From c1ddb1febeb31a79d8f69634575bcc36f38103d4 Mon Sep 17 00:00:00 2001 From: "Matthias P. Braendli" Date: Thu, 23 Apr 2015 14:34:24 +0200 Subject: mot-encoder: Add charset converter for EBU Latin --- src/charset-test.cpp | 57 +++++++++ src/charset.h | 104 ++++++++++++++++ src/mot-encoder.cpp | 98 ++++++++------- src/utf8.h | 34 ++++++ src/utf8/checked.h | 327 ++++++++++++++++++++++++++++++++++++++++++++++++++ src/utf8/core.h | 329 +++++++++++++++++++++++++++++++++++++++++++++++++++ src/utf8/unchecked.h | 228 +++++++++++++++++++++++++++++++++++ 7 files changed, 1135 insertions(+), 42 deletions(-) create mode 100644 src/charset-test.cpp create mode 100644 src/charset.h create mode 100644 src/utf8.h create mode 100644 src/utf8/checked.h create mode 100644 src/utf8/core.h create mode 100644 src/utf8/unchecked.h diff --git a/src/charset-test.cpp b/src/charset-test.cpp new file mode 100644 index 0000000..d65fcc7 --- /dev/null +++ b/src/charset-test.cpp @@ -0,0 +1,57 @@ +/* + Copyright (C) 2015 Matthias P. Braendli (http://opendigitalradio.org) + + This program is free software: you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation, either version 3 of the License, or + (at your option) any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program. If not, see . + + charset.cpp + This is test code for the character set conversion. It does not + get compiled by default. + Please use + g++ -Wall -Wextra -std=c++11 -o charset-test charset-test.cpp + to compile this, and create a input.txt file with test data. + + Authors: + Matthias P. Braendli +*/ + +#include "charset.h" +#include +#include + +#include + +using namespace std; + +int main(int argc, char** argv) +{ + string test_file_path("input.txt"); + + ifstream fs8(test_file_path); + if (!fs8.is_open()) { + cerr << "Could not open " << test_file_path << endl; + return 1; + } + + CharsetConverter conv; + + string line; + // Play with all the lines in the file + while (getline(fs8, line)) { + cout << conv.convert(line) << endl; + } + + return 0; +} + + diff --git a/src/charset.h b/src/charset.h new file mode 100644 index 0000000..0eb1edf --- /dev/null +++ b/src/charset.h @@ -0,0 +1,104 @@ +/* + Copyright (C) 2015 Matthias P. Braendli (http://opendigitalradio.org) + + This program is free software: you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation, either version 3 of the License, or + (at your option) any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program. If not, see . + + charset.h + Define the EBU charsets for DLS encoding + + Authors: + Matthias P. Braendli +*/ + +#ifndef __CHARSET_H_ +#define __CHARSET_H_ + +#include "utf8.h" +#include +#include +#include + +// The first 32 entries are control characters and are not specified in +// the table. +#define CHARSET_TABLE_OFFSET 32 +#define CHARSET_TABLE_ENTRIES (255 - CHARSET_TABLE_OFFSET) +const char* utf8_encoded_EBU_Latin[CHARSET_TABLE_ENTRIES] = { +" ", "!", "\"","#", "¤", "%", "&", "'", "(", ")", "*", "+", ",", "-", ".", "/", +"0", "1", "2", "3", "4", "5", "6", "7", "8", "9", ":", ";", "<", "=", ">", "?", +"@", "A", "B", "C", "D", "E", "F", "G", "H", "I", "J", "K", "L", "M", "N", "O", +"P", "Q", "R", "S", "T", "U", "V", "W", "X", "Y", "Z", "[", "\\","]", "—", "_", +"‖", "a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o", +"p", "q", "r", "s", "t", "u", "v", "w", "x", "y", "z", "{", "|", "}", "⎺", " ", +"á", "à", "é", "è", "í", "ì", "ó", "ò", "ú", "ù", "Ñ", "Ç", "Ş", "ß", "¡", "IJ", +"â", "ä", "ê", "ë", "î", "ï", "ô", "ö", "û", "ü", "ñ", "ç", "ş", "ǧ", "ı", "ij", +"ª", "α", "©", "‰", "Ǧ", "ě", "ň", "ő", "π", "€", "£", "$", "←", "↑", "→", "↓", +"º", "¹", "²", "³", "±", "İ", "ń", "ű", "μ", "¿", "÷", "°", "¼", "½", "¾", "§", +"Á", "À", "Ê", "È", "Í", "Ì", "Ó", "Ò", "Ú", "Ù", "Ř", "Č", "Š", "Ž", "Ð", "Ŀ", +"Â", "Ä", "Ê", "Ë", "Î", "Ï", "Ô", "Ö", "Û", "Ü", "ř", "č", "š", "ž", "đ", "ŀ", +"Ã", "Å", "Æ", "Œ", "ŷ", "Ý", "Õ", "Ø", "Þ", "Ŋ", "Ŕ", "Ć", "Ś", "Ź", "∓", "ð", +"ã", "å", "æ", "œ", "ŵ", "ý", "õ", "ø", "þ", "ŋ", "ŕ", "ć", "ś", "ź", "ł"}; + +class CharsetConverter +{ + public: + CharsetConverter() { + /* Build the converstion table that contains the known code points, + * at the indices corresponding to the EBU Latin table + */ + using namespace std; + for (size_t i = 0; i < CHARSET_TABLE_ENTRIES; i++) { + string table_entry(utf8_encoded_EBU_Latin[i]); + string::iterator it = table_entry.begin(); + uint32_t code_point = utf8::next(it, table_entry.end()); + m_conversion_table.push_back(code_point); + } + } + + /* Convert a UTF-8 encoded text line into a EBU Latin 1 encoded byte stream + */ + std::string convert(std::string line_utf8) { + using namespace std; + + // check for invalid utf-8, we only convert up to the first error + string::iterator end_it = utf8::find_invalid(line_utf8.begin(), line_utf8.end()); + + // Convert it to utf-32 + vector utf32line; + utf8::utf8to32(line_utf8.begin(), end_it, back_inserter(utf32line)); + + string encoded_line(utf32line.size(), '0'); + + // Try to convert each codepoint + for (size_t i = 0; i < utf32line.size(); i++) { + vector::iterator iter = find(m_conversion_table.begin(), + m_conversion_table.end(), utf32line[i]); + if (iter != m_conversion_table.end()) { + size_t index = std::distance(m_conversion_table.begin(), iter); + + encoded_line[i] = (char)(index + CHARSET_TABLE_OFFSET); + } + else { + encoded_line[i] = ' '; + } + } + return encoded_line; + } + + private: + + std::vector m_conversion_table; +}; + +#endif + diff --git a/src/mot-encoder.cpp b/src/mot-encoder.cpp index 8422514..6ddfcb8 100644 --- a/src/mot-encoder.cpp +++ b/src/mot-encoder.cpp @@ -1,7 +1,7 @@ /* Copyright (C) 2014 CSP Innovazione nelle ICT s.c.a r.l. (http://rd.csp.it/) - Copyright (C) 2014 Matthias P. Braendli (http://opendigitalradio.org) + Copyright (C) 2014, 2015 Matthias P. Braendli (http://opendigitalradio.org) Copyright (C) 2015 Stefan Pöschel (http://opendigitalradio.org) @@ -33,6 +33,10 @@ #include #include #include +#include +#include +#include +#include #include #include #include @@ -43,6 +47,7 @@ #include #include #include "config.h" +#include "charset.h" #define DEBUG 0 @@ -225,8 +230,8 @@ void writeMotPAD(int output_fd, unsigned short int mscdgsize, unsigned short int padlen); -void create_dls_pads(const char* text, const int padlen, const uint8_t charset); -void writeDLS(int output_fd, const char* dls_file, int padlen, uint8_t charset); +void create_dls_pads(const std::string& text, const int padlen, const uint8_t charset); +void writeDLS(int output_fd, const std::string& dls_file, int padlen, uint8_t charset); int get_xpadlengthmask(int padlen); @@ -247,10 +252,12 @@ static int cindex_body = 0; #define DLS_SEG_LEN_CRC 2 #define DLS_SEG_LEN_MAX (DLS_SEG_LEN_PREFIX + DLS_SEG_LEN_CHAR_MAX + DLS_SEG_LEN_CRC) +CharsetConverter charset_converter; + typedef std::vector pad_t; static std::deque dls_pads; static bool dls_toggle = false; -static char dlstext_prev[MAXDLS + 1]; +std::string dlstext_prev(MAXDLS + 1, ' '); static int verbose = 0; @@ -311,7 +318,7 @@ int main(int argc, char *argv[]) const char* dir = NULL; const char* output = "/tmp/pad.fifo"; - const char* dls_file = NULL; + std::string dls_file; const struct option longopts[] = { {"charset", required_argument, 0, 'c'}, @@ -377,17 +384,17 @@ int main(int argc, char *argv[]) return 1; } - if (dir && dls_file) { + if (dir && not dls_file.empty()) { fprintf(stderr, "mot-encoder encoding Slideshow from %s and DLS from %s to %s\n", - dir, dls_file, output); + dir, dls_file.c_str(), output); } else if (dir) { fprintf(stderr, "mot-encoder encoding Slideshow from %s to %s. No DLS.\n", dir, output); } - else if (dls_file) { + else if (not dls_file.empty()) { fprintf(stderr, "mot-encoder encoding DLS from %s to %s. No Slideshow.\n", - dls_file, output); + dls_file.c_str(), output); } else { fprintf(stderr, "mot-encoder Error: No DLS nor slideshow to encode !\n"); @@ -471,7 +478,7 @@ int main(int argc, char *argv[]) // Sort the list in fidx order slides_to_transmit.sort(); - if (dls_file) { + if (not dls_file.empty()) { // Maybe we have no slides, always update DLS writeDLS(output_fd, dls_file, padlen, charset); sleep(sleepdelay); @@ -496,7 +503,7 @@ int main(int argc, char *argv[]) } // Always retransmit DLS after each slide, we want it to be updated frequently - if (dls_file) { + if (not dls_file.empty()) { writeDLS(output_fd, dls_file, padlen, charset); } @@ -509,7 +516,7 @@ int main(int argc, char *argv[]) slides_to_transmit.resize(0); } - else if (dls_file) { // only DLS + else if (not dls_file.empty()) { // only DLS // Always retransmit DLS, we want it to be updated frequently writeDLS(output_fd, dls_file, padlen, charset); @@ -893,58 +900,65 @@ void packMscDG(unsigned char* b, MSCDG* msc, unsigned short int* bsize) } -void writeDLS(int output_fd, const char* dls_file, int padlen, uint8_t charset) +void writeDLS(int output_fd, const std::string& dls_file, int padlen, uint8_t charset) { - char dlstext[MAXDLS + 1]; - int dlslen; - int i; - bool dlstext_new; - - int dlsfd = open(dls_file, O_RDONLY); - if (dlsfd == -1) { - fprintf(stderr, "mot-encoder Error: Cannot open dls file\n"); + std::ifstream dls_fstream(dls_file.c_str()); + if (!dls_fstream.is_open()) { + std::cerr << "Could not open " << dls_file << std::endl; return; } - dlslen = read(dlsfd, dlstext, MAXDLS); + std::vector dls_lines; - close(dlsfd); - - dlstext[dlslen] = 0x00; + std::string line; + // Read and convert lines one by one because the converter doesn't understand + // line endings + while (std::getline(dls_fstream, line)) { + if (not line.empty()) { + if (charset == CHARSET_COMPLETE_EBU_LATIN) { + dls_lines.push_back(charset_converter.convert(line)); + } + else { + dls_lines.push_back(line); + } + // TODO handle the other charsets accordingly + } + } - // Remove trailing line breaks from the file - char* endp = dlstext + dlslen; - while ( endp > dlstext && - (*endp == '\0' || *endp == '\n')) { - if (*endp == '\n') { - *endp = '\0'; + std::stringstream ss; + for (size_t i = 0; i < dls_lines.size(); i++) { + if (i != 0) { + ss << "\n"; } - endp--; + ss << dls_lines[i]; } + std::string dlstext = ss.str(); + using namespace std; + // (Re)Create data groups (and thereby toggle the toggle bit) only on (first call or) new text - dlstext_new = dls_pads.empty() || strcmp(dlstext, dlstext_prev); + bool dlstext_is_new = dls_pads.empty() || (dlstext != dlstext_prev); if (verbose) { - fprintf(stderr, "mot-encoder writing %s DLS text \"%s\"\n", dlstext_new ? "new" : "old", dlstext); + fprintf(stderr, "mot-encoder writing %s DLS text \"%s\"\n", dlstext_is_new ? "new" : "old", dlstext.c_str()); } - if (dlstext_new) { + if (dlstext_is_new) { create_dls_pads(dlstext, padlen, charset); - strcpy(dlstext_prev, dlstext); + dlstext_prev = dlstext; } - for (i = 0; i < dls_pads.size(); i++) { + for (size_t i = 0; i < dls_pads.size(); i++) { size_t dummy = write(output_fd, &dls_pads[i].front(), dls_pads[i].size()); } } -int dls_count(const char* text) { - size_t text_len = strlen(text); +int dls_count(const std::string& text) { + size_t text_len = text.size(); return text_len / DLS_SEG_LEN_CHAR_MAX + (text_len % DLS_SEG_LEN_CHAR_MAX ? 1 : 0); } -size_t dls_get(const char* text, const uint8_t charset, const unsigned int seg_index, uint8_t *seg_data) { +size_t dls_get(const std::string& text, const uint8_t charset, const unsigned int seg_index, uint8_t *seg_data) { int seg_count = dls_count(text); if (seg_index >= seg_count) return 0; @@ -952,7 +966,7 @@ size_t dls_get(const char* text, const uint8_t charset, const unsigned int seg_i bool first_seg = seg_index == 0; bool last_seg = seg_index == seg_count - 1; - const char *seg_text_start = text + seg_index * DLS_SEG_LEN_CHAR_MAX; + const char *seg_text_start = text.c_str() + seg_index * DLS_SEG_LEN_CHAR_MAX; size_t seg_text_len = strnlen(seg_text_start, DLS_SEG_LEN_CHAR_MAX); size_t seg_len = DLS_SEG_LEN_PREFIX + seg_text_len + DLS_SEG_LEN_CRC; @@ -991,7 +1005,7 @@ size_t dls_get(const char* text, const uint8_t charset, const unsigned int seg_i } -void create_dls_pads(const char* text, const int padlen, const uint8_t charset) { +void create_dls_pads(const std::string& text, const int padlen, const uint8_t charset) { uint8_t seg_data[DLS_SEG_LEN_MAX]; int xpadlengthmask = get_xpadlengthmask(padlen); diff --git a/src/utf8.h b/src/utf8.h new file mode 100644 index 0000000..4e44514 --- /dev/null +++ b/src/utf8.h @@ -0,0 +1,34 @@ +// Copyright 2006 Nemanja Trifunovic + +/* +Permission is hereby granted, free of charge, to any person or organization +obtaining a copy of the software and accompanying documentation covered by +this license (the "Software") to use, reproduce, display, distribute, +execute, and transmit the Software, and to prepare derivative works of the +Software, and to permit third-parties to whom the Software is furnished to +do so, all subject to the following: + +The copyright notices in the Software and this entire statement, including +the above license grant, this restriction and the following disclaimer, +must be included in all copies of the Software, in whole or in part, and +all derivative works of the Software, unless such copies or derivative +works are solely in the form of machine-executable object code generated by +a source language processor. + +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, TITLE AND NON-INFRINGEMENT. IN NO EVENT +SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE +FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER +DEALINGS IN THE SOFTWARE. +*/ + + +#ifndef UTF8_FOR_CPP_2675DCD0_9480_4c0c_B92A_CC14C027B731 +#define UTF8_FOR_CPP_2675DCD0_9480_4c0c_B92A_CC14C027B731 + +#include "utf8/checked.h" +#include "utf8/unchecked.h" + +#endif // header guard diff --git a/src/utf8/checked.h b/src/utf8/checked.h new file mode 100644 index 0000000..1331155 --- /dev/null +++ b/src/utf8/checked.h @@ -0,0 +1,327 @@ +// Copyright 2006 Nemanja Trifunovic + +/* +Permission is hereby granted, free of charge, to any person or organization +obtaining a copy of the software and accompanying documentation covered by +this license (the "Software") to use, reproduce, display, distribute, +execute, and transmit the Software, and to prepare derivative works of the +Software, and to permit third-parties to whom the Software is furnished to +do so, all