/// <summary>
/// Delete documents from index document by company's user
/// </summary>
/// <param name="lstBlobs"></param>
/// <param name="usrUser"></param>
/// <returns></returns>
public List <BlobStructure> DeleteDocumentsFromIndex(List <BlobStructure> lstBlobs, User usrUser)
{
var idxSearch = CreateIfNotExistIndex(usrUser, cnsIndexTypes.Documents);
// Initialize client search
var client = new SearchIndexClient(ConfigurationManager.AppSettings["SearchServiceName"],
idxSearch.IndexName,
new SearchCredentials(ConfigurationManager.AppSettings["SearchApiKey"]));
// Only uploaded files
foreach (var blbBlob in lstBlobs.Where(x => x.IsUploaded))
{
var dcmMetadata = blbBlob.dctMetadata.ToObject <clsDocumentPdfIndex>();
var strFileNameWithOutExt = blbBlob.fntNameStr.Split('.')[0];
// If files not are valid documents
if (!strExtensions.Equals(dcmMetadata.Extension))
{
continue;
}
try
{
var lstActions = new List <IndexAction <clsDocumentPdfIndex> >();
// download blob like stream var
var streamFile = new Blobs.BlobStorage().fntGetStreamBlob(blbBlob);
streamFile.Position = 0;
// Create a index action by each page
using (PdfReader reader = new PdfReader(streamFile))
{
for (int indexPage = 1; indexPage <= reader.NumberOfPages; indexPage++)
{
// add index page before extension file
dcmMetadata.Id = MakeSafeId(strFileNameWithOutExt + "_" + indexPage + "." + dcmMetadata.Extension);
lstActions.Add(IndexAction.Delete(dcmMetadata));
}
}
// Index batch process
for (int i = 0; i < (int)Math.Ceiling(lstActions.Count / (double)intRegistersSkip); i++)
{
client.Documents.Index(new IndexBatch <clsDocumentPdfIndex>(lstActions.Skip(i * intRegistersSkip).Take(intRegistersSkip)));
}
lstBlobs.First(l => l.fntNameStr == blbBlob.fntNameStr).IsIndexed = true;
}
catch (Exception)
{
lstBlobs.First(l => l.fntNameStr == blbBlob.fntNameStr).IsIndexed = false;
}
}
return(lstBlobs);
}
//private readonly string[] lstExtensions = { "pdf", "doc", "docx" };
/// <summary>
/// Add documents to index document by company's user
/// </summary>
/// <param name="lstBlobs">list of blobs</param>
/// <param name="usrUser">user data</param>
/// <returns></returns>
public List <BlobStructure> AddDocumentsToIndex(List <BlobStructure> lstBlobs, User usrUser)
{
var idxSearch = CreateIfNotExistIndex(usrUser, cnsIndexTypes.Documents);
// Initialize client search
var client = new SearchIndexClient(ConfigurationManager.AppSettings["SearchServiceName"],
idxSearch.IndexName,
new SearchCredentials(ConfigurationManager.AppSettings["SearchApiKey"]));
// Only uploaded files
foreach (var blbBlob in lstBlobs.Where(x => x.IsUploaded))
{
// remove path metadata (local path is not stored)
blbBlob.dctMetadata.Remove(cnsBlobMetadata.Path);
// transform tags to collection string
var strTags = blbBlob.dctMetadata[cnsBlobMetadata.Tags];
if (strTags != null)
{
blbBlob.dctMetadata[cnsBlobMetadata.Tags] = strTags.ToString().Split(',');
}
// get full blob name with out extension
var strFileNameWithOutExt = blbBlob.fntNameStr.Split('.')[0];
// If files not are valid documents
if (!strExtensions.Equals(blbBlob.dctMetadata[cnsBlobMetadata.Extension].ToString()))
{
continue;
}
try
{
var lstActions = new List <IndexAction <clsDocumentPdfIndex> >();
// download blob like stream var
var streamFile = new Blobs.BlobStorage().fntGetStreamBlob(blbBlob);
streamFile.Position = 0;
// Create a index action by each page
using (PdfReader reader = new PdfReader(streamFile))
{
for (int indexPage = 1; indexPage <= reader.NumberOfPages; indexPage++)
{
// transform dictionary to object
var dcmMetadata = blbBlob.dctMetadata.ToObject <clsDocumentPdfIndex>();
dcmMetadata.Container = blbBlob.cntContainer.fntFullNameStr;
dcmMetadata.Uri = blbBlob.fntUri.AbsoluteUri;
dcmMetadata.FullName = blbBlob.fntNameStr;
// add index page before extension file
dcmMetadata.Id = MakeSafeId(strFileNameWithOutExt + "_" + indexPage + "." + dcmMetadata.Extension);
dcmMetadata.Content = PdfTextExtractor.GetTextFromPage(reader, indexPage);
dcmMetadata.Page = indexPage;
lstActions.Add(IndexAction.MergeOrUpload(dcmMetadata));
}
}
// Index batch process
for (int i = 0; i < (int)Math.Ceiling(lstActions.Count / (double)intRegistersSkip); i++)
{
client.Documents.Index(new IndexBatch <clsDocumentPdfIndex>(lstActions.Skip(i * intRegistersSkip).Take(intRegistersSkip)));
}
lstBlobs.First(l => l.fntNameStr == blbBlob.fntNameStr).IsIndexed = true;
}
catch (Exception ex)
{
lstBlobs.First(l => l.fntNameStr == blbBlob.fntNameStr).IsIndexed = false;
}
}
return(lstBlobs);
}
