//Parse IMDb page data
private void parseIMDbPage(string html, bool GetExtraInfo)
{
Id = match(@"<link rel=""canonical"" href=""http://www.imdb.com/title/(tt\d{7})/"" />", html);
if (!string.IsNullOrEmpty(Id))
{
status = true;
Title = match(@"<title>(IMDb \- )*(.*?) \(.*?</title>", html, 2);
OriginalTitle = match(@"title-extra"">(.*?)<", html);
Year = match(@"<title>.*?\(.*?(\d{4}).*?\).*?</title>", html);
Rating = match(@"ratingValue"">(\d.\d)<", html);
Genres = new ArrayList();
Genres = matchAll(@"<a.*?>(.*?)</a>", match(@"Genres:</h4>(.*?)</div>", html));
Directors = new ArrayList();
Directors = matchAll(@"<a.*?>(.*?)</a>", match(@"Directors?:[\n\r\s]*</h4>(.*?)(</div>|>.?and )", html));
Writers = matchAll(@"<a.*?>(.*?)</a>", match(@"Writers?:[\n\r\s]*</h4>(.*?)(</div>|>.?and )", html));
Stars = matchAll(@"<a.*?>(.*?)</a>", match(@"Stars?:(.*?)</div>", html));
Cast = matchAll(@"class=""name"">[\n\r\s]*<a.*?>(.*?)</a>", html);
Plot = match(@"<p itemprop=""description"">(.*?)</p>", html);
ReleaseDate = match(@"Release Date:</h4>.*?(\d{1,2} (January|February|March|April|May|June|July|August|September|October|November|December) (19|20)\d{2}).*(\(|<span)", html);
Runtime = match(@"Runtime:</h4>[\s]*.*?(\d{1,4}) min[\s]*.*?\<\/div\>", html);
if (String.IsNullOrEmpty(Runtime))
{
Runtime = match(@"infobar.*?([0-9]+) min.*?</div>", html);
}
Top250 = match(@"Top 250 #(\d{1,3})<", html);
Oscars = match(@"Won (\d{1,2}) Oscars\.", html);
Awards = match(@"(\d{1,4}) wins", html);
Nominations = match(@"(\d{1,4}) nominations", html);
Storyline = match(@"Storyline</h2>[\s]*<p>(.*?)[\s]*(<em|</p>)", html);
Tagline = match(@"Taglines?:</h4>(.*?)(<span|</div)", html);
MpaaRating = match(@"infobar"">.*?<img.*?alt=""(.*?)"" src="".*?certificates.*?"".*?>", html);
Votes = match(@"href=""ratings"".*?>(\d+,?\d*)</span> votes</a>", html);
Languages = new ArrayList();
Languages = matchAll(@"<a.*?>(.*?)</a>", match(@"Language.?:(.*?)(</div>|>.?and )", html));
Countries = new ArrayList();
Countries = matchAll(@"<a.*?>(.*?)</a>", match(@"Country:(.*?)(</div>|>.?and )", html));
Poster = match(@"img_primary"">[\n\r\s]*?<a.*?><img src=""(.*?)"".*?</td>", html);
if (!string.IsNullOrEmpty(Poster) && Poster.IndexOf("nopicture") < 0)
{
PosterSmall = Poster.Substring(0, Poster.IndexOf("_V1.")) + "_V1._SY150.jpg";
PosterLarge = Poster.Substring(0, Poster.IndexOf("_V1.")) + "_V1._SY500.jpg";
PosterFull = Poster.Substring(0, Poster.IndexOf("_V1.")) + "_V1._SY0.jpg";
}
else
{
Poster = string.Empty;
PosterSmall = string.Empty;
PosterLarge = string.Empty;
PosterFull = string.Empty;
}
ImdbURL = "http://www.imdb.com/title/" + Id + "/";
if (GetExtraInfo)
{
ReleaseDates = getReleaseDates();
MediaImages = getMediaImages();
}
}
}