Server full-Text search features a number of good improvements to SQL 2000 Full-text search.
The first is to be able to update data changes without having to rebuild the Full-text indexing index. You can update the index manually, or you can update it in the job, or use the background update option
(Background update index option) to update the index as the data changes. This allows Full-text search to be used for frequently updated data
Real-time search, for example, for forum posts retrieval. The second improvement is that Full-text search indexes can be used for fields of image types. When you store a file of the specified document type in the Image field,
Full-Text Search can call the corresponding filter for the type of document to obtain information about the file.
This paragraph is not very clear, for example, I put a word file in an image field, and add this field to the Full-text search index, then I
Can get the author of this file, file size, modification time and other information. Example:
SELECT Q.filename, Q.size, Q.docauthor, w.citizenship
From OpenQuery (Mylinkedserver, ' SELECT FileName, Size, DocAuthor
From SCOPE ("D:\" '')
WHERE CONTAINS ("SQL Server")
NEAR () text "")
and FileName like '%.doc% ') as Q,
Writers as wwhere Q.docauthor = W.writer_name
Supported file types are Office documents, text files, HTML documents, and messages. Next announcement Federated database Server Appendix SQL 7 full-text search capabilities how?
First, the huge consumption of hard disk resources, about the use of the source content size of 40%-50% of the disk space.
Second, it's not always updated (unlike Index server, it has a cidaemon that monitors file changes at any time) but can also be set to update every night or event triggers.
Third, support Chinese, but not too good, participle technology is not perfect, do ordinary inquiries can also, not suitable for search engine. Four, the query speed can also feel fast enough.
