| public static Document ( System f ) : Lucene.Net.Documents.Document | ||
| f | System | |
| return | Lucene.Net.Documents.Document | |
private static void IndexFile(IndexWriter writer, TermEnum uidIter, FileInfo file, Operation operation)
{
if (file.FullName.EndsWith(".html") || file.FullName.EndsWith(".htm") || file.FullName.EndsWith(".txt"))
{
// We've found a file we should index.
if (operation == Operation.IncrementalReindex ||
operation == Operation.RemoveStale)
{
// We should only get here with an open uidIter.
Debug.Assert(uidIter != null, "Expected uidIter != null for operation " + operation);
var uid = HTMLDocument.Uid(file); // construct uid for doc
while (uidIter.Term != null && uidIter.Term.Field == "uid" && String.CompareOrdinal(uidIter.Term.Text, uid) < 0)
{
if (operation == Operation.RemoveStale)
{
Console.Out.WriteLine("deleting " + HTMLDocument.Uid2url(uidIter.Term.Text));
writer.DeleteDocuments(uidIter.Term);
}
uidIter.Next();
}
// The uidIter TermEnum should now be pointing at either
// 1) a null term, meaning there are no more uids to check.
// 2) a term matching the current file.
// 3) a term not matching us.
if (uidIter.Term != null && uidIter.Term.Field == "uid" && String.CompareOrdinal(uidIter.Term.Text, uid) == 0)
{
// uidIter points to the current document, we should move one
// step ahead to keep state consistant, and carry on.
uidIter.Next();
}
else if (operation == Operation.IncrementalReindex)
{
// uidIter does not point to the current document, and we're
// currently indexing documents.
var doc = HTMLDocument.Document(file);
Console.Out.WriteLine("adding " + doc.Get("path"));
writer.AddDocument(doc);
}
}
else
{
// We're doing a complete reindexing. We aren't using uidIter,
// but for completeness we assert that it's null (as expected).
Debug.Assert(uidIter == null, "Expected uidIter == null for operation == " + operation);
var doc = HTMLDocument.Document(file);
Console.Out.WriteLine("adding " + doc.Get("path"));
writer.AddDocument(doc);
}
}
}
private static void IndexDocs(System.IO.FileInfo file)
{
if (System.IO.Directory.Exists(file.FullName))
{
// if a directory
System.String[] files = System.IO.Directory.GetFileSystemEntries(file.FullName); // list its files
System.Array.Sort(files); // sort the files
for (int i = 0; i < files.Length; i++)
{
// recursively index them
IndexDocs(new System.IO.FileInfo(System.IO.Path.Combine(file.FullName, files[i])));
}
}
else if (file.FullName.EndsWith(".html") || file.FullName.EndsWith(".htm") || file.FullName.EndsWith(".txt"))
{
// index .txt files
if (uidIter != null)
{
System.String uid = HTMLDocument.Uid(file); // construct uid for doc
while (uidIter.Term() != null && (System.Object)uidIter.Term().Field() == (System.Object) "uid" && String.CompareOrdinal(uidIter.Term().Text(), uid) < 0)
{
if (deleting)
{
// delete stale docs
System.Console.Out.WriteLine("deleting " + HTMLDocument.Uid2url(uidIter.Term().Text()));
reader.DeleteDocuments(uidIter.Term());
}
uidIter.Next();
}
if (uidIter.Term() != null && (System.Object)uidIter.Term().Field() == (System.Object) "uid" && String.CompareOrdinal(uidIter.Term().Text(), uid) == 0)
{
uidIter.Next(); // keep matching docs
}
else if (!deleting)
{
// add new docs
Document doc = HTMLDocument.Document(file);
System.Console.Out.WriteLine("adding " + doc.Get("path"));
writer.AddDocument(doc);
}
}
else
{
// creating a new index
Document doc = HTMLDocument.Document(file);
System.Console.Out.WriteLine("adding " + doc.Get("path"));
writer.AddDocument(doc); // add docs unconditionally
}
}
}