nerdcruft/libcruft-util
1
0
Fork 0
Code Issues Releases Activity

utf8: use mask test object for continuation ops

Browse Source
This commit is contained in:
Danny Robson 2017-10-02 15:40:54 +11:00
parent 45e2477e1b
commit 5e1fab3590
1 changed files with 5 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
utf8.cpp
View File

@ -111,15 +111,16 @@ util::utf8::decode (view<const std::byte*> src)
codepoint_t shift = 0; codepoint_t shift = 0;
// check every following data byte has the appropriate prefix // check every following data byte has the appropriate prefix
constexpr auto CONTINUATION = "0b10xxxxxx"_test;
for (int i = 1; i <= len; ++i) { for (int i = 1; i <= len; ++i) {
if ((std::to_integer<codepoint_t> (cursor[i]) & 0b11'000000u) != 0b10'000000u) if (!CONTINUATION.valid (std::to_integer<codepoint_t> (cursor[i])))
throw malformed_error {}; throw malformed_error {};
} }
switch (len) { switch (len) {
case 3: accum |= (std::to_integer<codepoint_t> (cursor[3]) & 0b00111111u) << (shift++ * 6u); case 3: accum |= CONTINUATION.value (std::to_integer<codepoint_t> (cursor[3])) << (shift++ * 6u);
case 2: accum |= (std::to_integer<codepoint_t> (cursor[2]) & 0b00111111u) << (shift++ * 6u); case 2: accum |= CONTINUATION.value (std::to_integer<codepoint_t> (cursor[2])) << (shift++ * 6u);
case 1: accum |= (std::to_integer<codepoint_t> (cursor[1]) & 0b00111111u) << (shift++ * 6u); case 1: accum |= CONTINUATION.value (std::to_integer<codepoint_t> (cursor[1])) << (shift++ * 6u);
} }
// describes the bits required to be present for a valid minimally // describes the bits required to be present for a valid minimally