subject to the following: + +The copyright notices in the Software and this entire statement, including +the above license grant, this restriction and the following disclaimer, +must be included in all copies of the Software, in whole or in part, and +all derivative works of the Software, unless such copies or derivative +works are solely in the form of machine-executable object code generated by +a source language processor. + +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, TITLE AND NON-INFRINGEMENT. IN NO EVENT +SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE +FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER +DEALINGS IN THE SOFTWARE. +*/ + + +#ifndef UTF8_FOR_CPP_CHECKED_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 +#define UTF8_FOR_CPP_CHECKED_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 + +#include "core.h" +#include + +namespace utf8 +{ + // Base for the exceptions that may be thrown from the library + class exception : public ::std::exception { + }; + + // Exceptions that may be thrown from the library functions. + class invalid_code_point : public exception { + uint32_t cp; + public: + invalid_code_point(uint32_t cp) : cp(cp) {} + virtual const char* what() const throw() { return "Invalid code point"; } + uint32_t code_point() const {return cp;} + }; + + class invalid_utf8 : public exception { + uint8_t u8; + public: + invalid_utf8 (uint8_t u) : u8(u) {} + virtual const char* what() const throw() { return "Invalid UTF-8"; } + uint8_t utf8_octet() const {return u8;} + }; + + class invalid_utf16 : public exception { + uint16_t u16; + public: + invalid_utf16 (uint16_t u) : u16(u) {} + virtual const char* what() const throw() { return "Invalid UTF-16"; } + uint16_t utf16_word() const {return u16;} + }; + + class not_enough_room : public exception { + public: + virtual const char* what() const throw() { return "Not enough space"; } + }; + + /// The library API - functions intended to be called by the users + + template + octet_iterator append(uint32_t cp, octet_iterator result) + { + if (!utf8::internal::is_code_point_valid(cp)) + throw invalid_code_point(cp); + + if (cp < 0x80) // one octet + *(result++) = static_cast(cp); + else if (cp < 0x800) { // two octets + *(result++) = static_cast((cp >> 6) | 0xc0); + *(result++) = static_cast((cp & 0x3f) | 0x80); + } + else if (cp < 0x10000) { // three octets + *(result++) = static_cast((cp >> 12) | 0xe0); + *(result++) = static_cast(((cp >> 6) & 0x3f) | 0x80); + *(result++) = static_cast((cp & 0x3f) | 0x80); + } + else { // four octets + *(result++) = static_cast((cp >> 18) | 0xf0); + *(result++) = static_cast(((cp >> 12) & 0x3f) | 0x80); + *(result++) = static_cast(((cp >> 6) & 0x3f) | 0x80); + *(result++) = static_cast((cp & 0x3f) | 0x80); + } + return result; + } + + template + output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out, uint32_t replacement) + { + while (start != end) { + octet_iterator sequence_start = start; + internal::utf_error err_code = utf8::internal::validate_next(start, end); + switch (err_code) { + case internal::UTF8_OK : + for (octet_iterator it = sequence_start; it != start; ++it) + *out++ = *it; + break; + case internal::NOT_ENOUGH_ROOM: + throw not_enough_room(); + case internal::INVALID_LEAD: + out = utf8::append (replacement, out); + ++start; + break; + case internal::INCOMPLETE_SEQUENCE: + case internal::OVERLONG_SEQUENCE: + case internal::INVALID_CODE_POINT: + out = utf8::append (replacement, out); + ++start; + // just one replacement mark for the sequence + while (start != end && utf8::internal::is_trail(*start)) + ++start; + break; + } + } + return out; + } + + template + inline output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out) + { + static const uint32_t replacement_marker = utf8::internal::mask16(0xfffd); + return utf8::replace_invalid(start, end, out, replacement_marker); + } + + template + uint32_t next(octet_iterator& it, octet_iterator end) + { + uint32_t cp = 0; + internal::utf_error err_code = utf8::internal::validate_next(it, end, cp); + switch (err_code) { + case internal::UTF8_OK : + break; + case internal::NOT_ENOUGH_ROOM : + throw not_enough_room(); + case internal::INVALID_LEAD : + case internal::INCOMPLETE_SEQUENCE : + case internal::OVERLONG_SEQUENCE : + throw invalid_utf8(*it); + case internal::INVALID_CODE_POINT : + throw invalid_code_point(cp); + } + return cp; + } + + template + uint32_t peek_next(octet_iterator it, octet_iterator end) + { + return utf8::next(it, end); + } + + template + uint32_t prior(octet_iterator& it, octet_iterator start) + { + // can't do much if it == start + if (it == start) + throw not_enough_room(); + + octet_iterator end = it; + // Go back until we hit either a lead octet or start + while (utf8::internal::is_trail(*(--it))) + if (it == start) + throw invalid_utf8(*it); // error - no lead byte in the sequence + return utf8::peek_next(it, end); + } + + /// Deprecated in versions that include "prior" + template + uint32_t previous(octet_iterator& it, octet_iterator pass_start) + { + octet_iterator end = it; + while (utf8::internal::is_trail(*(--it))) + if (it == pass_start) + throw invalid_utf8(*it); // error - no lead byte in the sequence + octet_iterator temp = it; + return utf8::next(temp, end); + } + + template + void advance (octet_iterator& it, distance_type n, octet_iterator end) + { + for (distance_type i = 0; i < n; ++i) + utf8::next(it, end); + } + + template + typename std::iterator_traits::difference_type + distance (octet_iterator first, octet_iterator last) + { + typename std::iterator_traits::difference_type dist; + for (dist = 0; first < last; ++dist) + utf8::next(first, last); + return dist; + } + + template + octet_iterator utf16to8 (u16bit_iterator start, u16bit_iterator end, octet_iterator result) + { + while (start != end) { + uint32_t cp = utf8::internal::mask16(*start++); + // Take care of surrogate pairs first + if (utf8::internal::is_lead_surrogate(cp)) { + if (start != end) { + uint32_t trail_surrogate = utf8::internal::mask16(*start++); + if (utf8::internal::is_trail_surrogate(trail_surrogate)) + cp = (cp << 10) + trail_surrogate + internal::SURROGATE_OFFSET; + else + throw invalid_utf16(static_cast(trail_surrogate)); + } + else + throw invalid_utf16(static_cast(cp)); + + } + // Lone trail surrogate + else if (utf8::internal::is_trail_surrogate(cp)) + throw invalid_utf16(static_cast(cp)); + + result = utf8::append(cp, result); + } + return result; + } + + template + u16bit_iterator utf8to16 (octet_iterator start, octet_iterator end, u16bit_iterator result) + { + while (start != end) { + uint32_t cp = utf8::next(start, end); + if (cp > 0xffff) { //make a surrogate pair + *result++ = static_cast((cp >> 10) + internal::LEAD_OFFSET); + *result++ = static_cast((cp & 0x3ff) + internal::TRAIL_SURROGATE_MIN); + } + else + *result++ = static_cast(cp); + } + return result; + } + + template + octet_iterator utf32to8 (u32bit_iterator start, u32bit_iterator end, octet_iterator result) + { + while (start != end) + result = utf8::append(*(start++), result); + + return result; + } + + template + u32bit_iterator utf8to32 (octet_iterator start, octet_iterator end, u32bit_iterator result) + { + while (start != end) + (*result++) = utf8::next(start, end); + + return result; + } + + // The iterator class + template + class iterator : public std::iterator { + octet_iterator it; + octet_iterator range_start; + octet_iterator range_end; + public: + iterator () {} + explicit iterator (const octet_iterator& octet_it, + const octet_iterator& range_start, + const octet_iterator& range_end) : + it(octet_it), range_start(range_start), range_end(range_end) + { + if (it < range_start || it > range_end) + throw std::out_of_range("Invalid utf-8 iterator position"); + } + // the default "big three" are OK + octet_iterator base () const { return it; } + uint32_t operator * () const + { + octet_iterator temp = it; + return utf8::next(temp, range_end); + } + bool operator == (const iterator& rhs) const + { + if (range_start != rhs.range_start || range_end != rhs.range_end) + throw std::logic_error("Comparing utf-8 iterators defined with different ranges"); + return (it == rhs.it); + } + bool operator != (const iterator& rhs) const + { + return !(operator == (rhs)); + } + iterator& operator ++ () + { + utf8::next(it, range_end); + return *this; + } + iterator operator ++ (int) + { + iterator temp = *this; + utf8::next(it, range_end); + return temp; + } + iterator& operator -- () + { + utf8::prior(it, range_start); + return *this; + } + iterator operator -- (int) + { + iterator temp = *this; + utf8::prior(it, range_start); + return temp; + } + }; // class iterator + +} // namespace utf8 + +#endif //header guard + + diff --git a/src/utf8/core.h b/src/utf8/core.h new file mode 100644 index 0000000..693d388 --- /dev/null +++ b/src/utf8/core.h @@ -0,0 +1,329 @@ +// Copyright 2006 Nemanja Trifunovic + +/* +Permission is hereby granted, free of charge, to any person or organization +obtaining a copy of the software and accompanying documentation covered by +this license (the "Software") to use, reproduce, display, distribute, +execute, and transmit the Software, and to prepare derivative works of the +Software, and to permit third-parties to whom the Software is furnished to +do so, all subject to the following: + +The copyright notices in the Software and this entire statement, including +the above license grant, this restriction and the following disclaimer, +must be included in all copies of the Software, in whole