2 Copyright (C) 2014-2020 Carl Hetherington <cth@carlh.net>
4 This file is part of DCP-o-matic.
6 DCP-o-matic is free software; you can redistribute it and/or modify
7 it under the terms of the GNU General Public License as published by
8 the Free Software Foundation; either version 2 of the License, or
9 (at your option) any later version.
11 DCP-o-matic is distributed in the hope that it will be useful,
12 but WITHOUT ANY WARRANTY; without even the implied warranty of
13 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 GNU General Public License for more details.
16 You should have received a copy of the GNU General Public License
17 along with DCP-o-matic. If not, see <http://www.gnu.org/licenses/>.
23 #include "exceptions.h"
24 #include "string_text_file.h"
25 #include "string_text_file_content.h"
27 #include <sub/collect.h>
28 #include <sub/ssa_reader.h>
29 #include <sub/stl_binary_reader.h>
30 #include <sub/subrip_reader.h>
31 #include <unicode/ucsdet.h>
32 #include <unicode/ucnv.h>
39 using std::shared_ptr;
42 using boost::scoped_array;
43 using boost::optional;
45 using namespace dcpomatic;
48 StringTextFile::StringTextFile (shared_ptr<const StringTextFileContent> content)
50 string ext = content->path(0).extension().string();
51 transform (ext.begin(), ext.end(), ext.begin(), ::tolower);
53 std::unique_ptr<sub::Reader> reader;
56 dcp::File f(content->path(0), "rb");
58 throw OpenFileError (f.path(), errno, OpenFileError::READ);
61 reader.reset(new sub::STLBinaryReader(f.get()));
67 /* Text-based file; sort out its character encoding before we try to parse it */
69 ArrayData in (content->path (0));
71 UErrorCode status = U_ZERO_ERROR;
72 UCharsetDetector* detector = ucsdet_open (&status);
73 ucsdet_setText (detector, reinterpret_cast<const char *>(in.data()), in.size(), &status);
75 UCharsetMatch const * match = ucsdet_detect (detector, &status);
76 char const * in_charset = ucsdet_getName (match, &status);
78 UConverter* to_utf16 = ucnv_open (in_charset, &status);
79 /* This is a guess; I think we should be able to encode any input in 4 times its input size */
80 scoped_array<uint16_t> utf16 (new uint16_t[in.size() * 2]);
81 int const utf16_len = ucnv_toUChars (
82 to_utf16, reinterpret_cast<UChar*>(utf16.get()), in.size() * 2,
83 reinterpret_cast<const char *>(in.data()), in.size(),
87 UConverter* to_utf8 = ucnv_open ("UTF-8", &status);
89 scoped_array<char> utf8 (new char[utf16_len * 2]);
90 ucnv_fromUChars (to_utf8, utf8.get(), utf16_len * 2, reinterpret_cast<UChar*>(utf16.get()), utf16_len, &status);
92 /* Fix OS X line endings */
93 size_t utf8_len = strlen (utf8.get ());
94 for (size_t i = 0; i < utf8_len; ++i) {
95 if (utf8[i] == '\r' && ((i == utf8_len - 1) || utf8[i + 1] != '\n')) {
100 ucsdet_close (detector);
101 ucnv_close (to_utf16);
102 ucnv_close (to_utf8);
105 reader.reset(new sub::SubripReader(utf8.get()));
106 } else if (ext == ".ssa" || ext == ".ass") {
107 reader.reset(new sub::SSAReader(utf8.get()));
112 _subtitles = sub::collect<vector<sub::Subtitle>>(reader->subtitles());
116 /** @return time of first subtitle, if there is one */
117 optional<ContentTime>
118 StringTextFile::first () const
120 if (_subtitles.empty()) {
124 return ContentTime::from_seconds(_subtitles[0].from.all_as_seconds());
128 StringTextFile::length () const
130 if (_subtitles.empty ()) {
134 return ContentTime::from_seconds (_subtitles.back().to.all_as_seconds ());