2 Copyright (C) 2014-2015 Carl Hetherington <cth@carlh.net>
4 This program is free software; you can redistribute it and/or modify
5 it under the terms of the GNU General Public License as published by
6 the Free Software Foundation; either version 2 of the License, or
7 (at your option) any later version.
9 This program is distributed in the hope that it will be useful,
10 but WITHOUT ANY WARRANTY; without even the implied warranty of
11 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12 GNU General Public License for more details.
14 You should have received a copy of the GNU General Public License
15 along with this program; if not, write to the Free Software
16 Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
20 /** @file src/subrip_reader.cc
21 * @brief SubripReader class.
24 #include "subrip_reader.h"
25 #include "exceptions.h"
27 #include <boost/algorithm/string.hpp>
28 #include <boost/lexical_cast.hpp>
29 #include <boost/regex.hpp>
30 #include <boost/bind.hpp>
40 using boost::lexical_cast;
41 using boost::to_upper;
42 using boost::optional;
43 using boost::function;
44 using boost::algorithm::replace_all;
47 /** @param s Subtitle string encoded in UTF-8 */
48 SubripReader::SubripReader (string s)
50 this->read (boost::bind(&get_line_string, &s));
53 /** @param f Subtitle file encoded in UTF-8 */
54 SubripReader::SubripReader (FILE* f)
56 this->read (boost::bind (&get_line_file, f));
60 SubripReader::read (function<optional<string> ()> get_line)
70 /* This reader extracts no information about where the subtitle
71 should be on screen, so its reference is TOP_OF_SUBTITLE.
73 rs.vertical_position.line = 0;
74 rs.vertical_position.reference = TOP_OF_SUBTITLE;
77 optional<string> line = get_line ();
82 trim_right_if (*line, boost::is_any_of ("\n\r"));
83 remove_unicode_bom (line);
85 /* Keep some history in case there is an error to report */
86 _context.push_back (*line);
87 if (_context.size() > 5) {
88 _context.pop_front ();
95 /* a blank line at the start is ok */
101 /* Reset stuff that should not persist across separate subtitles */
104 rs.underline = false;
105 rs.vertical_position.line = 0;
112 /* Further trim this line, removing spaces from the end */
113 trim_right_if (*line, boost::is_any_of (" "));
115 boost::algorithm::split (p, *line, boost::algorithm::is_any_of (" "), boost::token_compress_on);
116 if (p.size() != 3 && p.size() != 7) {
117 for (int i = 0; i < 2; ++i) {
118 optional<string> ex = get_line ();
120 _context.push_back (*ex);
123 throw SubripError (*line, "a time/position line", _context);
126 rs.from = convert_time (p[0]);
127 rs.to = convert_time (p[2]);
129 /* XXX: should not ignore coordinate specifications */
135 if (line->empty ()) {
138 convert_line (*line, rs);
139 rs.vertical_position.line = rs.vertical_position.line.get() + 1;
147 SubripReader::convert_time (string t)
150 boost::algorithm::split (a, t, boost::is_any_of (":"));
152 throw SubripError (t, "time in the format h:m:s,ms", _context);
156 boost::algorithm::split (b, a[2], boost::is_any_of (","));
158 throw SubripError (t, "time in the format h:m:s,ms", _context);
161 return Time::from_hms (
162 lexical_cast<int> (a[0]),
163 lexical_cast<int> (a[1]),
164 lexical_cast<int> (b[0]),
165 lexical_cast<int> (b[1])
170 SubripReader::convert_line (string t, RawSubtitle& p)
179 list<Colour> colours;
180 colours.push_back (Colour (1, 1, 1));
182 /* XXX: missing <font> support */
183 /* XXX: nesting of tags e.g. <b>foo<i>bar<b>baz</b>fred</i>jim</b> might
187 for (size_t i = 0; i < t.size(); ++i) {
190 if (t[i] == '<' || t[i] == '{') {
197 if (t[i] == '>' || t[i] == '}') {
201 } else if (tag == "/b") {
204 } else if (tag == "i") {
207 } else if (tag == "/i") {
210 } else if (tag == "u") {
213 } else if (tag == "/u") {
216 } else if (boost::starts_with (tag, "font")) {
218 boost::regex re (".*color=\"#([0123456789abcdef]+)\"");
220 if (boost::regex_search (tag, match, re) && string (match[1]).size() == 6) {
221 p.colour = Colour::from_rgb_hex (match[1]);
222 colours.push_back (p.colour);
224 } else if (tag == "/font") {
227 p.colour = colours.back ();
232 tag += tolower (t[i]);
238 /* Strip Unicode U+202B (right-to-left embedding) as sometimes it is rendered
239 as a missing character. This may be a hack.
241 replace_all (p.text, "\xe2\x80\xab", "");
246 /* Push p into _subs if it has some text, and clear the text out of p */
248 SubripReader::maybe_content (RawSubtitle& p)
250 if (!p.text.empty ()) {