generic-library/poco
1
0
Fork 0
mirror of https://github.com/pocoproject/poco.git synced 2024-12-12 10:13:51 +01:00
Code Issues Projects Releases Wiki Activity

feat(Foundation): Include utf8proc for enhanced Unicode support (#4710)

Browse Source
This commit is contained in:
Günter Obiltschnig 2024-09-27 12:46:13 +02:00
parent 6faf90773a
commit 92e06497f5
10 changed files with 18726 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
Foundation/CMakeLists.txt
View File

@ -29,6 +29,7 @@ POCO_MESSAGES(SRCS Logging src/pocomsg.mc)
if(POCO_UNBUNDLED)
find_package(PCRE2 REQUIRED)
find_package(ZLIB REQUIRED)
find_package(Utf8Proc REQUIRED)
#HACK: Unicode.cpp requires functions from these files. The can't be taken from the library
POCO_SOURCES(SRCS RegExp
@ -86,6 +87,10 @@ else()
src/trees.c
src/zutil.c
)
POCO_SOURCES(SRCS utf8proc
src/utf8proc.c
)
endif(POCO_UNBUNDLED)
@ -99,7 +104,7 @@ set_target_properties(Foundation
)
if(POCO_UNBUNDLED)
target_link_libraries(Foundation PUBLIC Pcre2::Pcre2 ZLIB::ZLIB)
target_link_libraries(Foundation PUBLIC Pcre2::Pcre2 ZLIB::ZLIB Utf8Proc::Utf8Proc)
target_compile_definitions(Foundation PUBLIC POCO_UNBUNDLED)
endif(POCO_UNBUNDLED)

6
Foundation/Makefile
View File

@ -44,11 +44,13 @@ pcre_objects = pcre2_auto_possess pcre2_chartables pcre2_compile pcre2_config \
pcre_utf8_objects = pcre2_ucd pcre2_tables
utf8proc_objects = utf8proc
ifdef POCO_UNBUNDLED
SYSLIBS += -lpcre2-8 -lz
SYSLIBS += -lpcre2-8 -lutf8proc -lz
objects += $(pcre_utf8_objects)
else
objects += $(zlib_objects) $(pcre_objects) $(pcre_utf8_objects)
objects += $(zlib_objects) $(pcre_objects) $(pcre_utf8_objects) $(utf8proc_objects)
endif
ifeq ($(findstring MinGW, $(POCO_CONFIG)), MinGW)

17
Foundation/include/Poco/UTF8String.h
View File

@ -37,6 +37,15 @@ struct Foundation_API UTF8
/// removeBOM() removes the UTF-8 Byte Order Mark sequence (0xEF, 0xBB, 0xBF)
/// from the beginning of the given string, if it's there.
{
enum NormalizationForm
/// Normalization form for normalize().
{
NORMALIZATION_FORM_D, /// Canonical Decomposition
NORMALIZATION_FORM_C, /// Canonical Decomposition, followed by Canonical Composition
NORMALIZATION_FORM_KD, /// Compatibility Decomposition
NORMALIZATION_FORM_KC /// Compatibility Decomposition, followed by Canonical Composition
};
static int icompare(const std::string& str, std::string::size_type pos, std::string::size_type n, std::string::const_iterator it2, std::string::const_iterator end2);
static int icompare(const std::string& str1, const std::string& str2);
static int icompare(const std::string& str1, std::string::size_type n1, const std::string& str2, std::string::size_type n2);
@ -74,6 +83,14 @@ struct Foundation_API UTF8
static std::string unescape(const std::string::const_iterator& begin, const std::string::const_iterator& end);
/// Creates an UTF8 string from a string that contains escaped characters.
static std::string normalize(const std::string& s, NormalizationForm form);
/// Normalizes the given UTF8 string according to the given normalization form.
/// Returns the normalized UTF8 string.
static std::string normalize(const std::string::const_iterator& begin, const std::string::const_iterator& end, NormalizationForm form);
/// Normalizes the given UTF8 string range according to the given normalization form.
/// Returns the normalized UTF8 string.
};

58
Foundation/src/UTF8String.cpp
View File

@ -20,7 +20,17 @@
#include "Poco/UTF8Encoding.h"
#include "Poco/NumberFormatter.h"
#include "Poco/Ascii.h"
#include "Poco/Buffer.h"
#include "Poco/Exception.h"
#include <algorithm>
#include <iterator>
#if defined(POCO_UNBUNDLED)
#include <utf8proc.h>
#else
#include "utf8proc.h"
#endif
#if !defined(POCO_OS_FAMILY_WINDOWS)
@ -410,4 +420,52 @@ std::string UTF8::unescape(const std::string::const_iterator& begin, const std::
}
namespace
{
std::string doNormalize(const char* str, std::size_t size, utf8proc_option_t options)
{
utf8proc_ssize_t n = utf8proc_decompose_custom(reinterpret_cast<const utf8proc_uint8_t*>(str), size, NULL, 0, options, NULL, NULL);
if (n < 0) throw Poco::RuntimeException("Normalization decompose failed"s, utf8proc_errmsg(n));
Poco::Buffer<utf8proc_int32_t> buffer(n + 1); // utf8proc_reencode() needs space for terminating NUL
n = utf8proc_decompose_custom(reinterpret_cast<const utf8proc_uint8_t*>(str), size, buffer.begin(), n, options, NULL, NULL);
if (n < 0) throw Poco::RuntimeException("Normalization decompose failed"s, utf8proc_errmsg(n));
n = utf8proc_reencode(buffer.begin(), n, options);
if (n < 0) throw Poco::RuntimeException("Normalization reeencode failed"s, utf8proc_errmsg(n));
return std::string(reinterpret_cast<char*>(buffer.begin()), n);
}
int formToOptions(UTF8::NormalizationForm form)
{
switch (form)
{
case UTF8::NORMALIZATION_FORM_D:
return UTF8PROC_STABLE | UTF8PROC_DECOMPOSE;
case UTF8::NORMALIZATION_FORM_C:
return UTF8PROC_STABLE | UTF8PROC_COMPOSE;
case UTF8::NORMALIZATION_FORM_KD:
return UTF8PROC_STABLE | UTF8PROC_DECOMPOSE | UTF8PROC_COMPAT;
case UTF8::NORMALIZATION_FORM_KC:
return UTF8PROC_STABLE | UTF8PROC_COMPOSE | UTF8PROC_COMPAT;
default:
return 0;
}
}
}
std::string UTF8::normalize(const std::string& s, NormalizationForm form)
{
return doNormalize(s.data(), s.size(), static_cast<utf8proc_option_t>(formToOptions(form)));
}
std::string UTF8::normalize(const std::string::const_iterator& begin, const std::string::const_iterator& end, NormalizationForm form)
{
return doNormalize(&*begin, static_cast<std::size_t>(std::distance(begin, end)), static_cast<utf8proc_option_t>(formToOptions(form)));
}
} // namespace Poco

