D4544.1767344557.diff
No OneTemporary
Actions

Size

1 KB

Referenced Files

None

Subscribers

None

D4544.1767344557.diff
View Options

	diff --git a/services/backup/src/Tools.cpp b/services/backup/src/Tools.cpp
	--- a/services/backup/src/Tools.cpp
	+++ b/services/backup/src/Tools.cpp
	@@ -3,6 +3,7 @@
	#include "GlobalConstants.h"
	#include "GlobalTools.h"

	+#include <algorithm>
	#include <chrono>
	#include <cstdlib>
	#include <random>
	@@ -53,19 +54,19 @@
	}

	std::size_t getUtf8Length(std::string &str) {
	- std::size_t result = 0;
	- const char *ptr = str.data();
	- const char *end = ptr + str.size();

	- while (ptr < end) {
	- int next = std::mblen(ptr, end - ptr);
	- if (next == -1) {
	- throw std::runtime_error("strlen_mb(): conversionn error");
	- }
	- ptr += next;
	- ++result;
	- }
	- return result;
	+ // How it works?
	+ // Every byte of multibyte utf8 character starts with information about its
	+ // own type. Additional bytes always start with a 2bit sequence "10xx xxxx"
	+ // (which is equal to hex 0x80). To count the length of utf8 string we need
	+ // to skip these bytes, so we use a bitmask.
	+ //
	+ // We need to compare first 2 bits of a byte, so we use bitwise operator
	+ // & with mask 0xC0. If returned value is equal to 0x80, we skip it.
	+
	+ return std::count_if(str.begin(), str.end(), [](char c) {
	+ return (static_cast<unsigned char>(c) & 0xC0) != 0x80;
	+ });
	}

	} // namespace tools