Chapter 8. Data administration

PID_00148447 PID_00148458 PID_00148469 X07_M2103_02287 Eureca Media, S.L. FUOC

Av. Tibidabo, 39-43, 08035 Barcelona

2009-09-01 1

GNUFDL

Remo Suppi Boldrito

preface. preface

An important aspect of an operating system is where and how the data is saved. When the availability of data needs to be efficient, it is necessary to use databases (DB).

A database is a structured set of data that can be organised in a simple and efficient manner by the database handler. Current databases are known as relational, since the data can be stored in different tables for ease of management and administration. For this purpose and with a view to standardising database access, a language known as structured query language (SQL) is used. This language allows a flexible and rapid interaction irrespective of the database applications.

At present, the most commonly used way consists of accessing a database from an application that runs SQL code. For example, it is very common to access a DB through a web page that contains PHP or Perl code (the most common ones). When a page is requested by a client, the PHP or Perl code embedded in the page is executed, the DB is accessed and the page is generated with its static content and the content extracted from the DB that is then sent to the client. Two of the most relevant current examples of databases are those provided by ProstgreSQL and MySQL, which are the ones we will analyse.

However, when we work on software development, there are other data-related aspects to consider, regarding their validity and environment (especially if there is a group of users that work on the same data). There are several packages for version control (revisions), but the purpose of all of them is to facilitate the administration of the different versions of each developed product together with the potential specialisations made for any particular client.

Versions control is provided to control the different versions of the source code. However, the same concepts apply to other spheres and not only for source code but also for documents, images etc. Although a version control system can be implemented manually, it is highly advisable to have tools that facilitate this management (cvs, Subversion, SourceSafe, Clear Case, Darcs, Plastic SCM, RCS etc.).

In this chapter, we will describe cvs (version control system) and Subversion for controlling and administering multiple file revisions, automating storage, reading, identifying and merging different revisions. These programs are useful when a text is revised frequently and includes source code, executables, libraries, documentation, graphs, articles and other files. [Pos03e, Mys, Ced]

The reasoning behind using cvs and Subversion is that cvs is one of the most commonly used traditional packages and Subversion is a version control system software designed specifically to replace the popular cvs and to resolve several of its deficiencies. Subversion is also known as svn since this is the name of the command line tool. An important feature of Subversion is that, unlike CVS, the files with versions do not each have an independent revision number. Instead, the entire repository has a single version number that identifies a shared status of all the repository's files at a certain point in time.

Bibliography

bibliography