819
Foundation/src/utf8proc.c Normal file
View File

@ -0,0 +1,819 @@
/* -*- mode: c; c-basic-offset: 2; tab-width: 2; indent-tabs-mode: nil -*- */
/*
* Copyright (c) 2014-2021 Steven G. Johnson, Jiahao Chen, Peter Colberg, Tony Kelman, Scott P. Jones, and other contributors.
* Copyright (c) 2009 Public Software Group e. V., Berlin, Germany
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
* DEALINGS IN THE SOFTWARE.
*/
/*
* This library contains derived data from a modified version of the
* Unicode data files.
*
* The original data files are available at
* https://www.unicode.org/Public/UNIDATA/
*
* Please notice the copyright statement in the file "utf8proc_data.c".
*/
/*
* File name: utf8proc.c
*
* Description:
* Implementation of libutf8proc.
*/
#include "utf8proc.h"
#ifndef SSIZE_MAX
#define SSIZE_MAX ((size_t)SIZE_MAX/2)
#endif
#ifndef UINT16_MAX
# define UINT16_MAX 65535U
#endif
#include "utf8proc_data.c"
UTF8PROC_DLLEXPORT const utf8proc_int8_t utf8proc_utf8class[256] = {
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
4, 4, 4, 4, 4, 4, 4, 4, 0, 0, 0, 0, 0, 0, 0, 0 };
#define UTF8PROC_HANGUL_SBASE 0xAC00
#define UTF8PROC_HANGUL_LBASE 0x1100
#define UTF8PROC_HANGUL_VBASE 0x1161
#define UTF8PROC_HANGUL_TBASE 0x11A7
#define UTF8PROC_HANGUL_LCOUNT 19
#define UTF8PROC_HANGUL_VCOUNT 21
#define UTF8PROC_HANGUL_TCOUNT 28
#define UTF8PROC_HANGUL_NCOUNT 588
#define UTF8PROC_HANGUL_SCOUNT 11172
/* END is exclusive */
#define UTF8PROC_HANGUL_L_START 0x1100
#define UTF8PROC_HANGUL_L_END 0x115A
#define UTF8PROC_HANGUL_L_FILLER 0x115F
#define UTF8PROC_HANGUL_V_START 0x1160
#define UTF8PROC_HANGUL_V_END 0x11A3
#define UTF8PROC_HANGUL_T_START 0x11A8
#define UTF8PROC_HANGUL_T_END 0x11FA
#define UTF8PROC_HANGUL_S_START 0xAC00
#define UTF8PROC_HANGUL_S_END 0xD7A4
/* Should follow semantic-versioning rules (semver.org) based on API
compatibility. (Note that the shared-library version number will
be different, being based on ABI compatibility.): */
#define STRINGIZEx(x) #x
#define STRINGIZE(x) STRINGIZEx(x)
UTF8PROC_DLLEXPORT const char *utf8proc_version(void) {
return STRINGIZE(UTF8PROC_VERSION_MAJOR) "." STRINGIZE(UTF8PROC_VERSION_MINOR) "." STRINGIZE(UTF8PROC_VERSION_PATCH) "";
}
UTF8PROC_DLLEXPORT const char *utf8proc_unicode_version(void) {
return "15.1.0";
}
UTF8PROC_DLLEXPORT const char *utf8proc_errmsg(utf8proc_ssize_t errcode) {
switch (errcode) {
case UTF8PROC_ERROR_NOMEM:
return "Memory for processing UTF-8 data could not be allocated.";
case UTF8PROC_ERROR_OVERFLOW:
return "UTF-8 string is too long to be processed.";
case UTF8PROC_ERROR_INVALIDUTF8:
return "Invalid UTF-8 string";
case UTF8PROC_ERROR_NOTASSIGNED:
return "Unassigned Unicode code point found in UTF-8 string.";
case UTF8PROC_ERROR_INVALIDOPTS:
return "Invalid options for UTF-8 processing chosen.";
default:
return "An unknown error occurred while processing UTF-8 data.";
}
}
#define utf_cont(ch) (((ch) & 0xc0) == 0x80)
UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_iterate(
const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, utf8proc_int32_t *dst
) {
utf8proc_int32_t uc;
const utf8proc_uint8_t *end;
*dst = -1;
if (!strlen) return 0;
end = str + ((strlen < 0) ? 4 : strlen);
uc = *str++;
if (uc < 0x80) {
*dst = uc;
return 1;
}
// Must be between 0xc2 and 0xf4 inclusive to be valid
if ((utf8proc_uint32_t)(uc - 0xc2) > (0xf4-0xc2)) return UTF8PROC_ERROR_INVALIDUTF8;
if (uc < 0xe0) { // 2-byte sequence
// Must have valid continuation character
if (str >= end || !utf_cont(*str)) return UTF8PROC_ERROR_INVALIDUTF8;
*dst = ((uc & 0x1f)<<6) | (*str & 0x3f);
return 2;
}
if (uc < 0xf0) { // 3-byte sequence
if ((str + 1 >= end) || !utf_cont(*str) || !utf_cont(str[1]))
return UTF8PROC_ERROR_INVALIDUTF8;
// Check for surrogate chars
if (uc == 0xed && *str > 0x9f)
return UTF8PROC_ERROR_INVALIDUTF8;
uc = ((uc & 0xf)<<12) | ((*str & 0x3f)<<6) | (str[1] & 0x3f);
if (uc < 0x800)
return UTF8PROC_ERROR_INVALIDUTF8;
*dst = uc;
return 3;
}
// 4-byte sequence
// Must have 3 valid continuation characters
if ((str + 2 >= end) || !utf_cont(*str) || !utf_cont(str[1]) || !utf_cont(str[2]))
return UTF8PROC_ERROR_INVALIDUTF8;
// Make sure in correct range (0x10000 - 0x10ffff)
if (uc == 0xf0) {
if (*str < 0x90) return UTF8PROC_ERROR_INVALIDUTF8;
} else if (uc == 0xf4) {
if (*str > 0x8f) return UTF8PROC_ERROR_INVALIDUTF8;
}
*dst = ((uc & 7)<<18) | ((*str & 0x3f)<<12) | ((str[1] & 0x3f)<<6) | (str[2] & 0x3f);
return 4;
}
UTF8PROC_DLLEXPORT utf8proc_bool utf8proc_codepoint_valid(utf8proc_int32_t uc) {
return (((utf8proc_uint32_t)uc)-0xd800 > 0x07ff) && ((utf8proc_uint32_t)uc < 0x110000);
}
UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_encode_char(utf8proc_int32_t uc, utf8proc_uint8_t *dst) {
if (uc < 0x00) {
return 0;
} else if (uc < 0x80) {
dst[0] = (utf8proc_uint8_t) uc;
return 1;
} else if (uc < 0x800) {
dst[0] = (utf8proc_uint8_t)(0xC0 + (uc >> 6));
dst[1] = (utf8proc_uint8_t)(0x80 + (uc & 0x3F));
return 2;
// Note: we allow encoding 0xd800-0xdfff here, so as not to change
// the API, however, these are actually invalid in UTF-8
} else if (uc < 0x10000) {
dst[0] = (utf8proc_uint8_t)(0xE0 + (uc >> 12));
dst[1] = (utf8proc_uint8_t)(0x80 + ((uc >> 6) & 0x3F));
dst[2] = (utf8proc_uint8_t)(0x80 + (uc & 0x3F));
return 3;
} else if (uc < 0x110000) {
dst[0] = (utf8proc_uint8_t)(0xF0 + (uc >> 18));
dst[1] = (utf8proc_uint8_t)(0x80 + ((uc >> 12) & 0x3F));
dst[2] = (utf8proc_uint8_t)(0x80 + ((uc >> 6) & 0x3F));
dst[3] = (utf8proc_uint8_t)(0x80 + (uc & 0x3F));
return 4;
} else return 0;
}
/* internal version used for inserting 0xff bytes between graphemes */
static utf8proc_ssize_t charbound_encode_char(utf8proc_int32_t uc, utf8proc_uint8_t *dst) {
if (uc < 0x00) {
if (uc == -1) { /* internal value used for grapheme breaks */
dst[0] = (utf8proc_uint8_t)0xFF;
return 1;
}
return 0;
} else if (uc < 0x80) {
dst[0] = (utf8proc_uint8_t)uc;
return 1;
} else if (uc < 0x800) {
dst[0] = (utf8proc_uint8_t)(0xC0 + (uc >> 6));
dst[1] = (utf8proc_uint8_t)(0x80 + (uc & 0x3F));
return 2;
} else if (uc < 0x10000) {
dst[0] = (utf8proc_uint8_t)(0xE0 + (uc >> 12));
dst[1] = (utf8proc_uint8_t)(0x80 + ((uc >> 6) & 0x3F));
dst[2] = (utf8proc_uint8_t)(0x80 + (uc & 0x3F));
return 3;
} else if (uc < 0x110000) {
dst[0] = (utf8proc_uint8_t)(0xF0 + (uc >> 18));
dst[1] = (utf8proc_uint8_t)(0x80 + ((uc >> 12) & 0x3F));
dst[2] = (utf8proc_uint8_t)(0x80 + ((uc >> 6) & 0x3F));
dst[3] = (utf8proc_uint8_t)(0x80 + (uc & 0x3F));
return 4;
} else return 0;
}
/* internal "unsafe" version that does not check whether uc is in range */
static const utf8proc_property_t *unsafe_get_property(utf8proc_int32_t uc) {
/* ASSERT: uc >= 0 && uc < 0x110000 */
return utf8proc_properties + (
utf8proc_stage2table[
utf8proc_stage1table[uc >> 8] + (uc & 0xFF)
]
);
}
UTF8PROC_DLLEXPORT const utf8proc_property_t *utf8proc_get_property(utf8proc_int32_t uc) {
return uc < 0 || uc >= 0x110000 ? utf8proc_properties : unsafe_get_property(uc);
}
/* return whether there is a grapheme break between boundclasses lbc and tbc
(according to the definition of extended grapheme clusters)
Rule numbering refers to TR29 Version 29 (Unicode 9.0.0):
http://www.unicode.org/reports/tr29/tr29-29.html
CAVEATS:
Please note that evaluation of GB10 (grapheme breaks between emoji zwj sequences)
and GB 12/13 (regional indicator code points) require knowledge of previous characters
and are thus not handled by this function. This may result in an incorrect break before
an E_Modifier class codepoint and an incorrectly missing break between two
REGIONAL_INDICATOR class code points if such support does not exist in the caller.
See the special support in grapheme_break_extended, for required bookkeeping by the caller.
*/
static utf8proc_bool grapheme_break_simple(int lbc, int tbc) {
return
(lbc == UTF8PROC_BOUNDCLASS_START) ? true : // GB1
(lbc == UTF8PROC_BOUNDCLASS_CR && // GB3
tbc == UTF8PROC_BOUNDCLASS_LF) ? false : // ---
(lbc >= UTF8PROC_BOUNDCLASS_CR && lbc <= UTF8PROC_BOUNDCLASS_CONTROL) ? true : // GB4
(tbc >= UTF8PROC_BOUNDCLASS_CR && tbc <= UTF8PROC_BOUNDCLASS_CONTROL) ? true : // GB5
(lbc == UTF8PROC_BOUNDCLASS_L && // GB6
(tbc == UTF8PROC_BOUNDCLASS_L || // ---
tbc == UTF8PROC_BOUNDCLASS_V || // ---
tbc == UTF8PROC_BOUNDCLASS_LV || // ---
tbc == UTF8PROC_BOUNDCLASS_LVT)) ? false : // ---
((lbc == UTF8PROC_BOUNDCLASS_LV || // GB7
lbc == UTF8PROC_BOUNDCLASS_V) && // ---
(tbc == UTF8PROC_BOUNDCLASS_V || // ---
tbc == UTF8PROC_BOUNDCLASS_T)) ? false : // ---
((lbc == UTF8PROC_BOUNDCLASS_LVT || // GB8
lbc == UTF8PROC_BOUNDCLASS_T) && // ---
tbc == UTF8PROC_BOUNDCLASS_T) ? false : // ---
(tbc == UTF8PROC_BOUNDCLASS_EXTEND || // GB9
tbc == UTF8PROC_BOUNDCLASS_ZWJ || // ---
tbc == UTF8PROC_BOUNDCLASS_SPACINGMARK || // GB9a
lbc == UTF8PROC_BOUNDCLASS_PREPEND) ? false : // GB9b
(lbc == UTF8PROC_BOUNDCLASS_E_ZWG && // GB11 (requires additional handling below)
tbc == UTF8PROC_BOUNDCLASS_EXTENDED_PICTOGRAPHIC) ? false : // ----
(lbc == UTF8PROC_BOUNDCLASS_REGIONAL_INDICATOR && // GB12/13 (requires additional handling below)
tbc == UTF8PROC_BOUNDCLASS_REGIONAL_INDICATOR) ? false : // ----
true; // GB999
}
static utf8proc_bool grapheme_break_extended(int lbc, int tbc, int licb, int ticb, utf8proc_int32_t *state)
