Custom template tags and filters

Django’s template language comes with a wide variety of built-intags and filters designed to address thepresentation logic needs of your application. Nevertheless, you mayfind yourself needing functionality that is not covered by the coreset of template primitives. You can extend the template engine bydefining custom tags and filters using Python, and then make themavailable to your templates using the {% load %} tag.

Code layout

The most common place to specify custom template tags and filters is insidea Django app. If they relate to an existing app, it makes sense to bundle themthere; otherwise, they can be added to a new app. When a Django app is addedto INSTALLED_APPS, any tags it defines in the conventional locationdescribed below are automatically made available to load within templates.

The app should contain a templatetags directory, at the same level asmodels.py, views.py, etc. If this doesn’t already exist, create it -don’t forget the init.py file to ensure the directory is treated as aPython package.

Development server won’t automatically restart

After adding the templatetags module, you will need to restart yourserver before you can use the tags or filters in templates.

Your custom tags and filters will live in a module inside the templatetagsdirectory. The name of the module file is the name you’ll use to load the tagslater, so be careful to pick a name that won’t clash with custom tags andfilters in another app.

For example, if your custom tags/filters are in a file calledpoll_extras.py, your app layout might look like this:

  1. polls/
  2. __init__.py
  3. models.py
  4. templatetags/
  5. __init__.py
  6. poll_extras.py
  7. views.py

And in your template you would use the following:

  1. {% load poll_extras %}

The app that contains the custom tags must be in INSTALLED_APPS inorder for the {% load %} tag to work. This is a security feature:It allows you to host Python code for many template libraries on a single hostmachine without enabling access to all of them for every Django installation.

There’s no limit on how many modules you put in the templatetags package.Just keep in mind that a {% load %} statement will loadtags/filters for the given Python module name, not the name of the app.

To be a valid tag library, the module must contain a module-level variablenamed register that is a template.Library instance, in which all thetags and filters are registered. So, near the top of your module, put thefollowing:

  1. from django import template
  2.  
  3. register = template.Library()

Alternatively, template tag modules can be registered through the'libraries' argument toDjangoTemplates. This is useful ifyou want to use a different label from the template tag module name whenloading template tags. It also enables you to register tags without installingan application.

Behind the scenes

For a ton of examples, read the source code for Django’s default filtersand tags. They’re in django/template/defaultfilters.py anddjango/template/defaulttags.py, respectively.

For more information on the load tag, read its documentation.

Writing custom template filters

Custom filters are Python functions that take one or two arguments:

  • The value of the variable (input) – not necessarily a string.
  • The value of the argument – this can have a default value, or be leftout altogether.For example, in the filter {{ var|foo:"bar" }}, the filter foo would bepassed the variable var and the argument "bar".

Since the template language doesn’t provide exception handling, any exceptionraised from a template filter will be exposed as a server error. Thus, filterfunctions should avoid raising exceptions if there is a reasonable fallbackvalue to return. In case of input that represents a clear bug in a template,raising an exception may still be better than silent failure which hides thebug.

Here’s an example filter definition:

  1. def cut(value, arg):
  2. """Removes all values of arg from the given string"""
  3. return value.replace(arg, '')

And here’s an example of how that filter would be used:

  1. {{ somevariable|cut:"0" }}

Most filters don’t take arguments. In this case, leave the argument out of yourfunction:

  1. def lower(value): # Only one argument.
  2. """Converts a string into all lowercase"""
  3. return value.lower()

Registering custom filters

  • django.template.Library.filter()
  • Once you’ve written your filter definition, you need to register it withyour Library instance, to make it available to Django’s template language:
  1. register.filter('cut', cut)
  2. register.filter('lower', lower)

The Library.filter() method takes two arguments:

  • The name of the filter – a string.
  • The compilation function – a Python function (not the name of thefunction as a string).You can use register.filter() as a decorator instead:

  1. @register.filter(name='cut')def cut(value, arg): return value.replace(arg, '')

  2. @register.filterdef lower(value): return value.lower()

If you leave off the name argument, as in the second example above, Djangowill use the function’s name as the filter name.

