How to query berkeley db with multiple keys

Question

Hi all,
Is there a way to query a berkeleydb using multiple keys. For example, I want to query a order record with orderID and orderDirection, orderID is a string and orderDirection is a char.
Is there a way to do so?
Regards,
-Bruce

Answer 1

Hi Bruce,
Yes, you would use the same approach with a join cursor.
Here is a simple example, simplified for brevity. Suppose the primary database "books" stores information about books, and that these books could have multiple authors. Following is some example data and the structure of these databases:
books.db (primary db -- ISBN is primary key)
isbn_1 | {author_1, author_2; ...}
isbn_2 | {author_2; ...}
isbn_3 | {author_1, author_3; ...}
booksAuthors.db (secondary db -- author is the secondary key, the secondary key extractor generates multiple keys)
author_1 | isbn_1
author_1 | isbn_3
author_2 | isbn_1
author_2 | isbn_2
author_3 | isbn_3
To get the books authored by author_1 and author_2 for example, you would do the following:
- position a cursor in the booksAuthors.db secondary db, using DBcursor->get() with the DB_SET flag, on the records whose key (secondary key) is author_1;
The cursor will be positioned on the duplicates list for author_1:
author_1 | isbn_1
author_1 | isbn_3
- position a cursor in the booksAuthors.db secondary db, using DBcursor->get() with the DB_SET flag, on the records whose key (secondary key) is author_2, hence duplicates list for author_2:
author_2 | isbn_1
author_2 | isbn_2
- create a join cursor using DB->join, using the previous two cursors.
An excerpt from the above documentation page explains the way a join cursor works:
"Joined values are retrieved by doing a sequential iteration over the first cursor in the curslist parameter, and a nested iteration over each secondary cursor in the order they are specified in the curslist parameter. This requires database traversals to search for the current datum in all the cursors after the first."
Hence, the data item (primary key from the primary db) that results is isbn_1.
Regards,
Andrei