void CMusicInfoScraper::FindAlbuminfo()
{
CStdString strAlbum=m_strAlbum;
CStdString strHTML;
m_vecAlbums.erase(m_vecAlbums.begin(), m_vecAlbums.end());
CScraperParser parser;
if (!parser.Load(_P("q:\\system\\scrapers\\music\\"+m_info.strPath)))
return;
if (!m_info.settings.GetPluginRoot() || m_info.settings.GetSettings().IsEmpty())
{
m_info.settings.LoadSettingsXML(_P("q:\\system\\scrapers\\music\\"+m_info.strPath));
m_info.settings.SaveFromDefault();
}
parser.m_param[0] = strAlbum;
parser.m_param[1] = m_strArtist;
CUtil::URLEncode(parser.m_param[0]);
CUtil::URLEncode(parser.m_param[1]);
CScraperUrl scrURL;
scrURL.ParseString(parser.Parse("CreateAlbumSearchUrl"));
if (!CScraperUrl::Get(scrURL.m_url[0], strHTML, m_http) || strHTML.size() == 0)
{
CLog::Log(LOGERROR, "%s: Unable to retrieve web site",__FUNCTION__);
return;
}
parser.m_param[0] = strHTML;
CStdString strXML = parser.Parse("GetAlbumSearchResults",&m_info.settings);
if (strXML.IsEmpty())
{
CLog::Log(LOGERROR, "%s: Unable to parse web site",__FUNCTION__);
return;
}
if (strXML.Find("encoding=\"utf-8\"") < 0)
g_charsetConverter.unknownToUTF8(strXML);
// ok, now parse the xml file
TiXmlDocument doc;
doc.Parse(strXML.c_str(),0,TIXML_ENCODING_UTF8);
if (!doc.RootElement())
{
CLog::Log(LOGERROR, "%s: Unable to parse xml",__FUNCTION__);
return;
}
TiXmlHandle docHandle( &doc );
TiXmlElement* album = docHandle.FirstChild( "results" ).FirstChild( "entity" ).Element();
if (!album)
return;
while (album)
{
TiXmlNode* title = album->FirstChild("title");
TiXmlElement* link = album->FirstChildElement("url");
TiXmlNode* artist = album->FirstChild("artist");
TiXmlNode* year = album->FirstChild("year");
if (title && title->FirstChild())
{
CStdString strTitle = title->FirstChild()->Value();
CStdString strArtist;
CStdString strAlbumName;
if (artist && artist->FirstChild())
{
strArtist = artist->FirstChild()->Value();
strAlbumName.Format("%s - %s",strArtist.c_str(),strTitle.c_str());
}
else
strAlbumName = strTitle;
if (year && year->FirstChild())
strAlbumName.Format("%s (%s)",strAlbumName.c_str(),year->FirstChild()->Value());
CScraperUrl url;
if (!link)
url.ParseString(scrURL.m_xml);
while (link && link->FirstChild())
{
url.ParseElement(link);
link = link->NextSiblingElement("url");
}
CMusicAlbumInfo newAlbum(strTitle, strArtist, strAlbumName, url);
m_vecAlbums.push_back(newAlbum);
}
album = album->NextSiblingElement();
}
if (m_vecAlbums.size()>0)
m_bSuccessfull=true;
return;
}
void CMusicInfoScraper::FindArtistInfo()
{
CStdString strArtist=m_strArtist;
CStdString strHTML;
m_vecArtists.erase(m_vecArtists.begin(), m_vecArtists.end());
if (!m_scraper->Load())
return;
vector<CStdString> extras;
extras.push_back(m_strArtist);
g_charsetConverter.utf8To(m_scraper->GetParser().GetSearchStringEncoding(), m_strArtist, extras[0]);
CURL::Encode(extras[0]);
CLog::Log(LOGDEBUG, "%s: Searching for '%s' using %s scraper (file: '%s', content: '%s', version: '%s')",
__FUNCTION__, m_strArtist.c_str(), m_scraper->Name().c_str(), m_scraper->Path().c_str(),
ADDON::TranslateContent(m_scraper->Content()).c_str(), m_scraper->Version().c_str());
