Introduction

Purpose

The RDB library was created to handle input and output related tasks for RDB tables. The interface consists of multiple layers. The casual user may use the library through the RDB object interface exclusively. This interface allows the user to open, close, read from, and write to RDB tables. The interface handles provides default means of interpreting data types of the columns as well as default data storage management. Examples of basic table manipulations via the RDB object interface are available.

For some tasks the user may wish to modify the structure of a table or explicitly manage the location and memory allocation of the data. In these cases, the user will need to make use of the RDBColumn interface in addition to the RDB interface. Examples of more advance table manipulations using the RDBColumn interface are also available.

RDB format

RDB is an ASCII text format. An RDB table consists of three distinct elements:

A comment is any line beginning zero or more spaces followed by a sharp sign (#). Characters following the sharp sign are considered to be the body of the comment. Comments may have additional structure within the body. It is possible to have comment variables, also known as header variables. A comment variable is said to exist if the sharp sign is immediately followed by a colon (:). The first word, non-whitespace and non-equal sign (=) characters, following the colon is the comment variable. The value of the comment variable is seperated from the variable by optional space (' ') and tab ('\t') characters and a mandatory equal sign. The variable value is then any text between the equal sign and the newline. It is common to restrict comments to the initial section of the file

The header consists of two lines. The first contains a tab seperated list of column names. Each column name is seperated by a single tab. The second line contains tab seperated column definitions. Definitions consist of an optional numeric value indicating the width, an S(tring), N(umeric), or M(onth) character denoting the type of the data, an optional < or > character indicating left or right justification of data, and an optional text description.

The data consist of tab seperated values, one per column name.

For a more definitive description of the format see http://hea-www.harvard.edu/MST/simul/software/docs/html/perlrdb/rdb.html .


Generated on Tue Sep 15 11:07:16 2009 for rdbxx by  doxygen 1.5.6