ODBC, OLE DB, ADO, and ASP Questions and Answers

Answer: A

Correct answer: system data source.

A system data source is held in the registry of one machine and is available to every user of that machine and to services running on it, so it is local to a single computer. A user data source is narrower still, available only to the user who created it, and a file data source is the one that can be shared between computers.

Answer: B

Correct answer: False

The first half is right and the second is wrong. ADO does overlie the more complex OLE DB model, and that is its purpose - but it is callable from programming languages as well as scripting ones, from Visual Basic and C++ as readily as from VBScript. Being usable from almost any language able to work with COM is exactly what ADO was created for.

Answer: A

Correct answer: True

An abstraction is a generalisation that captures what a group of things have in common while suppressing their differences. The idea runs right through OLE DB: a rowset is an abstraction of rows however they are actually stored, which is what allows one interface to serve a relational database, a spreadsheet and a mail store alike.

Answer: D

Correct answer: Both 1 and 3 above are contained in the ODBC core SQL.

The core grammar takes in everything in the minimum grammar and adds the full SELECT with subqueries, CREATE VIEW and DROP VIEW, GRANT and REVOKE, and the standard aggregate functions. Stored procedures are in the extended grammar, one level above.

Answer: D

Correct answer: ADO

ADO is a thin cover over OLE DB, exposing a small set of objects that can be called from Visual Basic, from C++, from scripting languages such as VBScript and JScript, and from almost anything else able to use COM. Its purpose is exactly that reach: OLE DB’s own interfaces are usable in practice only from C++.

Answer: A

Correct answer: True

This is the definition of an interface in the object model: the set of objects it exposes together with the properties and methods those objects make available. A provider implements the interfaces it can support and no others, and a consumer discovers what is available and uses that - which is what gives OLE DB its flexibility.

Answer: B

Correct answer: file data source.

A file data source stores the connection details in a file rather than in the registry, so the file can be placed on a shared drive and used by anyone who can read it. That shareability is its whole purpose. A system data source is local to one computer but available to every user of it, and a user data source is available only to the user who created it.

Answer: D

Correct answer: All of the above are contained in the OBDC Level 2 API

The conformance levels are cumulative: a Level 2 driver supports everything in the core API and everything in Level 1 as well as its own additions. So committing and rolling back transactions, connecting to data sources, and connecting with driver-specific information are all available at Level 2, having been introduced at the levels below.

Answer: B

Correct answer: Driver

The driver is the DBMS-specific component: it receives an ODBC request and translates it into statements the particular data source understands, then translates the results back. The driver manager sits above it and chooses which driver to load, and the data source names the database and the DBMS that processes it.

Answer: B

Correct answer: False

The nesting is the wrong way round, and one term is redundant. A rowset is the OLE DB abstraction, generalising the recordset or relation to any kind of data source. A relation is not an abstraction of a recordset; the two are near synonyms at the same level, and the rowset is the generalisation of both.