CScraperUrl scrURL;
vector<CStdString> url = m_scraper->Run("CreateArtistSearchUrl",scrURL,m_http,&extras);
if (url.empty())
return;
scrURL.ParseString(url[0]);
vector<CStdString> xml = m_scraper->Run("GetArtistSearchResults",scrURL,m_http,&extras);
for (vector<CStdString>::iterator it = xml.begin();
it != xml.end(); ++it)
{
// ok, now parse the xml file
TiXmlDocument doc;
doc.Parse(it->c_str(),0,TIXML_ENCODING_UTF8);
if (!doc.RootElement())
{
CLog::Log(LOGERROR, "%s: Unable to parse xml",__FUNCTION__);
continue;
}
TiXmlHandle docHandle( &doc );
TiXmlElement* artist = docHandle.FirstChild( "results" ).FirstChild( "entity" ).Element();
while (artist)
{
TiXmlNode* title = artist->FirstChild("title");
TiXmlNode* year = artist->FirstChild("year");
TiXmlNode* genre = artist->FirstChild("genre");
TiXmlElement* link = artist->FirstChildElement("url");
if (title && title->FirstChild())
{
CStdString strTitle = title->FirstChild()->Value();
CScraperUrl url;
if (!link)
url.ParseString(scrURL.m_xml);
while (link && link->FirstChild())
{
url.ParseElement(link);
link = link->NextSiblingElement("url");
}
CMusicArtistInfo newArtist(strTitle, url);
if (genre && genre->FirstChild())
newArtist.GetArtist().strGenre = genre->FirstChild()->Value();
if (year && year->FirstChild())
newArtist.GetArtist().strBorn = year->FirstChild()->Value();
m_vecArtists.push_back(newArtist);
}
artist = artist->NextSiblingElement();
}
}
if (m_vecArtists.size()>0)
m_bSucceeded=true;
return;
}
void CMusicInfoScraper::FindAlbumInfo()
{
CStdString strAlbum=m_strAlbum;
CStdString strHTML;
m_vecAlbums.erase(m_vecAlbums.begin(), m_vecAlbums.end());
if (!m_scraper->Load() || !m_scraper->GetParser().HasFunction("CreateAlbumSearchUrl"))
return;
CLog::Log(LOGDEBUG, "%s: Searching for '%s - %s' using %s scraper (path: '%s', content: '%s', version: '%s')",
__FUNCTION__, m_strArtist.c_str(), strAlbum.c_str(), m_scraper->Name().c_str(), m_scraper->Path().c_str(),
ADDON::TranslateContent(m_scraper->Content()).c_str(), m_scraper->Version().c_str());
vector<CStdString> extras;
extras.push_back(strAlbum);
extras.push_back(m_strArtist);
g_charsetConverter.utf8To(m_scraper->GetParser().GetSearchStringEncoding(), strAlbum, extras[0]);
g_charsetConverter.utf8To(m_scraper->GetParser().GetSearchStringEncoding(), m_strArtist, extras[1]);
CURL::Encode(extras[0]);
CURL::Encode(extras[1]);
CScraperUrl scrURL;
vector<CStdString> url = m_scraper->Run("CreateAlbumSearchUrl",scrURL,m_http,&extras);
if (url.empty())
return;
scrURL.ParseString(url[0]);
vector<CStdString> xml = m_scraper->Run("GetAlbumSearchResults",scrURL,m_http);
for (vector<CStdString>::iterator it = xml.begin();
it != xml.end(); ++it)
{
// ok, now parse the xml file
TiXmlDocument doc;
doc.Parse(it->c_str(),0,TIXML_ENCODING_UTF8);
TiXmlHandle docHandle( &doc );
TiXmlElement* album = docHandle.FirstChild( "results" ).FirstChild( "entity" ).Element();
while (album)
{
TiXmlNode* title = album->FirstChild("title");
TiXmlElement* link = album->FirstChildElement("url");
