/*
- Copyright (C) 2014 Carl Hetherington <cth@carlh.net>
+ Copyright (C) 2014-2020 Carl Hetherington <cth@carlh.net>
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
*/
+/** @file src/subrip_reader.cc
+ * @brief SubripReader class.
+ */
+
#include "subrip_reader.h"
#include "exceptions.h"
+#include "util.h"
+#include "sub_assert.h"
+#include "raw_convert.h"
+#include "ssa_reader.h"
#include <boost/algorithm/string.hpp>
#include <boost/lexical_cast.hpp>
#include <boost/regex.hpp>
+#include <boost/bind.hpp>
#include <cstdio>
#include <vector>
+#include <iostream>
using std::string;
using std::vector;
using std::hex;
using boost::lexical_cast;
using boost::to_upper;
+using boost::optional;
+using boost::function;
+using boost::algorithm::replace_all;
using namespace sub;
+/** @param s Subtitle string encoded in UTF-8 */
+SubripReader::SubripReader (string s)
+{
+ this->read (boost::bind(&get_line_string, &s));
+}
+
+/** @param f Subtitle file encoded in UTF-8 */
SubripReader::SubripReader (FILE* f)
+{
+ this->read (boost::bind (&get_line_file, f));
+}
+
+void
+SubripReader::read (function<optional<string> ()> get_line)
{
enum {
COUNTER,
CONTENT
} state = COUNTER;
- char buffer[256];
+ RawSubtitle rs;
- Time from;
- Time to;
+ rs.vertical_position.line = 0;
+ rs.vertical_position.reference = TOP_OF_SUBTITLE;
- string line;
- int line_number = 0;
-
- while (!feof (f)) {
- char* r = fgets (buffer, sizeof (buffer), f);
- if (r == 0 || feof (f)) {
+ while (true) {
+ optional<string> line = get_line ();
+ if (!line) {
break;
}
- line = string (buffer);
- trim_right_if (line, boost::is_any_of ("\n\r"));
-
- if (
- line.length() >= 3 &&
- static_cast<unsigned char> (line[0]) == 0xef &&
- static_cast<unsigned char> (line[1]) == 0xbb &&
- static_cast<unsigned char> (line[2]) == 0xbf
- ) {
+ trim_right_if (*line, boost::is_any_of ("\n\r"));
+ remove_unicode_bom (line);
- /* Skip Unicode byte order mark */
- line = line.substr (3);
+ /* Keep some history in case there is an error to report */
+ _context.push_back (*line);
+ if (_context.size() > 5) {
+ _context.pop_front ();
}
switch (state) {
case COUNTER:
{
- if (line.empty ()) {
+ if (line->empty ()) {
/* a blank line at the start is ok */
break;
}
state = METADATA;
+
+ /* Reset stuff that should not persist across separate subtitles */
+ rs.bold = false;
+ rs.italic = false;
+ rs.underline = false;
+ rs.vertical_position.line = 0;
+ rs.vertical_position.reference = TOP_OF_SUBTITLE;
}
break;
case METADATA:
{
vector<string> p;
- boost::algorithm::split (p, line, boost::algorithm::is_any_of (" "));
+
+ /* Further trim this line, removing spaces from the end */
+ trim_right_if (*line, boost::is_any_of (" "));
+
+ boost::algorithm::split (p, *line, boost::algorithm::is_any_of (" "), boost::token_compress_on);
if (p.size() != 3 && p.size() != 7) {
- throw SubripError (line, "a time/position line");
+ for (int i = 0; i < 2; ++i) {
+ optional<string> ex = get_line ();
+ if (ex) {
+ _context.push_back (*ex);
+ }
+ }
+ throw SubripError (*line, "a time/position line", _context);
}
- from = convert_time (p[0]);
- to = convert_time (p[2]);
+ rs.from = convert_time (p[0]);
+ rs.to = convert_time (p[2]);
/* XXX: should not ignore coordinate specifications */
break;
}
case CONTENT:
- if (line.empty ()) {
+ if (line->empty ()) {
state = COUNTER;
- line_number = 0;
} else {
- convert_line (line, line_number, from, to);