or in part, and +all derivative works of the Software, unless such copies or derivative +works are solely in the form of machine-executable object code generated by +a source language processor. + +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, TITLE AND NON-INFRINGEMENT. IN NO EVENT +SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE +FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER +DEALINGS IN THE SOFTWARE. +*/ + + +#ifndef UTF8_FOR_CPP_CORE_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 +#define UTF8_FOR_CPP_CORE_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 + +#include + +namespace utf8 +{ + // The typedefs for 8-bit, 16-bit and 32-bit unsigned integers + // You may need to change them to match your system. + // These typedefs have the same names as ones from cstdint, or boost/cstdint + typedef unsigned char uint8_t; + typedef unsigned short uint16_t; + typedef unsigned int uint32_t; + +// Helper code - not intended to be directly called by the library users. May be changed at any time +namespace internal +{ + // Unicode constants + // Leading (high) surrogates: 0xd800 - 0xdbff + // Trailing (low) surrogates: 0xdc00 - 0xdfff + const uint16_t LEAD_SURROGATE_MIN = 0xd800u; + const uint16_t LEAD_SURROGATE_MAX = 0xdbffu; + const uint16_t TRAIL_SURROGATE_MIN = 0xdc00u; + const uint16_t TRAIL_SURROGATE_MAX = 0xdfffu; + const uint16_t LEAD_OFFSET = LEAD_SURROGATE_MIN - (0x10000 >> 10); + const uint32_t SURROGATE_OFFSET = 0x10000u - (LEAD_SURROGATE_MIN << 10) - TRAIL_SURROGATE_MIN; + + // Maximum valid value for a Unicode code point + const uint32_t CODE_POINT_MAX = 0x0010ffffu; + + template + inline uint8_t mask8(octet_type oc) + { + return static_cast(0xff & oc); + } + template + inline uint16_t mask16(u16_type oc) + { + return static_cast(0xffff & oc); + } + template + inline bool is_trail(octet_type oc) + { + return ((utf8::internal::mask8(oc) >> 6) == 0x2); + } + + template + inline bool is_lead_surrogate(u16 cp) + { + return (cp >= LEAD_SURROGATE_MIN && cp <= LEAD_SURROGATE_MAX); + } + + template + inline bool is_trail_surrogate(u16 cp) + { + return (cp >= TRAIL_SURROGATE_MIN && cp <= TRAIL_SURROGATE_MAX); + } + + template + inline bool is_surrogate(u16 cp) + { + return (cp >= LEAD_SURROGATE_MIN && cp <= TRAIL_SURROGATE_MAX); + } + + template + inline bool is_code_point_valid(u32 cp) + { + return (cp <= CODE_POINT_MAX && !utf8::internal::is_surrogate(cp)); + } + + template + inline typename std::iterator_traits::difference_type + sequence_length(octet_iterator lead_it) + { + uint8_t lead = utf8::internal::mask8(*lead_it); + if (lead < 0x80) + return 1; + else if ((lead >> 5) == 0x6) + return 2; + else if ((lead >> 4) == 0xe) + return 3; + else if ((lead >> 3) == 0x1e) + return 4; + else + return 0; + } + + template + inline bool is_overlong_sequence(uint32_t cp, octet_difference_type length) + { + if (cp < 0x80) { + if (length != 1) + return true; + } + else if (cp < 0x800) { + if (length != 2) + return true; + } + else if (cp < 0x10000) { + if (length != 3) + return true; + } + + return false; + } + + enum utf_error {UTF8_OK, NOT_ENOUGH_ROOM, INVALID_LEAD, INCOMPLETE_SEQUENCE, OVERLONG_SEQUENCE, INVALID_CODE_POINT}; + + /// Helper for get_sequence_x + template + utf_error increase_safely(octet_iterator& it, octet_iterator end) + { + if (++it == end) + return NOT_ENOUGH_ROOM; + + if (!utf8::internal::is_trail(*it)) + return INCOMPLETE_SEQUENCE; + + return UTF8_OK; + } + + #define UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(IT, END) {utf_error ret = increase_safely(IT, END); if (ret != UTF8_OK) return ret;} + + /// get_sequence_x functions decode utf-8 sequences of the length x + template + utf_error get_sequence_1(octet_iterator& it, octet_iterator end, uint32_t& code_point) + { + if (it == end) + return NOT_ENOUGH_ROOM; + + code_point = utf8::internal::mask8(*it); + + return UTF8_OK; + } + + template + utf_error get_sequence_2(octet_iterator& it, octet_iterator end, uint32_t& code_point) + { + if (it == end) + return NOT_ENOUGH_ROOM; + + code_point = utf8::internal::mask8(*it); + + UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end) + + code_point = ((code_point << 6) & 0x7ff) + ((*it) & 0x3f); + + return UTF8_OK; + } + + template + utf_error get_sequence_3(octet_iterator& it, octet_iterator end, uint32_t& code_point) + { + if (it == end) + return NOT_ENOUGH_ROOM; + + code_point = utf8::internal::mask8(*it); + + UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end) + + code_point = ((code_point << 12) & 0xffff) + ((utf8::internal::mask8(*it) << 6) & 0xfff); + + UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end) + + code_point += (*it) & 0x3f; + + return UTF8_OK; + } + + template + utf_error get_sequence_4(octet_iterator& it, octet_iterator