TiXmlNode* artist = album->FirstChild("artist");
TiXmlNode* year = album->FirstChild("year");
TiXmlElement* relevance = album->FirstChildElement("relevance");
if (title && title->FirstChild())
{
CStdString strTitle = title->FirstChild()->Value();
CStdString strArtist;
CStdString strAlbumName;
if (artist && artist->FirstChild())
{
strArtist = artist->FirstChild()->Value();
strAlbumName.Format("%s - %s",strArtist.c_str(),strTitle.c_str());
}
else
strAlbumName = strTitle;
if (year && year->FirstChild())
strAlbumName.Format("%s (%s)",strAlbumName.c_str(),year->FirstChild()->Value());
CScraperUrl url;
if (!link)
url.ParseString(scrURL.m_xml);
while (link && link->FirstChild())
{
url.ParseElement(link);
link = link->NextSiblingElement("url");
}
CMusicAlbumInfo newAlbum(strTitle, strArtist, strAlbumName, url);
if (relevance && relevance->FirstChild())
{
float scale=1;
const char* newscale = relevance->Attribute("scale");
if (newscale)
scale = (float)atof(newscale);
newAlbum.SetRelevance((float)atof(relevance->FirstChild()->Value())/scale);
}
m_vecAlbums.push_back(newAlbum);
}
album = album->NextSiblingElement();
}
}
if (m_vecAlbums.size()>0)
m_bSucceeded=true;
return;
}
int CIMDB::InternalFindMovie(const CStdString &strMovie, IMDB_MOVIELIST& movielist, bool& sortMovieList, const CStdString& strFunction, CScraperUrl* pUrl)
{
movielist.clear();
CScraperUrl scrURL;
CStdString strName = strMovie;
CStdString movieTitle, movieTitleAndYear, movieYear;
CUtil::CleanString(strName, movieTitle, movieTitleAndYear, movieYear, true);
movieTitle.ToLower();
CLog::Log(LOGDEBUG, "%s: Searching for '%s' using %s scraper (file: '%s', content: '%s', language: '%s', date: '%s', framework: '%s')",
__FUNCTION__, movieTitle.c_str(), m_info.strTitle.c_str(), m_info.strPath.c_str(), m_info.strContent.c_str(), m_info.strLanguage.c_str(), m_info.strDate.c_str(), m_info.strFramework.c_str());
if (!pUrl)
{
if (m_parser.HasFunction("CreateSearchUrl"))
{
GetURL(strMovie, movieTitle, movieYear, scrURL);
}
else if (m_info.strContent.Equals("musicvideos"))
{
if (!m_parser.HasFunction("FileNameScrape"))
return false;
CScraperUrl scrURL("filenamescrape");
scrURL.strTitle = strMovie;
movielist.push_back(scrURL);
return 1;
}
if (scrURL.m_xml.IsEmpty())
return 0;
}
else
scrURL = *pUrl;
vector<CStdString> strHTML;
for (unsigned int i=0;i<scrURL.m_url.size();++i)
{
CStdString strCurrHTML;
if (!CScraperUrl::Get(scrURL.m_url[i],strCurrHTML,m_http) || strCurrHTML.size() == 0)
return 0;
strHTML.push_back(strCurrHTML);
}
// now grab our details using the scraper
for (unsigned int i=0;i<strHTML.size();++i)
m_parser.m_param[i] = strHTML[i];
m_parser.m_param[strHTML.size()] = scrURL.m_url[0].m_url;
CStdString strXML = m_parser.Parse(strFunction,&m_info.settings);
//CLog::Log(LOGDEBUG,"scraper: %s returned %s",strFunction.c_str(),strXML.c_str());
if (strXML.IsEmpty())
{
CLog::Log(LOGERROR, "%s: Unable to parse web site",__FUNCTION__);
return 0;
}
if (!XMLUtils::HasUTF8Declaration(strXML))
g_charsetConverter.unknownToUTF8(strXML);
// ok, now parse the xml file
TiXmlDocument doc;
