integration/luprex/core/cpp/util.cpp

#include "wrap-string.hpp"
#include "wrap-vector.hpp"
#include "util.hpp"
#include "fast-float.hpp"


#include <algorithm>
#include <sys/types.h>
#include <sys/stat.h>
#include <iomanip>
#include <cassert>
#include <cstdlib>
#include <cmath>
#include <charconv>


#ifdef WIN32
#endif

#ifndef WIN32
#include <time.h>
#include <unistd.h>
#endif

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;
}

int64_t to_int64(string_view value, int64_t errval) {
    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 errval;
    if (r.ptr != last) return errval;
    return result;
}

uint64_t to_hex64(string_view value, uint64_t errval) {
    uint64_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 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_isalpha(c)) && (c!='_')) return false;
    for (int i = 1; i < int(str.size()); i++) {
        char c = str[i];
        if ((!ascii_isalpha(c)) && (!ascii_isdigit(c)) && (c!='_')) return false;
    }
    return true;
}

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) == "--";
}

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;
}

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_ascii_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;
}

bool valid_utf8(string_view s)
{
    const unsigned char *bytes = (const unsigned char *)s.data();
    const unsigned char *tail = bytes + s.size();
    unsigned int codepoint;
    int seqlen;

    while (bytes < tail) {
        if ((bytes[0] & 0x80) == 0x00) {
            // U+0000 to U+007F
            codepoint = (bytes[0] & 0x7F);
            seqlen = 1;
        } else if ((bytes[0] & 0xE0) == 0xC0) {
            // U+0080 to U+07FF
            codepoint = (bytes[0] & 0x1F);
            seqlen = 2;
        } else if ((bytes[0] & 0xF0) == 0xE0) {
            // U+0800 to U+FFFF
            codepoint = (bytes[0] & 0x0F);
            seqlen = 3;
        } else if ((bytes[0] & 0xF8) == 0xF0) {
            // U+10000 to U+10FFFF
            codepoint = (bytes[0] & 0x07);
            seqlen = 4;
        } else {
            return false;
        }

        if (bytes + seqlen > tail) {
            return false;
        }

        for (int i = 1; i < seqlen; ++i) {
            if ((bytes[i] & 0xC0) != 0x80) return false;
            codepoint = (codepoint << 6) | (bytes[i] & 0x3F);
        }

        if ((codepoint > 0x10FFFF) ||
            ((codepoint >= 0xD800) && (codepoint <= 0xDFFF)) ||
            ((codepoint <= 0x007F) && (seqlen != 1)) ||
            ((codepoint >= 0x0080) && (codepoint <= 0x07FF) && (seqlen != 2)) ||
            ((codepoint >= 0x0800) && (codepoint <= 0xFFFF) && (seqlen != 3)) ||
            ((codepoint >= 0x10000) && (codepoint <= 0x1FFFFF) && (seqlen != 4))) {
            return false;
        }

        bytes += seqlen;
    }
    return true;
}

} // 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 ? '\'' : '"');
    for (char c : s) {
        if (c >= 32) {
            if (c == '"') {
                (*os) << (usesinglequote ? "\"" : "\\\"");
            } else if (c == '\'') {
                (*os) << (usesinglequote ? "\\'" : "'");
            } else {
                (*os) << c;
            }
        } else {
            unsigned int value = ((unsigned char)c);
            switch (c) {
            case '\n': (*os) << "\\n"; break;
            case '\t': (*os) << "\\t"; break;
            case '\r': (*os) << "\\r"; break;
            default:
                (*os) << "\\" << std::setfill('0') << std::setw(3) << value;
                break;
            }
        }
    }
    (*os) << (usesinglequote ? '\'' : '"');
}

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;
}

eng::string id_vector_debug_string(const IdVector &idv) {
    eng::ostringstream oss;
    bool first = true;
    for (int64_t id : idv) {
        if (!first) oss << ",";
        oss << id;
        first = false;
    }
    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(const eng::string &s) {
    uint64_t hash1 = 0;
    uint64_t hash2 = 0;
    SpookyHash::ChainHash128(s.c_str(), 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 << std::hex << std::setw(16) << std::setfill('0') << hv.first;
    oss << std::hex << std::setw(16) << std::setfill('0') << 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;
}

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;
}

eng::string XYZ::debug_string() const {
    eng::ostringstream oss;
    oss << "(" << x << "," << y << "," << z << ")";
    return oss.str();
}


} // namespace util

std::ostream &operator<<(std::ostream &oss, const util::hex64 &v) {
    oss << "0x" << std::setw(16) << std::setfill('0') << std::hex;
    return oss;
}

std::ostream &operator<<(std::ostream &oss, const util::hex32 &v) {
    oss << "0x" << std::setw(8) << std::setfill('0') << std::hex;
    return oss;
}

std::ostream &operator<<(std::ostream &oss, const util::hex16 &v) {
    oss << "0x" << std::setw(4) << std::setfill('0') << std::hex;
    return oss;
}

std::ostream &operator<<(std::ostream &oss, const util::hex8 &v) {
    oss << "0x" << std::setw(2) << std::setfill('0') << std::hex;
    return oss;
}

LuaDefine(unittests_util, "", "some unit tests") {
        // test str_to_int64, str_to_double
    LuaAssert(L, sv::to_int64("123") == 123);
    LuaAssert(L, sv::to_int64("123.4") == INT64_MIN);
    LuaAssert(L, sv::to_int64("12ab") == INT64_MIN);
    LuaAssert(L, sv::to_int64("") == INT64_MIN);
    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);
    return 0;
}