Finally, register.filter() also accepts three keyword arguments,is_safe, needs_autoescape, and expects_localtime. These argumentsare described in filters and auto-escaping andfilters and time zones below.

Template filters that expect strings

  • django.template.defaultfilters.stringfilter()
  • If you’re writing a template filter that only expects a string as the firstargument, you should use the decorator stringfilter. This willconvert an object to its string value before being passed to your function:
  1. from django import template
  2. from django.template.defaultfilters import stringfilter
  3.  
  4. register = template.Library()
  5.  
  6. @register.filter
  7. @stringfilter
  8. def lower(value):
  9. return value.lower()

This way, you’ll be able to pass, say, an integer to this filter, and itwon’t cause an AttributeError (because integers don’t have lower()methods).

Filters and auto-escaping

When writing a custom filter, give some thought to how the filter will interactwith Django’s auto-escaping behavior. Note that two types of strings can bepassed around inside the template code:

  • Raw strings are the native Python strings. On output, they’re escaped ifauto-escaping is in effect and presented unchanged, otherwise.

  • Safe strings are strings that have been marked safe from furtherescaping at output time. Any necessary escaping has already been done.They’re commonly used for output that contains raw HTML that is intendedto be interpreted as-is on the client side.

Internally, these strings are of typeSafeString. You can test for themusing code like:

  1. from django.utils.safestring import SafeString
  2.  
  3. if isinstance(value, SafeString):
  4. # Do something with the "safe" string.
  5. ...

Template filter code falls into one of two situations:

  • Your filter does not introduce any HTML-unsafe characters (<, >,', " or &) into the result that were not already present. Inthis case, you can let Django take care of all the auto-escapinghandling for you. All you need to do is set the is_safe flag to Truewhen you register your filter function, like so:

  1. @register.filter(is_safe=True)def myfilter(value): return value

This flag tells Django that if a “safe” string is passed into yourfilter, the result will still be “safe” and if a non-safe string ispassed in, Django will automatically escape it, if necessary.

You can think of this as meaning “this filter is safe – it doesn’tintroduce any possibility of unsafe HTML.”

The reason is_safe is necessary is because there are plenty ofnormal string operations that will turn a SafeData object back intoa normal str object and, rather than try to catch them all, which wouldbe very difficult, Django repairs the damage after the filter has completed.

For example, suppose you have a filter that adds the string xx tothe end of any input. Since this introduces no dangerous HTML charactersto the result (aside from any that were already present), you shouldmark your filter with is_safe:

  1. @register.filter(is_safe=True)def add_xx(value): return '%sxx' % value

When this filter is used in a template where auto-escaping is enabled,Django will escape the output whenever the input is not already markedas “safe”.

By default, is_safe is False, and you can omit it from any filterswhere it isn’t required.

Be careful when deciding if your filter really does leave safe stringsas safe. If you’re removing characters, you might inadvertently leaveunbalanced HTML tags or entities in the result. For example, removing a> from the input might turn <a> into <a, which would need tobe escaped on output to avoid causing problems. Similarly, removing asemicolon (;) can turn &amp; into &amp, which is no longer avalid entity and thus needs further escaping. Most cases won’t be nearlythis tricky, but keep an eye out for any problems like that whenreviewing your code.

Marking a filter is_safe will coerce the filter’s return value toa string. If your filter should return a boolean or other non-stringvalue, marking it is_safe will probably have unintendedconsequences (such as converting a boolean False to the string‘False’).

  • Alternatively, your filter code can manually take care of any necessaryescaping. This is necessary when you’re introducing new HTML markup intothe result. You want to mark the output as safe from furtherescaping so that your HTML markup isn’t escaped further, so you’ll needto handle the input yourself.

To mark the output as a safe string, usedjango.utils.safestring.mark_safe().

Be careful, though. You need to do more than just mark the output assafe. You need to ensure it really is safe, and what you do depends onwhether auto-escaping is in effect. The idea is to write filters thatcan operate in templates where auto-escaping is either on or off inorder to make things easier for your template authors.

