Amebis/stdex
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

html: support for char16_t

Browse Source 
This allows processing UTF-16 HTML documents on non-Windows platforms.

Signed-off-by: Simon Rozman <simon@rozman.si>
This commit is contained in:
Simon Rozman
2025-09-11 15:10:50 +02:00
parent 0e091003b6
commit 4611976a42
1 changed files with 134 additions and 64 deletions
Download Patch File Download Diff File Expand all files Collapse all files

198
include/stdex/html.hpp
View File

@@ -50,8 +50,9 @@ namespace stdex
_Inout_ std::basic_string<char, TR, AX>& dst,
_In_reads_or_z_opt_(num_chars) const char* src, _In_ size_t num_chars)
{
stdex_assert(src || !num_chars);
for (size_t i = 0; i < num_chars && src[i]; ++i) {
num_chars = stdex::strnlen(src, num_chars);
dst.reserve(dst.size() + num_chars + (num_chars >> 2));
for (size_t i = 0; i < num_chars; ++i) {
switch (src[i]) {
case '&': dst += "&amp;"; break;
case ';': dst += "&semi;"; break;
@@ -77,8 +78,9 @@ namespace stdex
_Inout_ std::basic_string<wchar_t, TR, AX>& dst,
_In_reads_or_z_opt_(num_chars) const wchar_t* src, _In_ size_t num_chars)
{
stdex_assert(src || !num_chars);
for (size_t i = 0; i < num_chars && src[i]; ++i) {
num_chars = stdex::strnlen(src, num_chars);
dst.reserve(dst.size() + num_chars + (num_chars >> 2));
for (size_t i = 0; i < num_chars; ++i) {
switch (src[i]) {
case L'&': dst += L"&amp;"; break;
case L';': dst += L"&semi;"; break;
@@ -92,6 +94,34 @@ namespace stdex
}
}
///
/// Appends HTML escaped string
///
/// \param[in,out] dst String to append to
/// \param[in] src Source string
/// \param[in] num_chars Code unit limit in string `src`
///
template<class TR = std::char_traits<char16_t>, class AX = std::allocator<char16_t>>
void escape(
_Inout_ std::basic_string<char16_t, TR, AX>& dst,
_In_reads_or_z_opt_(num_chars) const char16_t* src, _In_ size_t num_chars)
{
num_chars = stdex::strnlen(src, num_chars);
dst.reserve(dst.size() + num_chars + (num_chars >> 2));
for (size_t i = 0; i < num_chars; ++i) {
switch (src[i]) {
case L'&': dst += u"&amp;"; break;
case L';': dst += u"&semi;"; break;
case L'\"': dst += u"&quot;"; break;
case L'\'': dst += u"&#x27;"; break;
case L'<': dst += u"&lt;"; break;
case L'>': dst += u"&gt;"; break;
case L'\u00a0': dst += u"&nbsp;"; break; // No-break space must be escaped as SGML entity, otherwise browsers treat it as a normal space.
default: dst += src[i]; break;
}
}
}
///
/// Appends HTML escaped string
///
@@ -156,6 +186,24 @@ namespace stdex
}
}
///
/// Appends HTML escaped character
///
/// \param[in,out] dst String to append to
/// \param[in] chr Source character
///
template<class TR = std::char_traits<char16_t>, class AX = std::allocator<char16_t>>
void escape_min(_Inout_ std::basic_string<char16_t, TR, AX>& dst, _In_ char16_t chr)
{
switch (chr) {
case L'&': dst += u"&amp;"; break;
case L'<': dst += u"&lt;"; break;
case L'>': dst += u"&gt;"; break;
case L'\u00a0': dst += u"&nbsp;"; break; // No-break space must be escaped as SGML entity, otherwise browsers treat it as a normal space.
default: dst += chr; break;
}
}
///
/// Appends HTML escaped string
///
