|
The view page are used to retrieve information from the application's
data source and display it to the end user. The data source of a view page
is specified by one or two database queries, which are stored external to the
XML document describing the page itself to allow for re-use of the queries
across the pages. View pages can be used to display a list of the rows
returned from a query (multi-row view page) or to display the details on a specific
row (single-row view page). It is also possible to combine both features
on the same page, thus allowing for simultaneous presentation of information
from two independent data sources on a single page. This is mostly useful
for displaying related data from the two sides of one-to-many relationships.
In addition to specifying the query for retrieving the intended
row(s) from the database, the developer should specify how the attributes
of the retrieved row(s) are displayed in the constructed page. Normally the
attributes are directly displayed on the page; however it is also possible to
perform computations on the attributes and display the result.
The following figure shows the XML definition for the contactlist
page.
<?xml version="1.0" encoding="UTF-8"?>
<page type="view">
<title>Address Book</title>
<icon>contactlist.gif</icon>
<body>
<querysource>xmlquery/contact.xml</querysource>
<searchquerysource>xmlquery/contactsearch.xml</searchquerysource>
<defaultorder>lastname ASC</defaultorder>
<fields>
<field name="name" caption="First Name"/>
<field name="lastname" caption="Last Name"/>
<field name="phone" caption="Phone#"/>
<cfield name="address" caption="Address" code="addrhead"/>
<link name="linkdetail" on="id" target="contactdetail"
caption="Detail" text="Detail..."/>
<link name="linkedit" on="id" target="contactedit"
caption="Edit" text="Edit..."/>
<link name="linkdelete" on="id" target="contactdelete"
caption="Delete" text="Delete..."/>
</fields>
</body>
<code name="addrhead" language="php">
<![CDATA[
$text = substr($source, 0, 30);
if ($text) {
$text .= "...";
}
return $text;
]]>
</code>
</page>
Every page opens with the <page> tag. The type of the page is
denoted in the type attribute. The <body> tag denotes the beginning of a
list of rows. Within the <body> tag, the <querysource> tag specifies the
query for retrieving rows and the <defaultorder> tag specifies the initial order
used in displaying the rows to the user. The <fields> tag encompasses definition of the columns displayed on the page. The source of each column
is one of the attributes available in the result set returned from the execution
of the query. Three tags are allowed inside the
<fields> tag:
-
<field> tag is used for directly displaying the value of
an
attribute
-
<cfield> tag is used for displaying the result of applying
some computation on the value of the attribute
-
<link> tag is used for displaying a hyperlink to some
other page and sending the value of the attribute as the parameter
The following figure displays the contactlist page in a browser.
Contact list page in browser
For the <field> tags, the name attribute specifies which attribute
from the result set should be displayed. The title appearing in the head
of the column is provided by the caption attribute.
For the <cfield> tags, the name attribute is used to
determine the
source attribute for the column. The code attribute specifies the name of
a function whose definition is provided at the end of the document. For
computed fields the value of the attribute is passed to the specified function
and then the value that the function returns is shown in the appropriate cell.
The function body is written in a programming language supported by the
environment in which the tool runs (currently only PHP is supported).
Computed fields are one of the mechanisms that can be employed for extending the
basic functionality of the tool. In this example, since the address attribute is
of text type and can contain large amounts of text, we have used a computed
field to only print its first 30 characters. The
last three columns in the constructed page are three links to other pages of the
address book application. Every link sends the value of one attribute to
its target page. The links are specified by the <link> tag and they should
include the following attributes:
-
"name" is a unique identifier for the link
-
"target" determines the target page
-
"on" denotes the name of the attribute whose value is passed to the
target page
-
"caption" is printed in the caption of the column
-
"text" is the text that is repeated in every cell. If this
attribute is omitted, then caption is used instead.
In this example, the first link passes the value of the "id"
attribute to the contactdetail page, whose responsibility is displaying the
complete information for a single contact. When the user clicks on this
link for one of the rows, the value of the "id" attribute for the associated row
is sent to the contactdetail page. Upon receiving this value, the
contactdetail page would be able to extract the necessary information for the
intended
contact and display it to the user. As it is shown in the browser,
the rows are paginated in numbered pages so that the user can easily browse the
rows. It is also possible to sort the rows based on any of the
shown attributes in ascending or descending orders. These features are automatically
included in the generated pages and they do not need to be defined explicitly. |
