| 1 | #include "duckdb/common/string_util.hpp" |
| 2 | |
| 3 | #include <algorithm> |
| 4 | #include <cctype> |
| 5 | #include <iomanip> |
| 6 | #include <memory> |
| 7 | #include <sstream> |
| 8 | #include <stdarg.h> |
| 9 | #include <string.h> |
| 10 | #include <string> |
| 11 | |
| 12 | using namespace duckdb; |
| 13 | using namespace std; |
| 14 | |
| 15 | bool StringUtil::Contains(const string &haystack, const string &needle) { |
| 16 | return (haystack.find(needle) != string::npos); |
| 17 | } |
| 18 | |
| 19 | void StringUtil::LTrim(string &str) { |
| 20 | auto it = str.begin(); |
| 21 | while (isspace(*it)) { |
| 22 | it++; |
| 23 | } |
| 24 | str.erase(str.begin(), it); |
| 25 | } |
| 26 | |
| 27 | // Remove trailing ' ', '\f', '\n', '\r', '\t', '\v' |
| 28 | void StringUtil::RTrim(string &str) { |
| 29 | str.erase(find_if(str.rbegin(), str.rend(), [](int ch) { return ch > 0 && !isspace(ch); }).base(), str.end()); |
| 30 | } |
| 31 | |
| 32 | void StringUtil::Trim(string &str) { |
| 33 | StringUtil::LTrim(str); |
| 34 | StringUtil::RTrim(str); |
| 35 | } |
| 36 | |
| 37 | bool StringUtil::StartsWith(const string &str, const string &prefix) { |
| 38 | if (prefix.size() > str.size()) { |
| 39 | return false; |
| 40 | } |
| 41 | return equal(prefix.begin(), prefix.end(), str.begin()); |
| 42 | } |
| 43 | |
| 44 | bool StringUtil::EndsWith(const string &str, const string &suffix) { |
| 45 | if (suffix.size() > str.size()) |
| 46 | return false; |
| 47 | return equal(suffix.rbegin(), suffix.rend(), str.rbegin()); |
| 48 | } |
| 49 | |
| 50 | string StringUtil::Repeat(const string &str, idx_t n) { |
| 51 | ostringstream os; |
| 52 | if (n == 0 || str.empty()) { |
| 53 | return (os.str()); |
| 54 | } |
| 55 | for (int i = 0; i < static_cast<int>(n); i++) { |
| 56 | os << str; |
| 57 | } |
| 58 | return (os.str()); |
| 59 | } |
| 60 | |
| 61 | vector<string> StringUtil::Split(const string &str, char delimiter) { |
| 62 | stringstream ss(str); |
| 63 | vector<string> lines; |
| 64 | string temp; |
| 65 | while (getline(ss, temp, delimiter)) { |
| 66 | lines.push_back(temp); |
| 67 | } // WHILE |
| 68 | return (lines); |
| 69 | } |
| 70 | |
| 71 | string StringUtil::Join(const vector<string> &input, const string &separator) { |
| 72 | return StringUtil::Join(input, input.size(), separator, [](const string &s) { return s; }); |
| 73 | } |
| 74 | |
| 75 | string StringUtil::Prefix(const string &str, const string &prefix) { |
| 76 | vector<string> lines = StringUtil::Split(str, '\n'); |
| 77 | if (lines.empty()) |
| 78 | return ("" ); |
| 79 | |
| 80 | ostringstream os; |
| 81 | for (idx_t i = 0, cnt = lines.size(); i < cnt; i++) { |
| 82 | if (i > 0) |
| 83 | os << endl; |
| 84 | os << prefix << lines[i]; |
| 85 | } // FOR |
| 86 | return (os.str()); |
| 87 | } |
| 88 | |
| 89 | // http://ubuntuforums.org/showpost.php?p=10215516&postcount=5 |
| 90 | string StringUtil::FormatSize(idx_t bytes) { |
| 91 | double BASE = 1024; |
| 92 | double KB = BASE; |
| 93 | double MB = KB * BASE; |
| 94 | double GB = MB * BASE; |
| 95 | |
| 96 | ostringstream os; |
| 97 | |
| 98 | if (bytes >= GB) { |
| 99 | os << fixed << setprecision(2) << (bytes / GB) << " GB" ; |
| 100 | } else if (bytes >= MB) { |
| 101 | os << fixed << setprecision(2) << (bytes / MB) << " MB" ; |
| 102 | } else if (bytes >= KB) { |