In order for your filter to know the current auto-escaping state, set theneeds_autoescape flag to True when you register your filter function.(If you don’t specify this flag, it defaults to False). This flag tellsDjango that your filter function wants to be passed an extra keywordargument, called autoescape, that is True if auto-escaping is ineffect and False otherwise. It is recommended to set the default of theautoescape parameter to True, so that if you call the functionfrom Python code it will have escaping enabled by default.

For example, let’s write a filter that emphasizes the first character ofa string:

  1. from django import template
  2. from django.utils.html import conditional_escape
  3. from django.utils.safestring import mark_safe
  4.  
  5. register = template.Library()
  6.  
  7. @register.filter(needs_autoescape=True)
  8. def initial_letter_filter(text, autoescape=True):
  9. first, other = text[0], text[1:]
  10. if autoescape:
  11. esc = conditional_escape
  12. else:
  13. esc = lambda x: x
  14. result = '<strong>%s</strong>%s' % (esc(first), esc(other))
  15. return mark_safe(result)

The needs_autoescape flag and the autoescape keyword argument meanthat our function will know whether automatic escaping is in effect when thefilter is called. We use autoescape to decide whether the input dataneeds to be passed through django.utils.html.conditional_escape or not.(In the latter case, we use the identity function as the “escape” function.)The conditional_escape() function is like escape() except it onlyescapes input that is not a SafeData instance. If a SafeDatainstance is passed to conditional_escape(), the data is returnedunchanged.

Finally, in the above example, we remember to mark the result as safeso that our HTML is inserted directly into the template without furtherescaping.

There’s no need to worry about the is_safe flag in this case(although including it wouldn’t hurt anything). Whenever you manuallyhandle the auto-escaping issues and return a safe string, theis_safe flag won’t change anything either way.

Warning

Avoiding XSS vulnerabilities when reusing built-in filters

Django’s built-in filters have autoescape=True by default in order toget the proper autoescaping behavior and avoid a cross-site scriptvulnerability.

In older versions of Django, be careful when reusing Django’s built-infilters as autoescape defaults to None. You’ll need to passautoescape=True to get autoescaping.

For example, if you wanted to write a custom filter calledurlize_and_linebreaks that combined the urlize andlinebreaksbr filters, the filter would look like:

  1. from django.template.defaultfilters import linebreaksbr, urlize
  2.  
  3. @register.filter(needs_autoescape=True)
  4. def urlize_and_linebreaks(text, autoescape=True):
  5. return linebreaksbr(
  6. urlize(text, autoescape=autoescape),
  7. autoescape=autoescape
  8. )

Then:

  1. {{ comment|urlize_and_linebreaks }}

would be equivalent to:

  1. {{ comment|urlize|linebreaksbr }}

Filters and time zones

If you write a custom filter that operates on datetimeobjects, you’ll usually register it with the expects_localtime flag set toTrue:

  1. @register.filter(expects_localtime=True)def businesshours(value): try: return 9 <= value.hour < 17 except AttributeError: return ''

When this flag is set, if the first argument to your filter is a time zoneaware datetime, Django will convert it to the current time zone before passingit to your filter when appropriate, according to rules for time zonesconversions in templates.

Writing custom template tags

Tags are more complex than filters, because tags can do anything. Djangoprovides a number of shortcuts that make writing most types of tags easier.First we’ll explore those shortcuts, then explain how to write a tag fromscratch for those cases when the shortcuts aren’t powerful enough.

Simple tags

  • django.template.Library.simple_tag()
  • Many template tags take a number of arguments – strings or template variables– and return a result after doing some processing based solely onthe input arguments and some external information. For example, acurrent_time tag might accept a format string and return the time as astring formatted accordingly.

To ease the creation of these types of tags, Django provides a helper function,simple_tag. This function, which is a method ofdjango.template.Library, takes a function that accepts any number ofarguments, wraps it in a render function and the other necessary bitsmentioned above and registers it with the template system.