{
if (state) {
int state_bc, state_icb; /* boundclass and indic_conjunct_break state */
if (*state == 0) { /* state initialization */
state_bc = lbc;
state_icb = licb == UTF8PROC_INDIC_CONJUNCT_BREAK_CONSONANT ? licb : UTF8PROC_INDIC_CONJUNCT_BREAK_NONE;
}
else { /* lbc and licb are already encoded in *state */
state_bc = *state & 0xff; // 1st byte of state is bound class
state_icb = *state >> 8; // 2nd byte of state is indic conjunct break
}
utf8proc_bool break_permitted = grapheme_break_simple(state_bc, tbc) &&
!(state_icb == UTF8PROC_INDIC_CONJUNCT_BREAK_LINKER
&& ticb == UTF8PROC_INDIC_CONJUNCT_BREAK_CONSONANT); // GB9c
// Special support for GB9c. Don't break between two consonants
// separated 1+ linker characters and 0+ extend characters in any order.
// After a consonant, we enter LINKER state after at least one linker.
if (ticb == UTF8PROC_INDIC_CONJUNCT_BREAK_CONSONANT
|| state_icb == UTF8PROC_INDIC_CONJUNCT_BREAK_CONSONANT
|| state_icb == UTF8PROC_INDIC_CONJUNCT_BREAK_EXTEND)
state_icb = ticb;
else if (state_icb == UTF8PROC_INDIC_CONJUNCT_BREAK_LINKER)
state_icb = ticb == UTF8PROC_INDIC_CONJUNCT_BREAK_EXTEND ?
UTF8PROC_INDIC_CONJUNCT_BREAK_LINKER : ticb;
// Special support for GB 12/13 made possible by GB999. After two RI
// class codepoints we want to force a break. Do this by resetting the
// second RI's bound class to UTF8PROC_BOUNDCLASS_OTHER, to force a break
// after that character according to GB999 (unless of course such a break is
// forbidden by a different rule such as GB9).
if (state_bc == tbc && tbc == UTF8PROC_BOUNDCLASS_REGIONAL_INDICATOR)
state_bc = UTF8PROC_BOUNDCLASS_OTHER;
// Special support for GB11 (emoji extend* zwj / emoji)
else if (state_bc == UTF8PROC_BOUNDCLASS_EXTENDED_PICTOGRAPHIC) {
if (tbc == UTF8PROC_BOUNDCLASS_EXTEND) // fold EXTEND codepoints into emoji
state_bc = UTF8PROC_BOUNDCLASS_EXTENDED_PICTOGRAPHIC;
else if (tbc == UTF8PROC_BOUNDCLASS_ZWJ)
state_bc = UTF8PROC_BOUNDCLASS_E_ZWG; // state to record emoji+zwg combo
else
state_bc = tbc;
}
else
state_bc = tbc;
*state = state_bc + (state_icb << 8);
return break_permitted;
}
else
return grapheme_break_simple(lbc, tbc);
}
UTF8PROC_DLLEXPORT utf8proc_bool utf8proc_grapheme_break_stateful(
utf8proc_int32_t c1, utf8proc_int32_t c2, utf8proc_int32_t *state) {
const utf8proc_property_t *p1 = utf8proc_get_property(c1);
const utf8proc_property_t *p2 = utf8proc_get_property(c2);
return grapheme_break_extended(p1->boundclass,
p2->boundclass,
p1->indic_conjunct_break,
p2->indic_conjunct_break,
state);
}
UTF8PROC_DLLEXPORT utf8proc_bool utf8proc_grapheme_break(
utf8proc_int32_t c1, utf8proc_int32_t c2) {
return utf8proc_grapheme_break_stateful(c1, c2, NULL);
}
static utf8proc_int32_t seqindex_decode_entry(const utf8proc_uint16_t **entry)
{
utf8proc_int32_t entry_cp = **entry;
if ((entry_cp & 0xF800) == 0xD800) {
*entry = *entry + 1;
entry_cp = ((entry_cp & 0x03FF) << 10) | (**entry & 0x03FF);
entry_cp += 0x10000;
}
return entry_cp;
}
static utf8proc_int32_t seqindex_decode_index(const utf8proc_uint32_t seqindex)
{
const utf8proc_uint16_t *entry = &utf8proc_sequences[seqindex];
return seqindex_decode_entry(&entry);
}
static utf8proc_ssize_t seqindex_write_char_decomposed(utf8proc_uint16_t seqindex, utf8proc_int32_t *dst, utf8proc_ssize_t bufsize, utf8proc_option_t options, int *last_boundclass) {
utf8proc_ssize_t written = 0;
const utf8proc_uint16_t *entry = &utf8proc_sequences[seqindex & 0x3FFF];
int len = seqindex >> 14;
if (len >= 3) {
len = *entry;
entry++;
}
for (; len >= 0; entry++, len--) {
utf8proc_int32_t entry_cp = seqindex_decode_entry(&entry);
written += utf8proc_decompose_char(entry_cp, dst+written,
(bufsize > written) ? (bufsize - written) : 0, options,
last_boundclass);
if (written < 0) return UTF8PROC_ERROR_OVERFLOW;
}
return written;
}
UTF8PROC_DLLEXPORT utf8proc_int32_t utf8proc_tolower(utf8proc_int32_t c)
{
utf8proc_int32_t cl = utf8proc_get_property(c)->lowercase_seqindex;
return cl != UINT16_MAX ? seqindex_decode_index((utf8proc_uint32_t)cl) : c;
}
UTF8PROC_DLLEXPORT utf8proc_int32_t utf8proc_toupper(utf8proc_int32_t c)
{
utf8proc_int32_t cu = utf8proc_get_property(c)->uppercase_seqindex;
return cu != UINT16_MAX ? seqindex_decode_index((utf8proc_uint32_t)cu) : c;
}
UTF8PROC_DLLEXPORT utf8proc_int32_t utf8proc_totitle(utf8proc_int32_t c)
{
utf8proc_int32_t cu = utf8proc_get_property(c)->titlecase_seqindex;
return cu != UINT16_MAX ? seqindex_decode_index((utf8proc_uint32_t)cu) : c;
}
UTF8PROC_DLLEXPORT int utf8proc_islower(utf8proc_int32_t c)
{
const utf8proc_property_t *p = utf8proc_get_property(c);
return p->lowercase_seqindex != p->uppercase_seqindex && p->lowercase_seqindex == UINT16_MAX;
}
UTF8PROC_DLLEXPORT int utf8proc_isupper(utf8proc_int32_t c)
{
const utf8proc_property_t *p = utf8proc_get_property(c);
return p->lowercase_seqindex != p->uppercase_seqindex && p->uppercase_seqindex == UINT16_MAX && p->category != UTF8PROC_CATEGORY_LT;
}
/* return a character width analogous to wcwidth (except portable and
hopefully less buggy than most system wcwidth functions). */
UTF8PROC_DLLEXPORT int utf8proc_charwidth(utf8proc_int32_t c) {
return utf8proc_get_property(c)->charwidth;
}
UTF8PROC_DLLEXPORT utf8proc_bool utf8proc_charwidth_ambiguous(utf8proc_int32_t c) {
return utf8proc_get_property(c)->ambiguous_width;
}
UTF8PROC_DLLEXPORT utf8proc_category_t utf8proc_category(utf8proc_int32_t c) {
return (utf8proc_category_t) utf8proc_get_property(c)->category;
}
UTF8PROC_DLLEXPORT const char *utf8proc_category_string(utf8proc_int32_t c) {
static const char s[][3] = {"Cn","Lu","Ll","Lt","Lm","Lo","Mn","Mc","Me","Nd","Nl","No","Pc","Pd","Ps","Pe","Pi","Pf","Po","Sm","Sc","Sk","So","Zs","Zl","Zp","Cc","Cf","Cs","Co"};
return s[utf8proc_category(c)];
}
#define utf8proc_decompose_lump(replacement_uc) \
return utf8proc_decompose_char((replacement_uc), dst, bufsize, \
options & ~(unsigned int)UTF8PROC_LUMP, last_boundclass)
UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose_char(utf8proc_int32_t uc, utf8proc_int32_t *dst, utf8proc_ssize_t bufsize, utf8proc_option_t options, int *last_boundclass) {
const utf8proc_property_t *property;
utf8proc_propval_t category;
utf8proc_int32_t hangul_sindex;
if (uc < 0 || uc >= 0x110000) return UTF8PROC_ERROR_NOTASSIGNED;
property = unsafe_get_property(uc);
category = property->category;
hangul_sindex = uc - UTF8PROC_HANGUL_SBASE;
if (options & (UTF8PROC_COMPOSE|UTF8PROC_DECOMPOSE)) {
if (hangul_sindex >= 0 && hangul_sindex < UTF8PROC_HANGUL_SCOUNT) {
utf8proc_int32_t hangul_tindex;
if (bufsize >= 1) {
dst[0] = UTF8PROC_HANGUL_LBASE +
hangul_sindex / UTF8PROC_HANGUL_NCOUNT;
if (bufsize >= 2) dst[1] = UTF8PROC_HANGUL_VBASE +
(hangul_sindex % UTF8PROC_HANGUL_NCOUNT) / UTF8PROC_HANGUL_TCOUNT;
}
hangul_tindex = hangul_sindex % UTF8PROC_HANGUL_TCOUNT;
if (!hangul_tindex) return 2;
if (bufsize >= 3) dst[2] = UTF8PROC_HANGUL_TBASE + hangul_tindex;
return 3;
}
}
if (options & UTF8PROC_REJECTNA) {
if (!category) return UTF8PROC_ERROR_NOTASSIGNED;
}
if (options & UTF8PROC_IGNORE) {
if (property->ignorable) return 0;
}
if (options & UTF8PROC_STRIPNA) {
if (!category) return 0;
}
if (options & UTF8PROC_LUMP) {
if (category == UTF8PROC_CATEGORY_ZS) utf8proc_decompose_lump(0x0020);
if (uc == 0x2018 || uc == 0x2019 || uc == 0x02BC || uc == 0x02C8)
utf8proc_decompose_lump(0x0027);
if (category == UTF8PROC_CATEGORY_PD || uc == 0x2212)
utf8proc_decompose_lump(0x002D);
if (uc == 0x2044 || uc == 0x2215) utf8proc_decompose_lump(0x002F);
if (uc == 0x2236) utf8proc_decompose_lump(0x003A);
if (uc == 0x2039 || uc == 0x2329 || uc == 0x3008)
utf8proc_decompose_lump(0x003C);
if (uc == 0x203A || uc == 0x232A || uc == 0x3009)
utf8proc_decompose_lump(0x003E);
if (uc == 0x2216) utf8proc_decompose_lump(0x005C);
if (uc == 0x02C4 || uc == 0x02C6 || uc == 0x2038 || uc == 0x2303)
utf8proc_decompose_lump(0x005E);
if (category == UTF8PROC_CATEGORY_PC || uc == 0x02CD)
utf8proc_decompose_lump(0x005F);
if (uc == 0x02CB) utf8proc_decompose_lump(0x0060);
if (uc == 0x2223) utf8proc_decompose_lump(0x007C);
if (uc == 0x223C) utf8proc_decompose_lump(0x007E);
if ((options & UTF8PROC_NLF2LS) && (options & UTF8PROC_NLF2PS)) {
if (category == UTF8PROC_CATEGORY_ZL ||
category == UTF8PROC_CATEGORY_ZP)
utf8proc_decompose_lump(0x000A);
}
}
if (options & UTF8PROC_STRIPMARK) {
if (category == UTF8PROC_CATEGORY_MN ||
category == UTF8PROC_CATEGORY_MC ||
category == UTF8PROC_CATEGORY_ME) return 0;
}
if (options & UTF8PROC_CASEFOLD) {
if (property->casefold_seqindex != UINT16_MAX) {
return seqindex_write_char_decomposed(property->casefold_seqindex, dst, bufsize, options, last_boundclass);
}
}
if (options & (UTF8PROC_COMPOSE|UTF8PROC_DECOMPOSE)) {
if (property->decomp_seqindex != UINT16_MAX &&
(!property->decomp_type || (options & UTF8PROC_COMPAT))) {
return seqindex_write_char_decomposed(property->decomp_seqindex, dst, bufsize, options, last_boundclass);
}
}
if (options & UTF8PROC_CHARBOUND) {
utf8proc_bool boundary;
boundary = grapheme_break_extended(0, property->boundclass, 0, property->indic_conjunct_break,
last_boundclass);
if (boundary) {
if (bufsize >= 1) dst[0] = -1; /* sentinel value for grapheme break */
if (bufsize >= 2) dst[1] = uc;
return 2;
}
}
if (bufsize >= 1) *dst = uc;
