Msg#: 3769486 posted 9:06 am on Oct 20, 2008 (gmt 0)
Suppose we need a dynamic full text search engine on a web page. The engine would search a database consisting of a large amount of short documents. The user must be able to simply search the data using one or more keywords like Google. The web page must not be fully reloaded when the user performs a search -- only the search result area must be updated.
Which web development techniques would you use to implement such a search and why? What kinds of things must be carefully considered when implementing the search and the underlying database in order to make the functionality as intuitive, easy and fast for the end user as possible?
According to my understanding the part where only search result area will be updated must be done through AJAX right!? And for full text search we can use MyISAM FULL-TEXT-SEARCH index but this will only work in the case where documents content is saved into the database. What if we are only saving document name in database and have to search content from physical file?