@@ -168,8 +216,9 @@ namespace stdex
_Inout_ std::basic_string<char, TR, AX>& dst,
_In_reads_or_z_opt_(num_chars) const char* src, _In_ size_t num_chars)
{
stdex_assert(src || !num_chars);
for (size_t i = 0; i < num_chars && src[i]; ++i) {
num_chars = stdex::strnlen(src, num_chars);
dst.reserve(dst.size() + num_chars + (num_chars >> 2));
for (size_t i = 0; i < num_chars; ++i) {
switch (src[i]) {
case '&': dst += "&amp;"; break;
case '<': dst += "&lt;"; break;
@@ -192,8 +241,9 @@ namespace stdex
_Inout_ std::basic_string<wchar_t, TR, AX>& dst,
_In_reads_or_z_opt_(num_chars) const wchar_t* src, _In_ size_t num_chars)
{
stdex_assert(src || !num_chars);
for (size_t i = 0; i < num_chars && src[i]; ++i) {
num_chars = stdex::strnlen(src, num_chars);
dst.reserve(dst.size() + num_chars + (num_chars >> 2));
for (size_t i = 0; i < num_chars; ++i) {
switch (src[i]) {
case L'&': dst += L"&amp;"; break;
case L'<': dst += L"&lt;"; break;
@@ -204,6 +254,31 @@ namespace stdex
}
}
///
/// Appends HTML escaped string
///
/// \param[in,out] dst String to append to
/// \param[in] src Source string
/// \param[in] num_chars Code unit limit in string `src`
///
template<class TR = std::char_traits<char16_t>, class AX = std::allocator<char16_t>>
void escape_min(
_Inout_ std::basic_string<char16_t, TR, AX>& dst,
_In_reads_or_z_opt_(num_chars) const char16_t* src, _In_ size_t num_chars)
{
num_chars = stdex::strnlen(src, num_chars);
dst.reserve(dst.size() + num_chars + (num_chars >> 2));
for (size_t i = 0; i < num_chars; ++i) {
switch (src[i]) {
case L'&': dst += u"&amp;"; break;
case L'<': dst += u"&lt;"; break;
case L'>': dst += u"&gt;"; break;
case L'\u00a0': dst += u"&nbsp;"; break; // No-break space must be escaped as SGML entity, otherwise browsers treat it as a normal space.
default: dst += src[i]; break;
}
}
}
///
/// Appends HTML escaped string
///
@@ -427,7 +502,7 @@ namespace stdex
case 'D': case 'd':
case 'E': case 'e':
case 'F': case 'f': {
wchar_t chr = 0;
char32_t chr = 0;
size_t end = std::min(num_chars, i + 6);
for (; i < end; ++i) {
@@ -487,46 +562,21 @@ namespace stdex
/// \param[in] src Source string
/// \param[in] num_chars Code unit limit in string `src`
///
template<class TR = std::char_traits<char>, class AX = std::allocator<char>>
template<class T, class TR = std::char_traits<T>, class AX = std::allocator<T>>
void css_escape(
_Inout_ std::basic_string<char, TR, AX>& dst,
_In_reads_or_z_opt_(num_chars) const char* src, _In_ size_t num_chars)
_Inout_ std::basic_string<T, TR, AX>& dst,
_In_reads_or_z_opt_(num_chars) const T* src, _In_ size_t num_chars)
{
stdex_assert(src || !num_chars);
for (size_t i = 0; i < num_chars && src[i]; ++i) {
num_chars = stdex::strnlen(src, num_chars);
dst.reserve(dst.size() + num_chars + (num_chars >> 3));
for (size_t i = 0; i < num_chars; ++i) {
switch (src[i]) {
case '\\': dst += "\\\\"; break;
case '\n': dst += "\\n"; break;
case '\r': dst += "\\r"; break;
case '\t': dst += "\\t"; break;
case '\"': dst += "\\\""; break;
case '\'': dst += "\\'"; break;
default: dst += src[i]; break;
}
}
}
///
/// Appends escaped CSS string
///
/// \param[in,out] dst String to append to