- line_number++;
+ convert_line (*line, rs);
+ rs.vertical_position.line = rs.vertical_position.line.get() + 1;
}
break;
}
vector<string> a;
boost::algorithm::split (a, t, boost::is_any_of (":"));
if (a.size() != 3) {
- throw SubripError (t, "time in the format h:m:s,ms");
+ throw SubripError (t, "time in the format h:m:s,ms", _context);
}
vector<string> b;
boost::algorithm::split (b, a[2], boost::is_any_of (","));
+ if (b.size() != 2) {
+ throw SubripError (t, "time in the format h:m:s,ms", _context);
+ }
+
+ int h, m, s, ms;
+
+ try {
+ h = lexical_cast<int>(a[0]);
+ } catch (boost::bad_lexical_cast &) {
+ throw SubripError (t, "integer hour value", _context);
+ }
+
+ try {
+ m = lexical_cast<int>(a[1]);
+ } catch (boost::bad_lexical_cast &) {
+ throw SubripError (t, "integer minute value", _context);
+ }
+
+ try {
+ s = lexical_cast<int>(b[0]);
+ } catch (boost::bad_lexical_cast &) {
+ throw SubripError (t, "integer second value", _context);
+ }
+
+ try {
+ ms = lexical_cast<int>(b[1]);
+ } catch (boost::bad_lexical_cast &) {
+ throw SubripError (t, "integer millisecond value", _context);
+ }
- return Time::from_hms (
- lexical_cast<int> (a[0]),
- lexical_cast<int> (a[1]),
- lexical_cast<int> (b[0]),
- lexical_cast<int> (b[1])
- );
+ return Time::from_hms (h, m, s, ms);
}
void
-SubripReader::convert_line (string t, int line_number, Time from, Time to)
+SubripReader::convert_line (string t, RawSubtitle& p)
{
enum {
TEXT,
string tag;
- RawSubtitle p;
- p.font = "Arial";
- p.font_size.set_points (48);
- p.from = from;
- p.to = to;
- p.vertical_position.line = line_number;
- /* XXX: arbitrary */
- p.vertical_position.lines = 32;
- p.vertical_position.reference = TOP_OF_SUBTITLE;
-
list<Colour> colours;
colours.push_back (Colour (1, 1, 1));
p.underline = false;
} else if (boost::starts_with (tag, "font")) {
maybe_content (p);
- boost::regex re (".*color=\"#([0123456789abcdef]+)\"");
+ boost::regex re (".*color=\"#([[:xdigit:]]+)\"");
boost::smatch match;
if (boost::regex_search (tag, match, re) && string (match[1]).size() == 6) {
p.colour = Colour::from_rgb_hex (match[1]);
colours.push_back (p.colour);
+ } else {
+ re = boost::regex (
+ ".*color=\"rgba\\("
+ "[[:space:]]*([[:digit:]]+)[[:space:]]*,"
+ "[[:space:]]*([[:digit:]]+)[[:space:]]*,"
+ "[[:space:]]*([[:digit:]]+)[[:space:]]*,"
+ "[[:space:]]*([[:digit:]]+)[[:space:]]*"
+ "\\)\""
+ );
+ if (boost::regex_search (tag, match, re) && match.size() == 5) {
+ p.colour.r = raw_convert<int>(string(match[1])) / 255.0;
+ p.colour.g = raw_convert<int>(string(match[2])) / 255.0;
+ p.colour.b = raw_convert<int>(string(match[3])) / 255.0;
+ colours.push_back (p.colour);
+ } else {
+ throw SubripError (tag, "a colour in the format #rrggbb or rgba(rr,gg,bb,aa)", _context);
+ }
}
} else if (tag == "/font") {
+ maybe_content (p);
+ SUB_ASSERT (!colours.empty());
colours.pop_back ();
p.colour = colours.back ();
+ } else if (tag.size() > 0 && tag[0] == '\\') {
+ SSAReader::parse_style (p, tag, 288, 288);
}
tag.clear ();
state = TEXT;
} else {
- tag += t[i];
+ tag += tolower (t[i]);
}
break;
}
}
+ /* Strip Unicode U+202B (right-to-left embedding) as sometimes it is rendered
+ as a missing character. This may be a hack.
+ */
+ replace_all (p.text, "\xe2\x80\xab", "");
+
maybe_content (p);
}
+/* Push p into _subs if it has some text, and clear the text out of p */
void
SubripReader::maybe_content (RawSubtitle& p)
{