Our current_time function could thus be written like this:

  1. import datetime
  2. from django import template
  3.  
  4. register = template.Library()
  5.  
  6. @register.simple_tag
  7. def current_time(format_string):
  8. return datetime.datetime.now().strftime(format_string)

A few things to note about the simple_tag helper function:

  • Checking for the required number of arguments, etc., has already beendone by the time our function is called, so we don’t need to do that.
  • The quotes around the argument (if any) have already been stripped away,so we receive a plain string.
  • If the argument was a template variable, our function is passed thecurrent value of the variable, not the variable itself.Unlike other tag utilities, simple_tag passes its output throughconditional_escape() if the template context is inautoescape mode, to ensure correct HTML and protect you from XSSvulnerabilities.

If additional escaping is not desired, you will need to usemark_safe() if you are absolutely sure that yourcode does not contain XSS vulnerabilities. For building small HTML snippets,use of format_html() instead of mark_safe() isstrongly recommended.

If your template tag needs to access the current context, you can use thetakes_context argument when registering your tag:

  1. @register.simple_tag(takes_context=True)def current_time(context, format_string): timezone = context['timezone'] return your_get_current_time_method(timezone, format_string)

Note that the first argument must be called context.

For more information on how the takes_context option works, see the sectionon inclusion tags.

If you need to rename your tag, you can provide a custom name for it:

  1. register.simple_tag(lambda x: x - 1, name='minusone')
  2.  
  3. @register.simple_tag(name='minustwo')
  4. def some_function(value):
  5. return value - 2

simple_tag functions may accept any number of positional or keywordarguments. For example:

  1. @register.simple_tagdef my_tag(a, b, args, *kwargs): warning = kwargs['warning'] profile = kwargs['profile'] return

Then in the template any number of arguments, separated by spaces, may bepassed to the template tag. Like in Python, the values for keyword argumentsare set using the equal sign (“=”) and must be provided after thepositional arguments. For example:

  1. {% my_tag 123 "abcd" book.title warning=message|lower profile=user.profile %}

It’s possible to store the tag results in a template variable rather thandirectly outputting it. This is done by using the as argument followed bythe variable name. Doing so enables you to output the content yourself whereyou see fit:

  1. {% current_time "%Y-%m-%d %I:%M %p" as the_time %}
  2. <p>The time is {{ the_time }}.</p>

Inclusion tags

  • django.template.Library.inclusion_tag()
  • Another common type of template tag is the type that displays some data byrendering another template. For example, Django’s admin interface uses customtemplate tags to display the buttons along the bottom of the “add/change” formpages. Those buttons always look the same, but the link targets changedepending on the object being edited – so they’re a perfect case for using asmall template that is filled with details from the current object. (In theadmin’s case, this is the submit_row tag.)

These sorts of tags are called “inclusion tags”.

Writing inclusion tags is probably best demonstrated by example. Let’s write atag that outputs a list of choices for a given Poll object, such as wascreated in the tutorials. We’ll use the tag like this:

  1. {% show_results poll %}

…and the output will be something like this:

  1. <ul>
  2. <li>First choice</li>
  3. <li>Second choice</li>
  4. <li>Third choice</li>
  5. </ul>

First, define the function that takes the argument and produces a dictionary ofdata for the result. The important point here is we only need to return adictionary, not anything more complex. This will be used as a template contextfor the template fragment. Example:

def show_results(poll):
    choices = poll.choice_set.all()
    return {'choices': choices}

Next, create the template used to render the tag’s output. This template is afixed feature of the tag: the tag writer specifies it, not the templatedesigner. Following our example, the template is very short:

<ul>
{% for choice in choices %}
    <li> {{ choice }} </li>
{% endfor %}
</ul>

Now, create and register the inclusion tag by calling the inclusion_tag()method on a Library object. Following our example, if the above template isin a file called results.html in a directory that’s searched by thetemplate loader, we’d register the tag like this:

# Here, register is a django.template.Library instance, as before
@register.inclusion_tag('results.html')
def show_results(poll):
    ...

Alternatively it is possible to register the inclusion tag using adjango.template.Template instance:

from django.template.loader import get_template
t = get_template('results.html')
register.inclusion_tag(t)(show_results)

…when first creating the function.