return 1;
}
UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose(
const utf8proc_uint8_t *str, utf8proc_ssize_t strlen,
utf8proc_int32_t *buffer, utf8proc_ssize_t bufsize, utf8proc_option_t options
) {
return utf8proc_decompose_custom(str, strlen, buffer, bufsize, options, NULL, NULL);
}
UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose_custom(
const utf8proc_uint8_t *str, utf8proc_ssize_t strlen,
utf8proc_int32_t *buffer, utf8proc_ssize_t bufsize, utf8proc_option_t options,
utf8proc_custom_func custom_func, void *custom_data
) {
/* strlen will be ignored, if UTF8PROC_NULLTERM is set in options */
utf8proc_ssize_t wpos = 0;
if ((options & UTF8PROC_COMPOSE) && (options & UTF8PROC_DECOMPOSE))
return UTF8PROC_ERROR_INVALIDOPTS;
if ((options & UTF8PROC_STRIPMARK) &&
!(options & UTF8PROC_COMPOSE) && !(options & UTF8PROC_DECOMPOSE))
return UTF8PROC_ERROR_INVALIDOPTS;
{
utf8proc_int32_t uc;
utf8proc_ssize_t rpos = 0;
utf8proc_ssize_t decomp_result;
int boundclass = UTF8PROC_BOUNDCLASS_START;
while (1) {
if (options & UTF8PROC_NULLTERM) {
rpos += utf8proc_iterate(str + rpos, -1, &uc);
/* checking of return value is not necessary,
as 'uc' is < 0 in case of error */
if (uc < 0) return UTF8PROC_ERROR_INVALIDUTF8;
if (rpos < 0) return UTF8PROC_ERROR_OVERFLOW;
if (uc == 0) break;
} else {
if (rpos >= strlen) break;
rpos += utf8proc_iterate(str + rpos, strlen - rpos, &uc);
if (uc < 0) return UTF8PROC_ERROR_INVALIDUTF8;
}
if (custom_func != NULL) {
uc = custom_func(uc, custom_data); /* user-specified custom mapping */
}
decomp_result = utf8proc_decompose_char(
uc, buffer + wpos, (bufsize > wpos) ? (bufsize - wpos) : 0, options,
&boundclass
);
if (decomp_result < 0) return decomp_result;
wpos += decomp_result;
/* prohibiting integer overflows due to too long strings: */
if (wpos < 0 ||
wpos > (utf8proc_ssize_t)(SSIZE_MAX/sizeof(utf8proc_int32_t)/2))
return UTF8PROC_ERROR_OVERFLOW;
}
}
if ((options & (UTF8PROC_COMPOSE|UTF8PROC_DECOMPOSE)) && bufsize >= wpos) {
utf8proc_ssize_t pos = 0;
while (pos < wpos-1) {
utf8proc_int32_t uc1, uc2;
const utf8proc_property_t *property1, *property2;
uc1 = buffer[pos];
uc2 = buffer[pos+1];
property1 = unsafe_get_property(uc1);
property2 = unsafe_get_property(uc2);
if (property1->combining_class > property2->combining_class &&
property2->combining_class > 0) {
buffer[pos] = uc2;
buffer[pos+1] = uc1;
if (pos > 0) pos--; else pos++;
} else {
pos++;
}
}
}
return wpos;
}
UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_normalize_utf32(utf8proc_int32_t *buffer, utf8proc_ssize_t length, utf8proc_option_t options) {
/* UTF8PROC_NULLTERM option will be ignored, 'length' is never ignored */
if (options & (UTF8PROC_NLF2LS | UTF8PROC_NLF2PS | UTF8PROC_STRIPCC)) {
utf8proc_ssize_t rpos;
utf8proc_ssize_t wpos = 0;
utf8proc_int32_t uc;
for (rpos = 0; rpos < length; rpos++) {
uc = buffer[rpos];
if (uc == 0x000D && rpos < length-1 && buffer[rpos+1] == 0x000A) rpos++;
if (uc == 0x000A || uc == 0x000D || uc == 0x0085 ||
((options & UTF8PROC_STRIPCC) && (uc == 0x000B || uc == 0x000C))) {
if (options & UTF8PROC_NLF2LS) {
if (options & UTF8PROC_NLF2PS) {
buffer[wpos++] = 0x000A;
} else {
buffer[wpos++] = 0x2028;
}
} else {
if (options & UTF8PROC_NLF2PS) {
buffer[wpos++] = 0x2029;
} else {
buffer[wpos++] = 0x0020;
}
}
} else if ((options & UTF8PROC_STRIPCC) &&
(uc < 0x0020 || (uc >= 0x007F && uc < 0x00A0))) {
if (uc == 0x0009) buffer[wpos++] = 0x0020;
} else {
buffer[wpos++] = uc;
}
}
length = wpos;
}
if (options & UTF8PROC_COMPOSE) {
utf8proc_int32_t *starter = NULL;
utf8proc_int32_t current_char;
const utf8proc_property_t *starter_property = NULL, *current_property;
utf8proc_propval_t max_combining_class = -1;
utf8proc_ssize_t rpos;
utf8proc_ssize_t wpos = 0;
utf8proc_int32_t composition;
for (rpos = 0; rpos < length; rpos++) {
current_char = buffer[rpos];
current_property = unsafe_get_property(current_char);
if (starter && current_property->combining_class > max_combining_class) {
/* combination perhaps possible */
utf8proc_int32_t hangul_lindex;
utf8proc_int32_t hangul_sindex;
hangul_lindex = *starter - UTF8PROC_HANGUL_LBASE;
if (hangul_lindex >= 0 && hangul_lindex < UTF8PROC_HANGUL_LCOUNT) {
utf8proc_int32_t hangul_vindex;
hangul_vindex = current_char - UTF8PROC_HANGUL_VBASE;
if (hangul_vindex >= 0 && hangul_vindex < UTF8PROC_HANGUL_VCOUNT) {
*starter = UTF8PROC_HANGUL_SBASE +
(hangul_lindex * UTF8PROC_HANGUL_VCOUNT + hangul_vindex) *
UTF8PROC_HANGUL_TCOUNT;
starter_property = NULL;
continue;
}
}
hangul_sindex = *starter - UTF8PROC_HANGUL_SBASE;
if (hangul_sindex >= 0 && hangul_sindex < UTF8PROC_HANGUL_SCOUNT &&
(hangul_sindex % UTF8PROC_HANGUL_TCOUNT) == 0) {
utf8proc_int32_t hangul_tindex;
hangul_tindex = current_char - UTF8PROC_HANGUL_TBASE;
if (hangul_tindex >= 0 && hangul_tindex < UTF8PROC_HANGUL_TCOUNT) {
*starter += hangul_tindex;
starter_property = NULL;
continue;
}
}
if (!starter_property) {
starter_property = unsafe_get_property(*starter);
}
if (starter_property->comb_index < 0x8000 &&
current_property->comb_index != UINT16_MAX &&
current_property->comb_index >= 0x8000) {
int sidx = starter_property->comb_index;
int idx = current_property->comb_index & 0x3FFF;
if (idx >= utf8proc_combinations[sidx] && idx <= utf8proc_combinations[sidx + 1] ) {
idx += sidx + 2 - utf8proc_combinations[sidx];
if (current_property->comb_index & 0x4000) {
composition = (utf8proc_combinations[idx] << 16) | utf8proc_combinations[idx+1];
} else
composition = utf8proc_combinations[idx];
if (composition > 0 && (!(options & UTF8PROC_STABLE) ||
!(unsafe_get_property(composition)->comp_exclusion))) {
*starter = composition;
starter_property = NULL;
continue;
}
}
}
}
buffer[wpos] = current_char;
if (current_property->combining_class) {
if (current_property->combining_class > max_combining_class) {
max_combining_class = current_property->combining_class;
}
} else {
starter = buffer + wpos;
starter_property = NULL;
max_combining_class = -1;
}
wpos++;
}
length = wpos;
}
return length;
}
UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_reencode(utf8proc_int32_t *buffer, utf8proc_ssize_t length, utf8proc_option_t options) {
/* UTF8PROC_NULLTERM option will be ignored, 'length' is never ignored
ASSERT: 'buffer' has one spare byte of free space at the end! */
length = utf8proc_normalize_utf32(buffer, length, options);
if (length < 0) return length;
{
utf8proc_ssize_t rpos, wpos = 0;
utf8proc_int32_t uc;
if (options & UTF8PROC_CHARBOUND) {
for (rpos = 0; rpos < length; rpos++) {
uc = buffer[rpos];
wpos += charbound_encode_char(uc, ((utf8proc_uint8_t *)buffer) + wpos);
}
} else {
for (rpos = 0; rpos < length; rpos++) {
uc = buffer[rpos];
wpos += utf8proc_encode_char(uc, ((utf8proc_uint8_t *)buffer) + wpos);
}
}
((utf8proc_uint8_t *)buffer)[wpos] = 0;
return wpos;
}
}
UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_map(
const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, utf8proc_uint8_t **dstptr, utf8proc_option_t options
) {
return utf8proc_map_custom(str, strlen, dstptr, options, NULL, NULL);
}
UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_map_custom(
const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, utf8proc_uint8_t **dstptr, utf8proc_option_t options,
utf8proc_custom_func custom_func, void *custom_data
) {
utf8proc_int32_t *buffer;
utf8proc_ssize_t result;
*dstptr = NULL;
result = utf8proc_decompose_custom(str, strlen, NULL, 0, options, custom_func, custom_data);
if (result < 0) return result;
buffer = (utf8proc_int32_t *) malloc(((utf8proc_size_t)result) * sizeof(utf8proc_int32_t) + 1);
if (!buffer) return UTF8PROC_ERROR_NOMEM;
result = utf8proc_decompose_custom(str, strlen, buffer, result, options, custom_func, custom_data);
if (result < 0) {
free(buffer);
return result;
}
result = utf8proc_reencode(buffer, result, options);
if (result < 0) {
free(buffer);
return result;
}
{
utf8proc_int32_t *newptr;
newptr = (utf8proc_int32_t *) realloc(buffer, (size_t)result+1);
if (newptr) buffer = newptr;
}
*dstptr = (utf8proc_uint8_t *)buffer;
return result;
}
UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFD(const utf8proc_uint8_t *str) {
utf8proc_uint8_t *retval;
utf8proc_map(str, 0, &retval, UTF8PROC_NULLTERM | UTF8PROC_STABLE |
UTF8PROC_DECOMPOSE);
return retval;
}
UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFC(const utf8proc_uint8_t *str) {
utf8proc_uint8_t *retval;
utf8proc_map(str, 0, &retval, UTF8PROC_NULLTERM | UTF8PROC_STABLE |
UTF8PROC_COMPOSE);
return retval;
}
UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFKD(const utf8proc_uint8_t *str) {
utf8proc_uint8_t *retval;
utf8proc_map(str, 0, &retval, UTF8PROC_NULLTERM | UTF8PROC_STABLE |
UTF8PROC_DECOMPOSE | UTF8PROC_COMPAT);
return retval;
}
UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFKC(const utf8proc_uint8_t *str) {
utf8proc_uint8_t *retval;
utf8proc_map(str, 0, &retval, UTF8PROC_NULLTERM | UTF8PROC_STABLE |
UTF8PROC_COMPOSE | UTF8PROC_COMPAT);
return retval;
}
UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFKC_Casefold(const utf8proc_uint8_t *str) {
utf8proc_uint8_t *retval;
utf8proc_map(str, 0, &retval, UTF8PROC_NULLTERM | UTF8PROC_STABLE |
UTF8PROC_COMPOSE | UTF8PROC_COMPAT | UTF8PROC_CASEFOLD | UTF8PROC_IGNORE);
return retval;
}

