuaAssert(L, util::hash_to_double(0x7000000000000000) == 7.0/16.0);
|
|
LuaAssert(L, util::hash_to_double(0xF000000000000000) == 15.0/16.0);
|
|
|
|
// Test read_number allowing everything.
|
|
LuaAssert(L, read_number_x("123x", true, true, true, true) == "123");
|
|
LuaAssert(L, read_number_x("123.3x", true, true, true, true) == "123.3");
|
|
LuaAssert(L, read_number_x("123.x", true, true, true, true) == "123.");
|
|
LuaAssert(L, read_number_x("123..x", true, true, true, true) == "123.");
|
|
LuaAssert(L, read_number_x("-123x", true, true, true, true) == "-123");
|
|
LuaAssert(L, read_number_x("+123x", true, true, true, true) == "+123");
|
|
LuaAssert(L, read_number_x("+-123x", true, true, true, true) == "");
|
|
LuaAssert(L, read_number_x("-123.02e05x", true, true, true, true) == "-123.02e05");
|
|
LuaAssert(L, read_number_x("-123e-5x", true, true, true, true) == "-123e-5");
|
|
LuaAssert(L, read_number_x("-123e+5x", true, true, true, true) == "-123e+5");
|
|
LuaAssert(L, read_number_x("-123e+x", true, true, true, true) == "");
|
|
|
|
// Test read_codepoint_utf8.
|
|
std::string_view str("𝞮ὥπq");
|
|
LuaAssert(L, str.size() == 10);
|
|
LuaAssert(L, sv::read_codepoint_utf8(str) == 0x1D7AE); // 4-byte char
|
|
LuaAssert(L, str.size() == 6);
|
|
LuaAssert(L, sv::read_codepoint_utf8(str) == 0x1F65); // 3-byte char
|
|
LuaAssert(L, str.size() == 3);
|
|
LuaAssert(L, sv::read_codepoint_utf8(str) == 0x3C0); // 2-byte char
|
|
LuaAssert(L, str.size() == 1);
|
|
LuaAssert(L, sv::read_codepoint_utf8(str) == 0x71); // 1-byte char
|
|
LuaAssert(L, str.size() == 0);
|
|
LuaAssert(L, sv::read_codepoint_utf8(str) == -1); // EOF
|
|
|
|
// Test read_codepoint_utf8 on an invalid unicode sequence.
|
|
std::string_view strbad("\xC0\xC0");
|
|
LuaAssert(L, sv::read_codepoint_utf8(strbad) == -2);
|
|
|
|
return 0;
|
|
}
|
|
|