| 103 | os << fixed << setprecision(2) << (bytes / KB) << " KB" ; |
| 104 | } else { |
| 105 | os << to_string(bytes) + " bytes" ; |
| 106 | } |
| 107 | return (os.str()); |
| 108 | } |
| 109 | |
| 110 | string StringUtil::Upper(const string &str) { |
| 111 | string copy(str); |
| 112 | transform(copy.begin(), copy.end(), copy.begin(), [](unsigned char c) { return toupper(c); }); |
| 113 | return (copy); |
| 114 | } |
| 115 | |
| 116 | string StringUtil::Lower(const string &str) { |
| 117 | string copy(str); |
| 118 | transform(copy.begin(), copy.end(), copy.begin(), [](unsigned char c) { return tolower(c); }); |
| 119 | return (copy); |
| 120 | } |
| 121 | |
| 122 | // http://stackoverflow.com/a/8098080 |
| 123 | string StringUtil::Format(const string fmt_str, ...) { |
| 124 | // Reserve two times as much as the length of the fmt_str |
| 125 | int final_n, n = ((int)fmt_str.size()) * 2; |
| 126 | string str; |
| 127 | unique_ptr<char[]> formatted; |
| 128 | va_list ap; |
| 129 | |
| 130 | while (1) { |
| 131 | // Wrap the plain char array into the unique_ptr |
| 132 | formatted.reset(new char[n + 1]); |
| 133 | strcpy(&formatted[0], fmt_str.c_str()); |
| 134 | va_start(ap, fmt_str); |
| 135 | final_n = vsnprintf(&formatted[0], n, fmt_str.c_str(), ap); |
| 136 | va_end(ap); |
| 137 | if (final_n < 0 || final_n >= n) |
| 138 | n += abs(final_n - n + 1); |
| 139 | else |
| 140 | break; |
| 141 | } |
| 142 | return string(formatted.get()); |
| 143 | } |
| 144 | |
| 145 | string StringUtil::VFormat(const string fmt_str, va_list args) { |
| 146 | va_list args_copy; |
| 147 | |
| 148 | unique_ptr<char[]> formatted; |
| 149 | // make a copy of the args as we can only use it once |
| 150 | va_copy(args_copy, args); |
| 151 | |
| 152 | // first get the amount of characters we need |
| 153 | const auto n = vsnprintf(nullptr, 0, fmt_str.c_str(), args) + 1; |
| 154 | |
| 155 | // now allocate the string and do the actual printing |
| 156 | formatted.reset(new char[n]); |
| 157 | (void)vsnprintf(&formatted[0], n, fmt_str.c_str(), args_copy); |
| 158 | return string(formatted.get()); |
| 159 | } |
| 160 | |
| 161 | vector<string> StringUtil::Split(const string &input, const string &split) { |
| 162 | vector<string> splits; |
| 163 | |
| 164 | idx_t last = 0; |
| 165 | idx_t input_len = input.size(); |
| 166 | idx_t split_len = split.size(); |
| 167 | while (last <= input_len) { |
| 168 | idx_t next = input.find(split, last); |
| 169 | if (next == string::npos) { |
| 170 | next = input_len; |
| 171 | } |
| 172 | |
| 173 | // Push the substring [last, next) on to splits |
| 174 | string substr = input.substr(last, next - last); |
| 175 | if (substr.empty() == false) { |
| 176 | splits.push_back(substr); |
| 177 | } |
| 178 | last = next + split_len; |
| 179 | } |
| 180 | return splits; |
| 181 | } |
| 182 | |
| 183 | string StringUtil::Replace(string source, const string &from, const string &to) { |
| 184 | if (from.empty()) |
| 185 | return source; |
| 186 | ; |
| 187 | idx_t start_pos = 0; |
| 188 | while ((start_pos = source.find(from, start_pos)) != string::npos) { |
| 189 | source.replace(start_pos, from.length(), to); |
| 190 | start_pos += to.length(); // In case 'to' contains 'from', like |
| 191 | // replacing 'x' with 'yx' |
| 192 | } |
| 193 | return source; |
| 194 | } |
| 195 | |