753
Foundation/src/utf8proc.h Normal file
View File

@ -0,0 +1,753 @@
/*
* Copyright (c) 2014-2021 Steven G. Johnson, Jiahao Chen, Peter Colberg, Tony Kelman, Scott P. Jones, and other contributors.
* Copyright (c) 2009 Public Software Group e. V., Berlin, Germany
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
* DEALINGS IN THE SOFTWARE.
*/
/**
* @mainpage
*
* utf8proc is a free/open-source (MIT/expat licensed) C library
* providing Unicode normalization, case-folding, and other operations
* for strings in the UTF-8 encoding, supporting up-to-date Unicode versions.
* See the utf8proc home page (http://julialang.org/utf8proc/)
* for downloads and other information, or the source code on github
* (https://github.com/JuliaLang/utf8proc).
*
* For the utf8proc API documentation, see: @ref utf8proc.h
*
* The features of utf8proc include:
*
* - Transformation of strings (utf8proc_map()) to:
* - decompose (@ref UTF8PROC_DECOMPOSE) or compose (@ref UTF8PROC_COMPOSE) Unicode combining characters (http://en.wikipedia.org/wiki/Combining_character)
* - canonicalize Unicode compatibility characters (@ref UTF8PROC_COMPAT)
* - strip "ignorable" (@ref UTF8PROC_IGNORE) characters, control characters (@ref UTF8PROC_STRIPCC), or combining characters such as accents (@ref UTF8PROC_STRIPMARK)
* - case-folding (@ref UTF8PROC_CASEFOLD)
* - Unicode normalization: utf8proc_NFD(), utf8proc_NFC(), utf8proc_NFKD(), utf8proc_NFKC()
* - Detecting grapheme boundaries (utf8proc_grapheme_break() and @ref UTF8PROC_CHARBOUND)
* - Character-width computation: utf8proc_charwidth()
* - Classification of characters by Unicode category: utf8proc_category() and utf8proc_category_string()
* - Encode (utf8proc_encode_char()) and decode (utf8proc_iterate()) Unicode codepoints to/from UTF-8.
*/
/** @file */
#ifndef UTF8PROC_H
#define UTF8PROC_H
/** @name API version
*
* The utf8proc API version MAJOR.MINOR.PATCH, following
* semantic-versioning rules (http://semver.org) based on API
* compatibility.
*
* This is also returned at runtime by utf8proc_version(); however, the
* runtime version may append a string like "-dev" to the version number
* for prerelease versions.
*
* @note The shared-library version number in the Makefile
* (and CMakeLists.txt, and MANIFEST) may be different,
* being based on ABI compatibility rather than API compatibility.
*/
/** @{ */
/** The MAJOR version number (increased when backwards API compatibility is broken). */
#define UTF8PROC_VERSION_MAJOR 2
/** The MINOR version number (increased when new functionality is added in a backwards-compatible manner). */
#define UTF8PROC_VERSION_MINOR 9
/** The PATCH version (increased for fixes that do not change the API). */
#define UTF8PROC_VERSION_PATCH 0
/** @} */
#include <stdlib.h>
#if defined(_MSC_VER) && _MSC_VER < 1800
// MSVC prior to 2013 lacked stdbool.h and stdint.h
typedef signed char utf8proc_int8_t;
typedef unsigned char utf8proc_uint8_t;
typedef short utf8proc_int16_t;
typedef unsigned short utf8proc_uint16_t;
typedef int utf8proc_int32_t;
typedef unsigned int utf8proc_uint32_t;
# ifdef _WIN64
typedef __int64 utf8proc_ssize_t;
typedef unsigned __int64 utf8proc_size_t;
# else
typedef int utf8proc_ssize_t;
typedef unsigned int utf8proc_size_t;
# endif
# ifndef __cplusplus
// emulate C99 bool
typedef unsigned char utf8proc_bool;
# ifndef __bool_true_false_are_defined
# define false 0
# define true 1
# define __bool_true_false_are_defined 1
# endif
# else
typedef bool utf8proc_bool;
# endif
#else
# include <stddef.h>
# include <stdbool.h>
# include <stdint.h>
typedef int8_t utf8proc_int8_t;
typedef uint8_t utf8proc_uint8_t;
typedef int16_t utf8proc_int16_t;
typedef uint16_t utf8proc_uint16_t;
typedef int32_t utf8proc_int32_t;
typedef uint32_t utf8proc_uint32_t;
typedef size_t utf8proc_size_t;
typedef ptrdiff_t utf8proc_ssize_t;
typedef bool utf8proc_bool;
#endif
#include <limits.h>
#ifdef UTF8PROC_STATIC
# define UTF8PROC_DLLEXPORT
#else
# ifdef _WIN32
# ifdef UTF8PROC_EXPORTS
# define UTF8PROC_DLLEXPORT __declspec(dllexport)
# else
# define UTF8PROC_DLLEXPORT __declspec(dllimport)
# endif
# elif __GNUC__ >= 4
# define UTF8PROC_DLLEXPORT __attribute__ ((visibility("default")))
# else
# define UTF8PROC_DLLEXPORT
# endif
#endif
#ifdef __cplusplus
extern "C" {
#endif
/**
* Option flags used by several functions in the library.
*/
typedef enum {
/** The given UTF-8 input is NULL terminated. */
UTF8PROC_NULLTERM = (1<<0),
/** Unicode Versioning Stability has to be respected. */
UTF8PROC_STABLE = (1<<1),
/** Compatibility decomposition (i.e. formatting information is lost). */
UTF8PROC_COMPAT = (1<<2),
/** Return a result with decomposed characters. */
UTF8PROC_COMPOSE = (1<<3),
/** Return a result with decomposed characters. */
UTF8PROC_DECOMPOSE = (1<<4),
/** Strip "default ignorable characters" such as SOFT-HYPHEN or ZERO-WIDTH-SPACE. */
UTF8PROC_IGNORE = (1<<5),
/** Return an error, if the input contains unassigned codepoints. */
UTF8PROC_REJECTNA = (1<<6),
/**
* Indicating that NLF-sequences (LF, CRLF, CR, NEL) are representing a
* line break, and should be converted to the codepoint for line
* separation (LS).
*/
UTF8PROC_NLF2LS = (1<<7),
/**
* Indicating that NLF-sequences are representing a paragraph break, and
* should be converted to the codepoint for paragraph separation
* (PS).
*/
UTF8PROC_NLF2PS = (1<<8),
/** Indicating that the meaning of NLF-sequences is unknown. */
UTF8PROC_NLF2LF = (UTF8PROC_NLF2LS | UTF8PROC_NLF2PS),
/** Strips and/or convers control characters.
*
* NLF-sequences are transformed into space, except if one of the
* NLF2LS/PS/LF options is given. HorizontalTab (HT) and FormFeed (FF)
* are treated as a NLF-sequence in this case. All other control
* characters are simply removed.
*/
UTF8PROC_STRIPCC = (1<<9),
/**
* Performs unicode case folding, to be able to do a case-insensitive
* string comparison.
*/
UTF8PROC_CASEFOLD = (1<<10),
/**
* Inserts 0xFF bytes at the beginning of each sequence which is
* representing a single grapheme cluster (see UAX#29).
*/
UTF8PROC_CHARBOUND = (1<<11),
/** Lumps certain characters together.
*
* E.g. HYPHEN U+2010 and MINUS U+2212 to ASCII "-". See lump.md for details.
*
* If NLF2LF is set, this includes a transformation of paragraph and
* line separators to ASCII line-feed (LF).
*/
UTF8PROC_LUMP = (1<<12),
/** Strips all character markings.
*
* This includes non-spacing, spacing and enclosing (i.e. accents).
* @note This option works only with @ref UTF8PROC_COMPOSE or
* @ref UTF8PROC_DECOMPOSE
*/
UTF8PROC_STRIPMARK = (1<<13),
/**
* Strip unassigned codepoints.
*/
UTF8PROC_STRIPNA = (1<<14),
} utf8proc_option_t;
/** @name Error codes
* Error codes being returned by almost all functions.
*/
/** @{ */
/** Memory could not be allocated. */
#define UTF8PROC_ERROR_NOMEM -1
/** The given string is too long to be processed. */
#define UTF8PROC_ERROR_OVERFLOW -2
/** The given string is not a legal UTF-8 string. */
#define UTF8PROC_ERROR_INVALIDUTF8 -3
/** The @ref UTF8PROC_REJECTNA flag was set and an unassigned codepoint was found. */
#define UTF8PROC_ERROR_NOTASSIGNED -4
/** Invalid options have been used. */
#define UTF8PROC_ERROR_INVALIDOPTS -5
/** @} */
/* @name Types */
/** Holds the value of a property. */
typedef utf8proc_int16_t utf8proc_propval_t;
/** Struct containing information about a codepoint. */
typedef struct utf8proc_property_struct {
/**
* Unicode category.
* @see utf8proc_category_t.
*/
utf8proc_propval_t category;
utf8proc_propval_t combining_class;
/**
* Bidirectional class.
* @see utf8proc_bidi_class_t.
*/
utf8proc_propval_t bidi_class;
/**
* @anchor Decomposition type.
* @see utf8proc_decomp_type_t.
*/
utf8proc_propval_t decomp_type;
utf8proc_uint16_t decomp_seqindex;
utf8proc_uint16_t casefold_seqindex;
utf8proc_uint16_t uppercase_seqindex;
utf8proc_uint16_t lowercase_seqindex;
utf8proc_uint16_t titlecase_seqindex;
utf8proc_uint16_t comb_index;
unsigned bidi_mirrored:1;
unsigned comp_exclusion:1;
/**
* Can this codepoint be ignored?
*
* Used by utf8proc_decompose_char() when @ref UTF8PROC_IGNORE is
* passed as an option.
*/
unsigned ignorable:1;
unsigned control_boundary:1;
/** The width of the codepoint. */
unsigned charwidth:2;
/** East Asian width class A */
unsigned ambiguous_width:1;
unsigned pad:1;
/**
* Boundclass.
* @see utf8proc_boundclass_t.
*/
unsigned boundclass:6;
unsigned indic_conjunct_break:2;
} utf8proc_property_t;
/** Unicode categories. */
typedef enum {
UTF8PROC_CATEGORY_CN = 0, /**< Other, not assigned */
UTF8PROC_CATEGORY_LU = 1, /**< Letter, uppercase */
UTF8PROC_CATEGORY_LL = 2, /**< Letter, lowercase */
UTF8PROC_CATEGORY_LT = 3, /**< Letter, titlecase */
UTF8PROC_CATEGORY_LM = 4, /**< Letter, modifier */
UTF8PROC_CATEGORY_LO = 5, /**< Letter, other */
UTF8PROC_CATEGORY_MN = 6, /**< Mark, nonspacing */
UTF8PROC_CATEGORY_MC = 7, /**< Mark, spacing combining */
UTF8PROC_CATEGORY_ME = 8, /**< Mark, enclosing */
UTF8PROC_CATEGORY_ND = 9, /**< Number, decimal digit */
UTF8PROC_CATEGORY_NL = 10, /**< Number, letter */
UTF8PROC_CATEGORY_NO = 11, /**< Number, other */
UTF8PROC_CATEGORY_PC = 12, /**< Punctuation, connector */
UTF8PROC_CATEGORY_PD = 13, /**< Punctuation, dash */
UTF8PROC_CATEGORY_PS = 14, /**< Punctuation, open */
UTF8PROC_CATEGORY_PE = 15, /**< Punctuation, close */
UTF8PROC_CATEGORY_PI = 16, /**< Punctuation, initial quote */
UTF8PROC_CATEGORY_PF = 17, /**< Punctuation, final quote */
UTF8PROC_CATEGORY_PO = 18, /**< Punctuation, other */
UTF8PROC_CATEGORY_SM = 19, /**< Symbol, math */
UTF8PROC_CATEGORY_SC = 20, /**< Symbol, currency */
UTF8PROC_CATEGORY_SK = 21, /**< Symbol, modifier */
UTF8PROC_CATEGORY_SO = 22, /**< Symbol, other */
UTF8PROC_CATEGORY_ZS = 23, /**< Separator, space */
UTF8PROC_CATEGORY_ZL = 24, /**< Separator, line */
UTF8PROC_CATEGORY_ZP = 25, /**< Separator, paragraph */
