/// <summary>
/// Determines if the specified page name is in the supplied list
/// </summary>
/// <param name="pageName">Page Name for which to search</param>
/// <param name="items">List in which to search</param>
/// <returns>True if found, False if not</returns>
private bool PageNameIsInList( string pageName, FindItItem[] items )
{
bool searchResult = false;
foreach ( FindItItem item in items )
{
if ( String.Compare(item.Name, pageName, true) == 0 )
searchResult = true;
}
return searchResult;
}
/// <summary>
/// For each page name in the list, it inserts a new row in the database or
/// or updates an existing row. Then, any previously-existing rows that
/// were not updated are set to inactive.
/// </summary>
/// <param name="pageID">Identifier of the page whose page names are being updated</param>
/// <param name="items">List of page names</param>
public int[] PageNameUpdateList( int pageID, FindItItem[] items )
{
int[] updateStats = new int[ 4 ];
foreach ( FindItItem item in items )
{
PageName pageName = this.PageNameSelectByPageIDAndNameFound( pageID, item.Name );
if ( pageName == null )
{
// Insert new page name
pageName = new PageName();
pageName.NameFound = item.Name;
pageName.NameConfirmed = item.Name;
pageName.NameBankID = item.NamebankID;
if ( pageName.NameBankID <= 0 )
pageName.NameBankID = null;
pageName.Active = !( pageName.NameBankID == null );
pageName.PageID = pageID;
pageName.Source = "uBio";
pageName.IsCommonName = false;
this.PageNameInsertAuto( pageName );
updateStats[ 0 ]++; // number inserted
}
else
{
// Update existing page name if the namebankid has changed
if ( ( pageName.NameBankID ?? 0 ) != item.NamebankID )
{
pageName.NameBankID = item.NamebankID;
if ( pageName.NameBankID <= 0 )
pageName.NameBankID = null;
pageName.Active = !( pageName.NameBankID == null );
this.PageNameUpdateAuto( pageName );
updateStats[ 1 ]++; // number updated
}
else
{
updateStats[ 3 ]++; // number unchanged
}
}
}
// Deactivate any names that are in the database, but were not not returned
// by the just-completed UBIO request (this means they've fallen out of the
// list of page names for this page)
CustomGenericList<PageName> pageNames = this.PageNameSelectByPageID( pageID );
foreach ( PageName pageName in pageNames )
{
if ( !this.PageNameIsInList( pageName.NameFound, items ) )
{
if ( pageName.NameBankID != null )
{
pageName.Active = false;
pageName.NameBankID = null;
this.PageNameUpdateAuto( pageName );
updateStats[ 2 ]++; // number deleted
}
else
{
updateStats[ 3 ]++; // number unchanged
}
}
}
return updateStats;
}
public FindItItem[] GetUBioNames( int pageID )
{
bool tooLarge = false;
string webServiceUrl = string.Empty;
PageSummaryView ps = new BHLProvider().PageSummarySelectByPageId( pageID );
string filepath = ps.OcrTextLocation;
if (new BHLProvider().GetFileAccessProvider(ConfigurationManager.AppSettings["UseRemoteFileAccessProvider"] == "true").GetFileSizeInKB(filepath) <= 5)
{
// OCR text not too large for URI, so send it in the UBIO request
// Get the OCR text and start to build the UBIO url
string ocrText = this.GetOcrText( ps );
StringBuilder webServiceUrlSB = new StringBuilder();
webServiceUrlSB.Append( "http://www.ubio.org/webservices/service.php?function=taxonFinder&includeLinks=0&freeText=" );
webServiceUrlSB.Append( System.Web.HttpUtility.UrlEncode( ocrText ) );
// Add the existing page names for this Page to the UBIO url
