+SubtitleDecoder::image_subtitle (ContentTime from, ContentTime to, shared_ptr<Image> image, dcpomatic::Rect<double> rect)
+{
+ _decoded_image_subtitles.push_back (shared_ptr<ContentImageSubtitle> (new ContentImageSubtitle (from, to, image, rect)));
+}
+
+void
+SubtitleDecoder::text_subtitle (list<dcp::SubtitleString> s)
+{
+ _decoded_text_subtitles.push_back (shared_ptr<ContentTextSubtitle> (new ContentTextSubtitle (s)));
+}
+
+template <class T>
+list<shared_ptr<T> >
+SubtitleDecoder::get (list<shared_ptr<T> > const & subs, ContentTime from, ContentTime to)
+{
+ if (subs.empty() || from < subs.front()->from() || to > (subs.back()->to() + ContentTime::from_seconds (10))) {
+ /* Either we have no decoded data, or what we do have is a long way from what we want: seek */
+ seek (from, true);
+ }
+
+ /* Now enough pass() calls will either:
+ * (a) give us what we want, or
+ * (b) hit the end of the decoder.
+ *
+ * XXX: with subs being sparse, this may need more care...
+ */
+ while (!pass() && (subs.empty() || (subs.front()->from() > from || to < subs.back()->to()))) {}
+
+ /* Now look for what we wanted in the data we have collected */
+ /* XXX: inefficient */
+
+ list<shared_ptr<T> > out;
+ for (typename list<shared_ptr<T> >::const_iterator i = subs.begin(); i != subs.end(); ++i) {
+ if ((*i)->from() <= to && (*i)->to() >= from) {
+ out.push_back (*i);
+ }
+ }
+
+ return out;
+}
+
+list<shared_ptr<ContentTextSubtitle> >
+SubtitleDecoder::get_text_subtitles (ContentTime from, ContentTime to)
+{
+ return get<ContentTextSubtitle> (_decoded_text_subtitles, from, to);
+}
+
+list<shared_ptr<ContentImageSubtitle> >
+SubtitleDecoder::get_image_subtitles (ContentTime from, ContentTime to)
+{
+ return get<ContentImageSubtitle> (_decoded_image_subtitles, from, to);
+}
+
+void
+SubtitleDecoder::seek (ContentTime, bool)