UTF8PROC_CATEGORY_CC = 26, /**< Other, control */
UTF8PROC_CATEGORY_CF = 27, /**< Other, format */
UTF8PROC_CATEGORY_CS = 28, /**< Other, surrogate */
UTF8PROC_CATEGORY_CO = 29, /**< Other, private use */
} utf8proc_category_t;
/** Bidirectional character classes. */
typedef enum {
UTF8PROC_BIDI_CLASS_L = 1, /**< Left-to-Right */
UTF8PROC_BIDI_CLASS_LRE = 2, /**< Left-to-Right Embedding */
UTF8PROC_BIDI_CLASS_LRO = 3, /**< Left-to-Right Override */
UTF8PROC_BIDI_CLASS_R = 4, /**< Right-to-Left */
UTF8PROC_BIDI_CLASS_AL = 5, /**< Right-to-Left Arabic */
UTF8PROC_BIDI_CLASS_RLE = 6, /**< Right-to-Left Embedding */
UTF8PROC_BIDI_CLASS_RLO = 7, /**< Right-to-Left Override */
UTF8PROC_BIDI_CLASS_PDF = 8, /**< Pop Directional Format */
UTF8PROC_BIDI_CLASS_EN = 9, /**< European Number */
UTF8PROC_BIDI_CLASS_ES = 10, /**< European Separator */
UTF8PROC_BIDI_CLASS_ET = 11, /**< European Number Terminator */
UTF8PROC_BIDI_CLASS_AN = 12, /**< Arabic Number */
UTF8PROC_BIDI_CLASS_CS = 13, /**< Common Number Separator */
UTF8PROC_BIDI_CLASS_NSM = 14, /**< Nonspacing Mark */
UTF8PROC_BIDI_CLASS_BN = 15, /**< Boundary Neutral */
UTF8PROC_BIDI_CLASS_B = 16, /**< Paragraph Separator */
UTF8PROC_BIDI_CLASS_S = 17, /**< Segment Separator */
UTF8PROC_BIDI_CLASS_WS = 18, /**< Whitespace */
UTF8PROC_BIDI_CLASS_ON = 19, /**< Other Neutrals */
UTF8PROC_BIDI_CLASS_LRI = 20, /**< Left-to-Right Isolate */
UTF8PROC_BIDI_CLASS_RLI = 21, /**< Right-to-Left Isolate */
UTF8PROC_BIDI_CLASS_FSI = 22, /**< First Strong Isolate */
UTF8PROC_BIDI_CLASS_PDI = 23, /**< Pop Directional Isolate */
} utf8proc_bidi_class_t;
/** Decomposition type. */
typedef enum {
UTF8PROC_DECOMP_TYPE_FONT = 1, /**< Font */
UTF8PROC_DECOMP_TYPE_NOBREAK = 2, /**< Nobreak */
UTF8PROC_DECOMP_TYPE_INITIAL = 3, /**< Initial */
UTF8PROC_DECOMP_TYPE_MEDIAL = 4, /**< Medial */
UTF8PROC_DECOMP_TYPE_FINAL = 5, /**< Final */
UTF8PROC_DECOMP_TYPE_ISOLATED = 6, /**< Isolated */
UTF8PROC_DECOMP_TYPE_CIRCLE = 7, /**< Circle */
UTF8PROC_DECOMP_TYPE_SUPER = 8, /**< Super */
UTF8PROC_DECOMP_TYPE_SUB = 9, /**< Sub */
UTF8PROC_DECOMP_TYPE_VERTICAL = 10, /**< Vertical */
UTF8PROC_DECOMP_TYPE_WIDE = 11, /**< Wide */
UTF8PROC_DECOMP_TYPE_NARROW = 12, /**< Narrow */
UTF8PROC_DECOMP_TYPE_SMALL = 13, /**< Small */
UTF8PROC_DECOMP_TYPE_SQUARE = 14, /**< Square */
UTF8PROC_DECOMP_TYPE_FRACTION = 15, /**< Fraction */
UTF8PROC_DECOMP_TYPE_COMPAT = 16, /**< Compat */
} utf8proc_decomp_type_t;
/** Boundclass property. (TR29) */
typedef enum {
UTF8PROC_BOUNDCLASS_START = 0, /**< Start */
UTF8PROC_BOUNDCLASS_OTHER = 1, /**< Other */
UTF8PROC_BOUNDCLASS_CR = 2, /**< Cr */
UTF8PROC_BOUNDCLASS_LF = 3, /**< Lf */
UTF8PROC_BOUNDCLASS_CONTROL = 4, /**< Control */
UTF8PROC_BOUNDCLASS_EXTEND = 5, /**< Extend */
UTF8PROC_BOUNDCLASS_L = 6, /**< L */
UTF8PROC_BOUNDCLASS_V = 7, /**< V */
UTF8PROC_BOUNDCLASS_T = 8, /**< T */
UTF8PROC_BOUNDCLASS_LV = 9, /**< Lv */
UTF8PROC_BOUNDCLASS_LVT = 10, /**< Lvt */
UTF8PROC_BOUNDCLASS_REGIONAL_INDICATOR = 11, /**< Regional indicator */
UTF8PROC_BOUNDCLASS_SPACINGMARK = 12, /**< Spacingmark */
UTF8PROC_BOUNDCLASS_PREPEND = 13, /**< Prepend */
UTF8PROC_BOUNDCLASS_ZWJ = 14, /**< Zero Width Joiner */
/* the following are no longer used in Unicode 11, but we keep
the constants here for backward compatibility */
UTF8PROC_BOUNDCLASS_E_BASE = 15, /**< Emoji Base */
UTF8PROC_BOUNDCLASS_E_MODIFIER = 16, /**< Emoji Modifier */
UTF8PROC_BOUNDCLASS_GLUE_AFTER_ZWJ = 17, /**< Glue_After_ZWJ */
UTF8PROC_BOUNDCLASS_E_BASE_GAZ = 18, /**< E_BASE + GLUE_AFTER_ZJW */
/* the Extended_Pictographic property is used in the Unicode 11
grapheme-boundary rules, so we store it in the boundclass field */
UTF8PROC_BOUNDCLASS_EXTENDED_PICTOGRAPHIC = 19,
UTF8PROC_BOUNDCLASS_E_ZWG = 20, /* UTF8PROC_BOUNDCLASS_EXTENDED_PICTOGRAPHIC + ZWJ */
} utf8proc_boundclass_t;
/** Indic_Conjunct_Break property. (TR44) */
typedef enum {
UTF8PROC_INDIC_CONJUNCT_BREAK_NONE = 0,
UTF8PROC_INDIC_CONJUNCT_BREAK_LINKER = 1,
UTF8PROC_INDIC_CONJUNCT_BREAK_CONSONANT = 2,
UTF8PROC_INDIC_CONJUNCT_BREAK_EXTEND = 3,
} utf8proc_indic_conjunct_break_t;
/**
* Function pointer type passed to utf8proc_map_custom() and
* utf8proc_decompose_custom(), which is used to specify a user-defined
* mapping of codepoints to be applied in conjunction with other mappings.
*/
typedef utf8proc_int32_t (*utf8proc_custom_func)(utf8proc_int32_t codepoint, void *data);
/**
* Array containing the byte lengths of a UTF-8 encoded codepoint based
* on the first byte.
*/
UTF8PROC_DLLEXPORT extern const utf8proc_int8_t utf8proc_utf8class[256];
/**
* Returns the utf8proc API version as a string MAJOR.MINOR.PATCH
* (http://semver.org format), possibly with a "-dev" suffix for
* development versions.
*/
UTF8PROC_DLLEXPORT const char *utf8proc_version(void);
/**
* Returns the utf8proc supported Unicode version as a string MAJOR.MINOR.PATCH.
*/
UTF8PROC_DLLEXPORT const char *utf8proc_unicode_version(void);
/**
* Returns an informative error string for the given utf8proc error code
* (e.g. the error codes returned by utf8proc_map()).
*/
UTF8PROC_DLLEXPORT const char *utf8proc_errmsg(utf8proc_ssize_t errcode);
/**
* Reads a single codepoint from the UTF-8 sequence being pointed to by `str`.
* The maximum number of bytes read is `strlen`, unless `strlen` is
* negative (in which case up to 4 bytes are read).
*
* If a valid codepoint could be read, it is stored in the variable
* pointed to by `codepoint_ref`, otherwise that variable will be set to -1.
* In case of success, the number of bytes read is returned; otherwise, a
* negative error code is returned.
*/
UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_iterate(const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, utf8proc_int32_t *codepoint_ref);
/**
* Check if a codepoint is valid (regardless of whether it has been
* assigned a value by the current Unicode standard).
*
* @return 1 if the given `codepoint` is valid and otherwise return 0.
*/
UTF8PROC_DLLEXPORT utf8proc_bool utf8proc_codepoint_valid(utf8proc_int32_t codepoint);
/**
* Encodes the codepoint as an UTF-8 string in the byte array pointed
* to by `dst`. This array must be at least 4 bytes long.
*
* In case of success the number of bytes written is returned, and
* otherwise 0 is returned.
*
* This function does not check whether `codepoint` is valid Unicode.
*/
UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_encode_char(utf8proc_int32_t codepoint, utf8proc_uint8_t *dst);
/**
* Look up the properties for a given codepoint.
*
* @param codepoint The Unicode codepoint.
*
* @returns
* A pointer to a (constant) struct containing information about
* the codepoint.
* @par
* If the codepoint is unassigned or invalid, a pointer to a special struct is
* returned in which `category` is 0 (@ref UTF8PROC_CATEGORY_CN).
*/
UTF8PROC_DLLEXPORT const utf8proc_property_t *utf8proc_get_property(utf8proc_int32_t codepoint);
/** Decompose a codepoint into an array of codepoints.
*
* @param codepoint the codepoint.
* @param dst the destination buffer.
* @param bufsize the size of the destination buffer.
* @param options one or more of the following flags:
* - @ref UTF8PROC_REJECTNA - return an error `codepoint` is unassigned
* - @ref UTF8PROC_IGNORE - strip "default ignorable" codepoints
* - @ref UTF8PROC_CASEFOLD - apply Unicode casefolding
* - @ref UTF8PROC_COMPAT - replace certain codepoints with their
* compatibility decomposition
* - @ref UTF8PROC_CHARBOUND - insert 0xFF bytes before each grapheme cluster
* - @ref UTF8PROC_LUMP - lump certain different codepoints together
* - @ref UTF8PROC_STRIPMARK - remove all character marks
* - @ref UTF8PROC_STRIPNA - remove unassigned codepoints
* @param last_boundclass
* Pointer to an integer variable containing
* the previous codepoint's (boundclass + indic_conjunct_break << 1) if the @ref UTF8PROC_CHARBOUND
* option is used. If the string is being processed in order, this can be initialized to 0 for
* the beginning of the string, and is thereafter updated automatically. Otherwise, this parameter is ignored.
*
* @return
* In case of success, the number of codepoints written is returned; in case
* of an error, a negative error code is returned (utf8proc_errmsg()).
* @par
* If the number of written codepoints would be bigger than `bufsize`, the
* required buffer size is returned, while the buffer will be overwritten with
* undefined data.
*/
UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose_char(
utf8proc_int32_t codepoint, utf8proc_int32_t *dst, utf8proc_ssize_t bufsize,
utf8proc_option_t options, int *last_boundclass
);
/**
* The same as utf8proc_decompose_char(), but acts on a whole UTF-8
* string and orders the decomposed sequences correctly.
*
* If the @ref UTF8PROC_NULLTERM flag in `options` is set, processing
* will be stopped, when a NULL byte is encountered, otherwise `strlen`
* bytes are processed. The result (in the form of 32-bit unicode
* codepoints) is written into the buffer being pointed to by
* `buffer` (which must contain at least `bufsize` entries). In case of
* success, the number of codepoints written is returned; in case of an
* error, a negative error code is returned (utf8proc_errmsg()).
* See utf8proc_decompose_custom() to supply additional transformations.
*
* If the number of written codepoints would be bigger than `bufsize`, the
* required buffer size is returned, while the buffer will be overwritten with
* undefined data.
*/
UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose(
const utf8proc_uint8_t *str, utf8proc_ssize_t strlen,
utf8proc_int32_t *buffer, utf8proc_ssize_t bufsize, utf8proc_option_t options
);
/**
* The same as utf8proc_decompose(), but also takes a `custom_func` mapping function
* that is called on each codepoint in `str` before any other transformations