doc.Parse(strXML.c_str(),0,TIXML_ENCODING_UTF8);
if (!doc.RootElement())
{
CLog::Log(LOGERROR, "%s: Unable to parse xml",__FUNCTION__);
return 0;
}
if (stricmp(doc.RootElement()->Value(),"error")==0)
{
TiXmlElement* title = doc.RootElement()->FirstChildElement("title");
CStdString strTitle;
if (title && title->FirstChild() && title->FirstChild()->Value())
strTitle = title->FirstChild()->Value();
TiXmlElement* message = doc.RootElement()->FirstChildElement("message");
CStdString strMessage;
if (message && message->FirstChild() && message->FirstChild()->Value())
strMessage = message->FirstChild()->Value();
CGUIDialogOK* dialog = (CGUIDialogOK*)g_windowManager.GetWindow(WINDOW_DIALOG_OK);
dialog->SetHeading(strTitle);
dialog->SetLine(0,strMessage);
g_application.getApplicationMessenger().DoModal(dialog,WINDOW_DIALOG_OK);
return -1;
}
TiXmlHandle docHandle( &doc );
TiXmlElement* xurl = doc.RootElement()->FirstChildElement("url");
while (xurl && xurl->FirstChild())
{
const char* szFunction = xurl->Attribute("function");
if (szFunction)
{
CScraperUrl scrURL(xurl);
InternalFindMovie(strMovie,movielist,sortMovieList,szFunction,&scrURL);
}
xurl = xurl->NextSiblingElement("url");
}
TiXmlElement *movie = docHandle.FirstChild( "results" ).FirstChild( "entity" ).Element();
if (!movie)
return 0;
while (movie)
{
// is our result already sorted correctly when handed over from scraper? if so, do not let xbmc sort it
if (sortMovieList)
{
TiXmlElement* results = docHandle.FirstChild("results").Element();
if (results)
{
CStdString szSorted = results->Attribute("sorted");
sortMovieList = (szSorted.CompareNoCase("yes") != 0);
}
}
CScraperUrl url;
TiXmlNode *title = movie->FirstChild("title");
TiXmlElement *link = movie->FirstChildElement("url");
TiXmlNode *year = movie->FirstChild("year");
TiXmlNode* id = movie->FirstChild("id");
TiXmlNode* language = movie->FirstChild("language");
if (title && title->FirstChild() && link && link->FirstChild())
{
url.strTitle = title->FirstChild()->Value();
while (link && link->FirstChild())
{
url.ParseElement(link);
link = link->NextSiblingElement("url");
}
if (id && id->FirstChild())
url.strId = id->FirstChild()->Value();
// calculate the relavance of this hit
CStdString compareTitle = url.strTitle;
compareTitle.ToLower();
CStdString matchTitle = movieTitle;
matchTitle.ToLower();
// see if we need to add year information
CStdString compareYear;
if(year && year->FirstChild())
compareYear = year->FirstChild()->Value();
if (!movieYear.IsEmpty() && !compareYear.IsEmpty())
{
matchTitle.AppendFormat(" (%s)", movieYear.c_str());
compareTitle.AppendFormat(" (%s)", compareYear.c_str());
}
url.relevance = fstrcmp(matchTitle.c_str(), compareTitle.c_str(), 0);
// reconstruct a title for the user
CStdString title = url.strTitle;
if (!compareYear.IsEmpty())
title.AppendFormat(" (%s)", compareYear.c_str());
if (language && language->FirstChild())
title.AppendFormat(" (%s)", language->FirstChild()->Value());
url.strTitle = title;
movielist.push_back(url);
}
movie = movie->NextSiblingElement();
}
return 1;
}