CustomGenericList<PageName> pageNames = new BHLProvider().PageNameSelectByPageID( pageID );
foreach ( PageName pageName in pageNames )
{
webServiceUrlSB.Append( System.Web.HttpUtility.UrlEncode( " " + pageName.NameFound ) );
}
// Get the final UBIO url
webServiceUrl = webServiceUrlSB.ToString();
// If the url is too large after all UrlEncoding is complete, just send the file path
if ( ( (long)webServiceUrl.Length / 1024 ) > 5 )
tooLarge = true;
}
else
{
tooLarge = true;
}
// OCR text is too large, so just send the file path
//string webServiceUrl = String.Format("http://names.ubio.org/webservices/service.php?function=findIT&url=http://www.botanicus.org/PrimeOcr/{0}/{1}/{2}/{3}.txt&strict=1&keyCode=78bd024866f1df74125f194b764d80333e0d64aa", ps.WebVirtualDirectory, ps.MARCBibID, ps.BarCode, ps.FileNamePrefix);
if ( tooLarge )
webServiceUrl = String.Format( "http://www.ubio.org/webservices/service.php?function=taxonFinder&includeLinks=0&url=http://www.botanicus.org/PrimeOcr/mbgserv14/{0}/{1}/{2}/{3}.txt", ps.WebVirtualDirectory, ps.FileRootFolder, ps.BarCode,
ps.FileNamePrefix );
List<FindItItem> fiil = new List<FindItItem>();
XmlTextReader reader = null;
try
{
HttpWebRequest req = (HttpWebRequest)WebRequest.Create( webServiceUrl );
req.Method = "POST";
req.Timeout = 15000;
HttpWebResponse resp = (HttpWebResponse)req.GetResponse();
reader = new XmlTextReader( (System.IO.Stream)resp.GetResponseStream() );
StringBuilder sb = new StringBuilder();
FindItItem fii = null;
string currentStage = "";
while ( reader.Read() )
{
if ( reader.NodeType == XmlNodeType.Whitespace )
continue;
if ( reader.HasValue )
sb.Append( reader.Value );
if ( currentStage == "nameString" && reader.Value != "" )
{
fii.Name = reader.Value;
}
if ( currentStage == "namebankID" && reader.Value != "" )
{
fii.NamebankID = Int32.Parse( reader.Value );
}
else if ( reader.NodeType != XmlNodeType.EndElement )
{
sb.Append( "\n" + reader.Name + ": " );
currentStage = reader.Name;
if ( reader.Name == "entity" )
{
fii = new FindItItem();
}
}
else
{
if ( reader.Name == "entity" )
{
fiil.Add( fii );
}
}
}
}
finally
{
if ( reader != null )
reader.Close();
}
return ( fiil.ToArray() );
}
public int[] PageNameUpdateList(int pageID, FindItItem[] items)
{
BHLProvider provider = new BHLProvider();
return provider.PageNameUpdateList(pageID, items);
}
private FindItItem ubioLookup( string name )
{
XmlTextReader reader = null;
try
{
HttpWebRequest request = (HttpWebRequest)WebRequest.Create(
"http://www.ubio.org/webservices/service.php?function=taxonFinder&includeLinks=0&freeText=" +
Server.UrlEncode( name ) );
request.Timeout = 15000;
HttpWebResponse response = (HttpWebResponse)request.GetResponse();
reader = new XmlTextReader( (Stream)response.GetResponseStream() );
XmlDocument doc = new XmlDocument();
doc.Load( reader );
FindItItem result = null;
// Taking this simple, straight forward approach because we're only expecting one result per check. Will
// make parsing algorithm more robust later if needed.
XmlNodeList nameList = doc.GetElementsByTagName( "nameString" );
XmlNodeList idList = doc.GetElementsByTagName( "namebankID" );
if ( nameList.Count == 1 )
{
result = new FindItItem();
result.Name = nameList[ 0 ].InnerText;
if ( idList.Count == 1 )
{
result.NamebankID = int.Parse( idList[ 0 ].InnerText );
}
}
return result;
}
catch
{
return null;
}
finally
{
if ( reader != null )
{
reader.Close();
}
}
}