* (along with a `custom_data` pointer that is passed through to `custom_func`).
* The `custom_func` argument is ignored if it is `NULL`. See also utf8proc_map_custom().
*/
UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose_custom(
const utf8proc_uint8_t *str, utf8proc_ssize_t strlen,
utf8proc_int32_t *buffer, utf8proc_ssize_t bufsize, utf8proc_option_t options,
utf8proc_custom_func custom_func, void *custom_data
);
/**
* Normalizes the sequence of `length` codepoints pointed to by `buffer`
* in-place (i.e., the result is also stored in `buffer`).
*
* @param buffer the (native-endian UTF-32) unicode codepoints to re-encode.
* @param length the length (in codepoints) of the buffer.
* @param options a bitwise or (`|`) of one or more of the following flags:
* - @ref UTF8PROC_NLF2LS - convert LF, CRLF, CR and NEL into LS
* - @ref UTF8PROC_NLF2PS - convert LF, CRLF, CR and NEL into PS
* - @ref UTF8PROC_NLF2LF - convert LF, CRLF, CR and NEL into LF
* - @ref UTF8PROC_STRIPCC - strip or convert all non-affected control characters
* - @ref UTF8PROC_COMPOSE - try to combine decomposed codepoints into composite
* codepoints
* - @ref UTF8PROC_STABLE - prohibit combining characters that would violate
* the unicode versioning stability
*
* @return
* In case of success, the length (in codepoints) of the normalized UTF-32 string is
* returned; otherwise, a negative error code is returned (utf8proc_errmsg()).
*
* @warning The entries of the array pointed to by `str` have to be in the
* range `0x0000` to `0x10FFFF`. Otherwise, the program might crash!
*/
UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_normalize_utf32(utf8proc_int32_t *buffer, utf8proc_ssize_t length, utf8proc_option_t options);
/**
* Reencodes the sequence of `length` codepoints pointed to by `buffer`
* UTF-8 data in-place (i.e., the result is also stored in `buffer`).
* Can optionally normalize the UTF-32 sequence prior to UTF-8 conversion.
*
* @param buffer the (native-endian UTF-32) unicode codepoints to re-encode.
* @param length the length (in codepoints) of the buffer.
* @param options a bitwise or (`|`) of one or more of the following flags:
* - @ref UTF8PROC_NLF2LS - convert LF, CRLF, CR and NEL into LS
* - @ref UTF8PROC_NLF2PS - convert LF, CRLF, CR and NEL into PS
* - @ref UTF8PROC_NLF2LF - convert LF, CRLF, CR and NEL into LF
* - @ref UTF8PROC_STRIPCC - strip or convert all non-affected control characters
* - @ref UTF8PROC_COMPOSE - try to combine decomposed codepoints into composite
* codepoints
* - @ref UTF8PROC_STABLE - prohibit combining characters that would violate
* the unicode versioning stability
* - @ref UTF8PROC_CHARBOUND - insert 0xFF bytes before each grapheme cluster
*
* @return
* In case of success, the length (in bytes) of the resulting nul-terminated
* UTF-8 string is returned; otherwise, a negative error code is returned
* (utf8proc_errmsg()).
*
* @warning The amount of free space pointed to by `buffer` must
* exceed the amount of the input data by one byte, and the
* entries of the array pointed to by `str` have to be in the
* range `0x0000` to `0x10FFFF`. Otherwise, the program might crash!
*/
UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_reencode(utf8proc_int32_t *buffer, utf8proc_ssize_t length, utf8proc_option_t options);
/**
* Given a pair of consecutive codepoints, return whether a grapheme break is
* permitted between them (as defined by the extended grapheme clusters in UAX#29).
*
* @param codepoint1 The first codepoint.
* @param codepoint2 The second codepoint, occurring consecutively after `codepoint1`.
* @param state Beginning with Version 29 (Unicode 9.0.0), this algorithm requires
* state to break graphemes. This state can be passed in as a pointer
* in the `state` argument and should initially be set to 0. If the
* state is not passed in (i.e. a null pointer is passed), UAX#29 rules
* GB10/12/13 which require this state will not be applied, essentially
* matching the rules in Unicode 8.0.0.
*
* @warning If the state parameter is used, `utf8proc_grapheme_break_stateful` must
* be called IN ORDER on ALL potential breaks in a string. However, it
* is safe to reset the state to zero after a grapheme break.
*/
UTF8PROC_DLLEXPORT utf8proc_bool utf8proc_grapheme_break_stateful(
utf8proc_int32_t codepoint1, utf8proc_int32_t codepoint2, utf8proc_int32_t *state);
/**
* Same as utf8proc_grapheme_break_stateful(), except without support for the
* Unicode 9 additions to the algorithm. Supported for legacy reasons.
*/
UTF8PROC_DLLEXPORT utf8proc_bool utf8proc_grapheme_break(
utf8proc_int32_t codepoint1, utf8proc_int32_t codepoint2);
/**
* Given a codepoint `c`, return the codepoint of the corresponding
* lower-case character, if any; otherwise (if there is no lower-case
* variant, or if `c` is not a valid codepoint) return `c`.
*/
UTF8PROC_DLLEXPORT utf8proc_int32_t utf8proc_tolower(utf8proc_int32_t c);
/**
* Given a codepoint `c`, return the codepoint of the corresponding
* upper-case character, if any; otherwise (if there is no upper-case
* variant, or if `c` is not a valid codepoint) return `c`.
*/
UTF8PROC_DLLEXPORT utf8proc_int32_t utf8proc_toupper(utf8proc_int32_t c);
/**
* Given a codepoint `c`, return the codepoint of the corresponding
* title-case character, if any; otherwise (if there is no title-case
* variant, or if `c` is not a valid codepoint) return `c`.
*/
UTF8PROC_DLLEXPORT utf8proc_int32_t utf8proc_totitle(utf8proc_int32_t c);
/**
* Given a codepoint `c`, return `1` if the codepoint corresponds to a lower-case character
* and `0` otherwise.
*/
UTF8PROC_DLLEXPORT int utf8proc_islower(utf8proc_int32_t c);
/**
* Given a codepoint `c`, return `1` if the codepoint corresponds to an upper-case character
* and `0` otherwise.
*/
UTF8PROC_DLLEXPORT int utf8proc_isupper(utf8proc_int32_t c);
/**
* Given a codepoint, return a character width analogous to `wcwidth(codepoint)`,
* except that a width of 0 is returned for non-printable codepoints
* instead of -1 as in `wcwidth`.
*
* @note
* If you want to check for particular types of non-printable characters,
* (analogous to `isprint` or `iscntrl`), use utf8proc_category(). */
UTF8PROC_DLLEXPORT int utf8proc_charwidth(utf8proc_int32_t codepoint);
/**
* Given a codepoint, return whether it has East Asian width class A (Ambiguous)
*
* Codepoints with this property are considered to have charwidth 1 (if they are printable)
* but some East Asian fonts render them as double width.
*/
UTF8PROC_DLLEXPORT utf8proc_bool utf8proc_charwidth_ambiguous(utf8proc_int32_t codepoint);
/**
* Return the Unicode category for the codepoint (one of the
* @ref utf8proc_category_t constants.)
*/
UTF8PROC_DLLEXPORT utf8proc_category_t utf8proc_category(utf8proc_int32_t codepoint);
/**
* Return the two-letter (nul-terminated) Unicode category string for
* the codepoint (e.g. `"Lu"` or `"Co"`).
*/
UTF8PROC_DLLEXPORT const char *utf8proc_category_string(utf8proc_int32_t codepoint);
/**
* Maps the given UTF-8 string pointed to by `str` to a new UTF-8
* string, allocated dynamically by `malloc` and returned via `dstptr`.
*
* If the @ref UTF8PROC_NULLTERM flag in the `options` field is set,
* the length is determined by a NULL terminator, otherwise the
* parameter `strlen` is evaluated to determine the string length, but
* in any case the result will be NULL terminated (though it might
* contain NULL characters with the string if `str` contained NULL
* characters). Other flags in the `options` field are passed to the
* functions defined above, and regarded as described. See also
* utf8proc_map_custom() to supply a custom codepoint transformation.
*
* In case of success the length of the new string is returned,
* otherwise a negative error code is returned.
*
* @note The memory of the new UTF-8 string will have been allocated
* with `malloc`, and should therefore be deallocated with `free`.
*/
UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_map(
const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, utf8proc_uint8_t **dstptr, utf8proc_option_t options
);
/**
* Like utf8proc_map(), but also takes a `custom_func` mapping function
* that is called on each codepoint in `str` before any other transformations
* (along with a `custom_data` pointer that is passed through to `custom_func`).
* The `custom_func` argument is ignored if it is `NULL`.
*/
UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_map_custom(
const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, utf8proc_uint8_t **dstptr, utf8proc_option_t options,
utf8proc_custom_func custom_func, void *custom_data
);
/** @name Unicode normalization
*
* Returns a pointer to newly allocated memory of a NFD, NFC, NFKD, NFKC or
* NFKC_Casefold normalized version of the null-terminated string `str`. These
* are shortcuts to calling utf8proc_map() with @ref UTF8PROC_NULLTERM
* combined with @ref UTF8PROC_STABLE and flags indicating the normalization.
*/
/** @{ */
/** NFD normalization (@ref UTF8PROC_DECOMPOSE). */
UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFD(const utf8proc_uint8_t *str);
/** NFC normalization (@ref UTF8PROC_COMPOSE). */
UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFC(const utf8proc_uint8_t *str);
/** NFKD normalization (@ref UTF8PROC_DECOMPOSE and @ref UTF8PROC_COMPAT). */
UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFKD(const utf8proc_uint8_t *str);
/** NFKC normalization (@ref UTF8PROC_COMPOSE and @ref UTF8PROC_COMPAT). */
UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFKC(const utf8proc_uint8_t *str);
/**
* NFKC_Casefold normalization (@ref UTF8PROC_COMPOSE and @ref UTF8PROC_COMPAT
* and @ref UTF8PROC_CASEFOLD and @ref UTF8PROC_IGNORE).
**/
UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFKC_Casefold(const utf8proc_uint8_t *str);
/** @} */
#ifdef __cplusplus
}
#endif
#endif

