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 std::stringstream;
41 using boost::lexical_cast;
42 using boost::to_upper;
43 using boost::optional;
44 using boost::function;
47 /** @param s Subtitle string encoded in UTF-8 */
48 SubripReader::SubripReader (string const & s)
51 this->read (boost::bind (&get_line_stringstream, &str));
54 /** @param f Subtitle file encoded in UTF-8 */
55 SubripReader::SubripReader (FILE* f)
57 this->read (boost::bind (&get_line_file, f));
61 SubripReader::read (function<optional<string> ()> get_line)
76 optional<string> line = get_line ();
81 trim_right_if (*line, boost::is_any_of ("\n\r"));
84 line->length() >= 3 &&
85 static_cast<unsigned char> (line.get()[0]) == 0xef &&
86 static_cast<unsigned char> (line.get()[1]) == 0xbb &&
87 static_cast<unsigned char> (line.get()[2]) == 0xbf
90 /* Skip Unicode byte order mark */
91 line = line->substr (3);
98 /* a blank line at the start is ok */
108 boost::algorithm::split (p, *line, boost::algorithm::is_any_of (" "));
109 if (p.size() != 3 && p.size() != 7) {
110 throw SubripError (*line, "a time/position line");
113 from = convert_time (p[0]);
114 to = convert_time (p[2]);
116 /* XXX: should not ignore coordinate specifications */
122 if (line->empty ()) {
126 convert_line (*line, line_number, from, to);
135 SubripReader::convert_time (string t)
138 boost::algorithm::split (a, t, boost::is_any_of (":"));
140 throw SubripError (t, "time in the format h:m:s,ms");
144 boost::algorithm::split (b, a[2], boost::is_any_of (","));
146 return Time::from_hms (
147 lexical_cast<int> (a[0]),
148 lexical_cast<int> (a[1]),
149 lexical_cast<int> (b[0]),
150 lexical_cast<int> (b[1])
155 SubripReader::convert_line (string t, int line_number, Time from, Time to)
166 p.font_size.set_points (48);
169 p.vertical_position.line = line_number;
171 p.vertical_position.lines = 32;
172 p.vertical_position.reference = TOP_OF_SUBTITLE;
174 list<Colour> colours;
175 colours.push_back (Colour (1, 1, 1));
177 /* XXX: missing <font> support */
178 /* XXX: nesting of tags e.g. <b>foo<i>bar<b>baz</b>fred</i>jim</b> might
182 for (size_t i = 0; i < t.size(); ++i) {
185 if (t[i] == '<' || t[i] == '{') {
192 if (t[i] == '>' || t[i] == '}') {
196 } else if (tag == "/b") {
199 } else if (tag == "i") {
202 } else if (tag == "/i") {
205 } else if (tag == "u") {
208 } else if (tag == "/u") {
211 } else if (boost::starts_with (tag, "font")) {
213 boost::regex re (".*color=\"#([0123456789abcdef]+)\"");
215 if (boost::regex_search (tag, match, re) && string (match[1]).size() == 6) {
216 p.colour = Colour::from_rgb_hex (match[1]);
217 colours.push_back (p.colour);
219 } else if (tag == "/font") {
222 p.colour = colours.back ();
237 SubripReader::maybe_content (RawSubtitle& p)
239 if (!p.text.empty ()) {