Ulf: the index typically does not contain the full text of the indexed documents.
True, but the index will typically give me the record pointer, isnt it?
So, if i have indexed a text file to give me record pointers & record length for records containing a particular value for the indexed field, i can quickly retrieve the record from the file. isn't?
Ulf:I don't understand what you mean by "jump from one index to another" - random access of the file contents?
I assumed this! Its worth while getting this confirmed.