16986
Foundation/src/utf8proc_data.c Normal file
View File

File diff suppressed because it is too large Load Diff

22
Foundation/testsuite/src/UTF8StringTest.cpp
View File

@ -12,7 +12,7 @@
#include "CppUnit/TestCaller.h"
#include "CppUnit/TestSuite.h"
#include "Poco/UTF8String.h"
#include <iostream>
using Poco::UTF8;
@ -104,6 +104,25 @@ void UTF8StringTest::testUnescape()
}
void UTF8StringTest::testNormalize()
{
const std::string INPUT("\303\274\303\266\303\244"); // "u"o"a
const std::string NFD("\165\314\210\157\314\210\141\314\210");
std::string nfd = UTF8::normalize(INPUT, UTF8::NORMALIZATION_FORM_D);
assertTrue (nfd == NFD);
nfd = UTF8::normalize(INPUT.begin(), INPUT.end(), UTF8::NORMALIZATION_FORM_D);
assertTrue (nfd == NFD);
std::string nfc = UTF8::normalize(nfd, UTF8::NORMALIZATION_FORM_C);
assertTrue (nfc == INPUT);
nfc = UTF8::normalize(nfd.begin(), nfd.end(), UTF8::NORMALIZATION_FORM_C);
assertTrue (nfc == INPUT);
}
void UTF8StringTest::setUp()
{
}
@ -122,6 +141,7 @@ CppUnit::Test* UTF8StringTest::suite()
CppUnit_addTest(pSuite, UTF8StringTest, testTransform);
CppUnit_addTest(pSuite, UTF8StringTest, testEscape);
CppUnit_addTest(pSuite, UTF8StringTest, testUnescape);
CppUnit_addTest(pSuite, UTF8StringTest, testNormalize);
return pSuite;
}

2
Foundation/testsuite/src/UTF8StringTest.h
View File

@ -30,6 +30,8 @@ public:
void testEscape();
void testUnescape();
void testNormalize();
void setUp();
void tearDown();

60
cmake/FindUtf8Proc.cmake Normal file
View File

@ -0,0 +1,60 @@
#.rst:
# FindUtf8Proc
# --------
#
# Find utf8proc
#
# Find the UTF-8 processing library
#
# ::
#
# This module defines the following variables:
# UTF8PROC_FOUND - True if UTF8PROC_INCLUDE_DIR & UTF8PROC_LIBRARY are found
# UTF8PROC_LIBRARIES - Set when UTF8PROC_LIBRARY is found
# UTF8PROC_INCLUDE_DIRS - Set when UTF8PROC_INCLUDE_DIR is found
#
#
#
# ::
#
# UTF8PROC_INCLUDE_DIR - where to find utf8proc.h
# UTF8PROC_LIBRARY - the utf8proc library
#=============================================================================
# This module is adapted from FindALSA.cmake. Below are the original license
# header.
#=============================================================================
# Copyright 2009-2011 Kitware, Inc.
# Copyright 2009-2011 Philip Lowman <philip@yhbt.com>
#
# Distributed under the OSI-approved BSD License (the "License");
# see accompanying file Copyright.txt for details.
#
# This software is distributed WITHOUT ANY WARRANTY; without even the
# implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
# See the License for more information.
#=============================================================================
find_path(
UTF8PROC_INCLUDE_DIR NAMES utf8proc.h DOC "The utf8proc include directory"
)
find_library(
UTF8PROC_LIBRARY NAMES utf8proc DOC "The utf8proc library"
)
# handle the QUIETLY and REQUIRED arguments and set UTF8PROC_FOUND to TRUE if
# all listed variables are TRUE
include(FindPackageHandleStandardArgs)
FIND_PACKAGE_HANDLE_STANDARD_ARGS(
Utf8Proc
FOUND_VAR Utf8Proc_FOUND
REQUIRED_VARS UTF8PROC_LIBRARY UTF8PROC_INCLUDE_DIR
)
if(Utf8Proc_FOUND)
set( UTF8PROC_LIBRARIES ${UTF8PROC_LIBRARY} )
set( UTF8PROC_INCLUDE_DIRS ${UTF8PROC_INCLUDE_DIR} )
endif()
mark_as_advanced(UTF8PROC_INCLUDE_DIR UTF8PROC_LIBRARY)