Implementing Data Sources

Templating System : Developer Guide : User Guide

Data sources are implemented in a Tcl script using regular Tcl variables, lists and arrays. The templating system includes a set of procedures to simplify the construction of data sources from relational database queries.

The Structure of Data Sources

The templating system can manipulate four basic types of structures as data sources:

onevalueA simple scalar, such as a user's first name or the total due on a purchase order.
onelistA list of simple scalars.
onerowA one-row data table, with values in one or more columns.
multirowA multi-row, multi-column data table.

onevalue

onevalue data sources are implemented simply by setting a Tcl variable:

set name "Walter Cronkite"

The query procedure may be used to set a onevalue data source based on a database query:

query name onevalue "select name from users where user_id = 123"

You can embed a onevalue data source in a template with simple variable substitution.

onerow

onerow data sources are implemented as Tcl arrays:

set name(first_name) Walter
set name(last_name) Cronkite

The query procedure may be used as a convenient way to store the result of a one-row database query into an array:

query name onerow "
  select 
    first_name, last_name 
  from 
    users 
  where  
    user_id = 123"

You can embed references to column values of a onerow data source in a template with simple variable substitution.

onelist

onelist data sources are implemented by creating a Tcl list:

set names [list "Walter" "Fred" "Susy" "Frieda"]

The query procedure may be used to set a onelist data source based on a one-column database query:

query name onevalue "select name from users"

You can iterate over a onelist data source in a template with the list tag.

multirow

multirow data sources are not represented by a single Tcl data structure. As such the templating system includes a special API for constructing and manipulating them.

multirow create cars make model year
multirow append cars "Toyota" "Camry" "1996"
multirow append cars "Volvo" "960" "1995"

The query procedure may be used as a convenient way to store the result of a multi-row, multi-column database query into a multirow data source:

query name multirow "
  select 
    make, model, year
  from 
    cars"

You can iterate over a multirow data source in a template with the multiple tag.


templating@arsdigita.com