Attribute and dump operators #

When you do not know what a variable holds, attribute and dump list it. Both are registered by eZTemplateAttributeOperator (lib/eztemplate/classes/eztemplateattributeoperator.php) through lib/eztemplate/classes/eztemplateautoload.php. They walk arrays and every object that has the attributes() and attribute() methods (nodes, content objects, attributes, users and the other persistent objects), one level per step, and print one line per key.

The output is produced by a formatter chosen in template.ini:

[AttributeOperator]
DefaultFormatter=html
OutputFormatter[html]=ezpAttributeOperatorHTMLFormatter
OutputFormatter[text]=ezpAttributeOperatorTextFormatter

ezpAttributeOperatorManager::getOutputFormatter() (kernel/private/eztemplate/ezpattributeoperatormanager.php) falls back to DefaultFormatter for a name that is not registered. An extension can add a formatter class that implements ezpAttributeOperatorFormatterInterface and list it under OutputFormatter[].

Contents #

1. attribute
2. dump
3. Safety
4. Related pages

attribute #

attribute #

Lists the keys of the piped array or object, by default without values and two levels deep.

Input: the value to inspect, piped. Returns the formatted listing; a scalar or an empty array is returned as the formatter exports it. Generated reference: attribute.

The fields of the Fit & Healthy front page (node 89), one level, as text:

{def $node = fetch( 'content', 'node', hash( 'node_id', 89 ) )}
{$node.data_map|attribute( 'show', 1, 'text' )}

Output:

title (object[eZContentObjectAttribute]=Object)
metadata (object[eZContentObjectAttribute]=Object)
teaser_title (object[eZContentObjectAttribute]=Object)
teaser_intro (object[eZContentObjectAttribute]=Object)
full_intro (object[eZContentObjectAttribute]=Object)
image (object[eZContentObjectAttribute]=Object)
tags (object[eZContentObjectAttribute]=Object)

Without show, only names and types:

{def $hash = hash( 'title', 'Fit & Healthy', 'tags', array( 'fit', 'healthy' ) )}
{$hash|attribute( '', 2, 'text' )}

Output:

title (string)
tags (array)
    0 (string)
    1 (string)

dump #

dump #

The same walk with other defaults: values shown, one level deep.

Generated reference: dump.

{def $hash = hash( 'title', 'Fit & Healthy', 'tags', array( 'fit', 'healthy' ) )}
{$hash|dump( 'show', 2, 'text' )}

Output:

title (string='Fit & Healthy')
tags (array=Array(2))
    0 (string='fit')
    1 (string='healthy')

With the default formatter, html, the result is a table:

{hash( 'node_id', 89, 'name', 'Fit & Healthy' )|dump}

Output:

<table><tr><th align="left">Attribute</th>
<th align="left">Type</th>
<th align="left">Value</th>
</tr>
<tr><td>node_id</td>
<td>integer</td>
<td>89</td>
</tr>
<tr><td>name</td>
<td>string</td>
<td>'Fit & Healthy'</td>
</tr>
</table>

A scalar comes back as it was:

[{'plain text'|dump( 'show', 1, 'text' )}] [{42|attribute( 'show', 1, 'text' )}]

Output:

['plain text'] [42]

Safety #

• The HTML formatter does not escape values: 'Fit & Healthy' above came out with a bare &. A value that holds markup or script is printed as it is. Use these operators on a development installation, or inside a block that only administrators see, and remove them before a template goes live.
• A deep level on a node walks into its object, its versions, its parents and their attributes: every level multiplies the number of database queries. Start with max_val 1.
• Some attributes are computed when they are read. Listing them with values runs that code, so a dump can be slow even when the page is small.

The examples were rendered by the interpreter and again with the template compiler on a development installation; both gave the same output.

Related pages #

• Template debugging
• Nodes, objects and attributes
• Type operators: get_type and get_class for one value
• All template operators (generated)
• Free documentation: this page is free to copy, translate and improve under the GNU FDL 1.3.