end, uint32_t& code_point) + { + if (it == end) + return NOT_ENOUGH_ROOM; + + code_point = utf8::internal::mask8(*it); + + UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end) + + code_point = ((code_point << 18) & 0x1fffff) + ((utf8::internal::mask8(*it) << 12) & 0x3ffff); + + UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end) + + code_point += (utf8::internal::mask8(*it) << 6) & 0xfff; + + UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end) + + code_point += (*it) & 0x3f; + + return UTF8_OK; + } + + #undef UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR + + template + utf_error validate_next(octet_iterator& it, octet_iterator end, uint32_t& code_point) + { + // Save the original value of it so we can go back in case of failure + // Of course, it does not make much sense with i.e. stream iterators + octet_iterator original_it = it; + + uint32_t cp = 0; + // Determine the sequence length based on the lead octet + typedef typename std::iterator_traits::difference_type octet_difference_type; + const octet_difference_type length = utf8::internal::sequence_length(it); + + // Get trail octets and calculate the code point + utf_error err = UTF8_OK; + switch (length) { + case 0: + return INVALID_LEAD; + case 1: + err = utf8::internal::get_sequence_1(it, end, cp); + break; + case 2: + err = utf8::internal::get_sequence_2(it, end, cp); + break; + case 3: + err = utf8::internal::get_sequence_3(it, end, cp); + break; + case 4: + err = utf8::internal::get_sequence_4(it, end, cp); + break; + } + + if (err == UTF8_OK) { + // Decoding succeeded. Now, security checks... + if (utf8::internal::is_code_point_valid(cp)) { + if (!utf8::internal::is_overlong_sequence(cp, length)){ + // Passed! Return here. + code_point = cp; + ++it; + return UTF8_OK; + } + else + err = OVERLONG_SEQUENCE; + } + else + err = INVALID_CODE_POINT; + } + + // Failure branch - restore the original value of the iterator + it = original_it; + return err; + } + + template + inline utf_error validate_next(octet_iterator& it, octet_iterator end) { + uint32_t ignored; + return utf8::internal::validate_next(it, end, ignored); + } + +} // namespace internal + + /// The library API - functions intended to be called by the users + + // Byte order mark + const uint8_t bom[] = {0xef, 0xbb, 0xbf}; + + template + octet_iterator find_invalid(octet_iterator start, octet_iterator end) + { + octet_iterator result = start; + while (result != end) { + utf8::internal::utf_error err_code = utf8::internal::validate_next(result, end); + if (err_code != internal::UTF8_OK) + return result; + } + return result; + } + + template + inline bool is_valid(octet_iterator start, octet_iterator end) + { + return (utf8::find_invalid(start, end) == end); + } + + template + inline bool starts_with_bom (octet_iterator it, octet_iterator end) + { + return ( + ((it != end) && (utf8::internal::mask8(*it++)) == bom[0]) && + ((it != end) && (utf8::internal::mask8(*it++)) == bom[1]) && + ((it != end) && (utf8::internal::mask8(*it)) == bom[2]) + ); + } + + //Deprecated in release 2.3 + template + inline bool is_bom (octet_iterator it) + { + return ( + (utf8::internal::mask8(*it++)) == bom[0] && + (utf8::internal::mask8(*it++)) == bom[1] && + (utf8::internal::mask8(*it)) == bom[2] + ); + } +} // namespace utf8 + +#endif // header guard + + diff --git a/src/utf8/unchecked.h b/src/utf8/unchecked.h new file mode 100644 index 0000000..cb24271 --- /dev/null +++ b/src/utf8/unchecked.h @@ -0,0 +1,228 @@ +// Copyright 2006 Nemanja Trifunovic + +/* +Permission is hereby granted, free of charge, to any person or organization +obtaining a copy of the software and accompanying documentation covered by +this license (the "Software") to use, reproduce, display, distribute, +execute, and transmit the Software, and to prepare derivative works of the +Software, and to permit third-parties to whom the Software is furnished to +do so, all subject to the following: + +The copyright notices in the Software and this entire statement, including +the above license grant, this restriction and the following disclaimer, +must be included in all copies of the Software, in whole or in part, and +all derivative works of the Software, unless such copies or derivative +works are solely in the form of machine-executable object code generated by +a source language processor. + +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, TITLE AND NON-INFRINGEMENT. IN NO EVENT +SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE +FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER +DEALINGS IN THE SOFTWARE. +*/ + + +#ifndef UTF8_FOR_CPP_UNCHECKED_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 +#define UTF8_FOR_CPP_UNCHECKED_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 + +#include "core.h" + +namespace utf8 +{ + namespace unchecked + { + template + octet_iterator append(uint32_t cp, octet_iterator result) + { + if (cp < 0x80) // one octet + *(result++) = static_cast(cp); + else if (cp < 0x800) { // two octets + *(result++) = static_cast((cp >> 6) | 0xc0); + *(result++) = static_cast((cp & 0x3f) | 0x80); + } + else if (cp < 0x10000) { // three octets + *(result++) = static_cast((cp >> 12) | 0xe0); + *(result++) = static_cast(((cp >> 6) & 0x3f) | 0x80); + *(result++) = static_cast((cp & 0x3f) | 0x80); + } + else { // four octets + *(result++) = static_cast((cp >> 18) | 0xf0); + *(result++) = static_cast(((cp >> 12) & 0x3f)| 0x80); + *(result++) = static_cast(((cp >> 6) & 0x3f) | 0x80); + *(result++) = static_cast((cp & 0x3f) | 0x80); + } + return result; + } + + template + uint32_t next(octet_iterator& it) + { + uint32_t cp = utf8::internal::mask8(*it); + typename std::iterator_traits::difference_type length = utf8::internal::sequence_length(it); + switch (length) { + case 1: + break; + case 2: + it++; + cp = ((cp << 6) & 0x7ff) + ((*it) & 0x3f); + break; + case 3: + ++it; + cp = ((cp << 12) & 0xffff) + ((utf8::internal::mask8(*it) << 6) & 0xfff); + ++it; + cp += (*it) & 0x3f; + break; + case 4: + ++it; + cp = ((cp << 18) & 0x1fffff) + ((utf8::internal::mask8(*it) << 12) & 0x3ffff); + ++it; + cp += (utf8::internal::mask8(*it) << 6) & 0xfff; + ++it; + cp += (*it) & 0x3f; + break; + } + ++it; + return cp; + } + + template + uint32_t peek_next(octet_iterator it) + { + return utf8::unchecked::next(it); + } + + template + uint32_t prior(octet_iterator& it) + { + while (utf8::internal::is_trail(*(--it))) ; + octet_iterator temp = it; + return utf8::unchecked::next(temp); + } + + // Deprecated in versions that include prior, but only for the sake of consistency (see utf8::previous) + template + inline uint32_t previous(octet_iterator& it) + { + return utf8::unchecked::prior(it); + } + + template + void advance (octet_iterator& it, distance_type n) + { + for (distance_type i = 0; i < n; ++i) + utf8::unchecked::next(it); + } + + template + typename std::iterator_traits::difference_type + distance (octet_iterator first, octet_iterator last) + { + typename std::iterator_traits::difference_type dist; + for (dist = 0; first < last; ++dist) + utf8::unchecked::next(first); + return dist; + } + + template + octet_iterator utf16to8 (u16bit_iterator start, u16bit_iterator end, octet_iterator result) + { + while (start != end) { + uint32_t cp = utf8::internal::mask16(*start++); + // Take care of surrogate pairs first + if (utf8::internal::is_lead_surrogate(cp)) { + uint32_t trail_surrogate = utf8::internal::mask16(*start++); + cp = (cp << 10) + trail_surrogate + internal::SURROGATE_OFFSET; + } + result = utf8::unchecked::append(cp, result); + } + return result; + } + + template + u16bit_iterator utf8to16 (octet_iterator start, octet_iterator end, u16bit_iterator result) + { + while (start < end) { + uint32_t cp = utf8::unchecked::next(start); + if (cp > 0xffff) { //make a surrogate pair + *result++ = static_cast((cp >> 10) + internal::LEAD_OFFSET); + *result++ = static_cast((cp & 0x3ff) + internal::TRAIL_SURROGATE_MIN); + } + else + *result++ = static_cast(cp); + } + return result; + } + + template + octet_iterator utf32to8 (u32bit_iterator start, u32bit_iterator end, octet_iterator result) + { + while (start != end) + result = utf8::unchecked::append(*(start++), result); + + return result; + } + + template + u32bit_iterator utf8to32 (octet_iterator start, octet_iterator end, u32bit_iterator result) + { + while (start < end) + (*result++) = utf8::unchecked::next(start); + + return result; + } + + // The iterator class + template + class iterator : public std::iterator { + octet_iterator it; + public: + iterator () {} + explicit iterator (const octet_iterator& octet_it): it(octet_it) {} + // the default "big three" are OK + octet_iterator base () const { return it; } + uint32_t operator * () const + { + octet_iterator temp = it; + return utf8::unchecked::next(temp); + } + bool operator == (const iterator& rhs) const + { + return (it == rhs.it); + } + bool operator != (const iterator& rhs) const + { + return !(operator == (rhs)); + } + iterator& operator ++ () + { + ::std::advance(it, utf8::internal::sequence_length(it)); + return *this; + } + iterator operator ++ (int) + { + iterator temp = *this; + ::std::advance(it, utf8::internal::sequence_length(it)); + return temp; + } + iterator& operator -- () + { + utf8::unchecked::prior(it); + return *this; + } + iterator operator -- (int) + { + iterator temp = *this; + utf8::unchecked::prior(it); + return temp; + } + }; // class iterator + + } // namespace utf8::unchecked +} // namespace utf8 + + +#endif // header guard + -- cgit v1.2.3