/// \param[in] src Source string
/// \param[in] num_chars Code unit limit in string `src`
///
template<class TR = std::char_traits<wchar_t>, class AX = std::allocator<wchar_t>>
void css_escape(
_Inout_ std::basic_string<wchar_t, TR, AX>& dst,
_In_reads_or_z_opt_(num_chars) const wchar_t* src, _In_ size_t num_chars)
{
stdex_assert(src || !num_chars);
for (size_t i = 0; i < num_chars && src[i]; ++i) {
switch (src[i]) {
case L'\\': dst += L"\\\\"; break;
case L'\n': dst += L"\\n"; break;
case L'\r': dst += L"\\r"; break;
case L'\t': dst += L"\\t"; break;
case L'\"': dst += L"\\\""; break;
case L'\'': dst += L"\\'"; break;
case '\\': dst += '\\'; dst+= '\\'; break;
case '\n': dst += '\\'; dst+= 'n'; break;
case '\r': dst += '\\'; dst+= 'r'; break;
case '\t': dst += '\\'; dst+= 't'; break;
case '\"': dst += '\\'; dst+= '"'; break;
case '\'': dst += '\\'; dst+= '\''; break;
default: dst += src[i]; break;
}
}
@@ -2015,27 +2065,47 @@ namespace stdex
///
/// \returns Number of code units appended
///
template<class TR = std::char_traits<char>, class AX = std::allocator<char>>
size_t append_tag(_Inout_ std::basic_string<char, TR, AX>& str) const
template<class T, class TR = std::char_traits<T>, class AX = std::allocator<T>>
size_t append_tag(_Inout_ std::basic_string<T, TR, AX>& str) const
{
size_t n = str.size();
// Use %X instead of %p to omit leading zeros and save space.
stdex::appendf(str, "%c%zX%c", stdex::locale_C, token_tag_start, reinterpret_cast<uintptr_t>(this), token_tag_end);
return str.size() - n;
}
str.reserve(str.size() + 2 + sizeof(this)*2);
str += token_tag_start;
///
/// Appends token tag to the source code
///
/// \param[in,out] str Source code
///
/// \returns Number of code units appended
///
template<class TR = std::char_traits<wchar_t>, class AX = std::allocator<wchar_t>>
size_t append_tag(_Inout_ std::basic_string<wchar_t, TR, AX>& str) const
{
// Use %X instead of %p to omit leading zeros and save space.
return stdex::appendf(str, L"%c%zX%c", stdex::locale_C, static_cast<wchar_t>(token_tag_start), reinterpret_cast<uintptr_t>(this), static_cast<wchar_t>(token_tag_end));
// Sure snprintf looks cleaner, but we don't have it for char16_t.
static const char digits[] =
"000102030405060708090A0B0C0D0E0F"
"101112131415161718191A1B1C1D1E1F"
"202122232425262728292A2B2C2D2E2F"
"303132333435363738393A3B3C3D3E3F"
"404142434445464748494A4B4C4D4E4F"
"505152535455565758595A5B5C5D5E5F"
"606162636465666768696A6B6C6D6E6F"
"707172737475767778797A7B7C7D7E7F"
"808182838485868788898A8B8C8D8E8F"
"909192939495969798999A9B9C9D9E9F"
"A0A1A2A3A4A5A6A7A8A9AAABACADAEAF"
"B0B1B2B3B4B5B6B7B8B9BABBBCBDBEBF"
"C0C1C2C3C4C5C6C7C8C9CACBCCCDCECF"
"D0D1D2D3D4D5D6D7D8D9DADBDCDDDEDF"
"E0E1E2E3E4E5E6E7E8E9EAEBECEDEEEF"
"F0F1F2F3F4F5F6F7F8F9FAFBFCFDFEFF";
char buf[16]; // Enough for 128-bit numbers
auto x = reinterpret_cast<uintptr_t>(this);
for (size_t i = _countof(buf); i;) {
size_t pos = (x & 0xFF) << 1;
buf[--i] = digits[pos + 1];
buf[--i] = digits[pos];
x >>= 8;
if (!x) {
for (i += pos < 32 ? 1 : 2; i < _countof(buf); ++i)
str += buf[i];
break;
}
}
str += token_tag_end;
return str.size() - n;
}
template<class T>