Sometimes, your inclusion tags might require a large number of arguments,making it a pain for template authors to pass in all the arguments and remembertheir order. To solve this, Django provides a takes_context option forinclusion tags. If you specify takes_context in creating a template tag,the tag will have no required arguments, and the underlying Python functionwill have one argument – the template context as of when the tag was called.

For example, say you’re writing an inclusion tag that will always be used in acontext that contains home_link and home_title variables that pointback to the main page. Here’s what the Python function would look like:

@register.inclusion_tag('link.html', takes_context=True)def jump_link(context):    return {        'link': context['home_link'],        'title': context['home_title'],    }

Note that the first parameter to the function must be called context.

In that register.inclusion_tag() line, we specified takes_context=Trueand the name of the template. Here’s what the template link.html might looklike:

Jump directly to <a href="{{ link }}">{{ title }}</a>.

Then, any time you want to use that custom tag, load its library and call itwithout any arguments, like so:

{% jump_link %}

Note that when you’re using takes_context=True, there’s no need to passarguments to the template tag. It automatically gets access to the context.

The takes_context parameter defaults to False. When it’s set toTrue, the tag is passed the context object, as in this example. That’s theonly difference between this case and the previous inclusion_tag example.

inclusion_tag functions may accept any number of positional or keywordarguments. For example:

@register.inclusion_tag('my_template.html')def my_tag(a, b, args, *kwargs):    warning = kwargs['warning']    profile = kwargs['profile']    …    return …

Then in the template any number of arguments, separated by spaces, may bepassed to the template tag. Like in Python, the values for keyword argumentsare set using the equal sign (“=”) and must be provided after thepositional arguments. For example:

{% my_tag 123 "abcd" book.title warning=message|lower profile=user.profile %}

Advanced custom template tags

Sometimes the basic features for custom template tag creation aren’t enough.Don’t worry, Django gives you complete access to the internals required to builda template tag from the ground up.

A quick overview

The template system works in a two-step process: compiling and rendering. Todefine a custom template tag, you specify how the compilation works and howthe rendering works.

When Django compiles a template, it splits the raw template text into‘’nodes’‘. Each node is an instance of django.template.Node and hasa render() method. A compiled template is a list of Node objects. Whenyou call render() on a compiled template object, the template callsrender() on each Node in its node list, with the given context. Theresults are all concatenated together to form the output of the template.

Thus, to define a custom template tag, you specify how the raw template tag isconverted into a Node (the compilation function), and what the node’srender() method does.

Writing the compilation function

For each template tag the template parser encounters, it calls a Pythonfunction with the tag contents and the parser object itself. This function isresponsible for returning a Node instance based on the contents of the tag.

For example, let’s write a full implementation of our template tag,{% current_time %}, that displays the current date/time, formatted accordingto a parameter given in the tag, in strftime() syntax. It’s a goodidea to decide the tag syntax before anything else. In our case, let’s say thetag should be used like this:

<p>The time is {% current_time "%Y-%m-%d %I:%M %p" %}.</p>

The parser for this function should grab the parameter and create a Nodeobject:

from django import template

def do_current_time(parser, token):
    try:
        # split_contents() knows not to split quoted strings.
        tag_name, format_string = token.split_contents()
    except ValueError:
        raise template.TemplateSyntaxError(
            "%r tag requires a single argument" % token.contents.split()[0]
        )
    if not (format_string[0] == format_string[-1] and format_string[0] in ('"', "'")):
        raise template.TemplateSyntaxError(
            "%r tag's argument should be in quotes" % tag_name
        )
    return CurrentTimeNode(format_string[1:-1])

Notes:

  • parser is the template parser object. We don’t need it in thisexample.
  • token.contents is a string of the raw contents of the tag. In ourexample, it’s 'current_time "%Y-%m-%d %I:%M %p"'.
  • The token.splitcontents() method separates the arguments on spaceswhile keeping quoted strings together. The more straightforwardtoken.contents.split() wouldn’t be as robust, as it would naivelysplit on _all spaces, including those within quoted strings. It’s a goodidea to always use token.split_contents().
  • This function is responsible for raisingdjango.template.TemplateSyntaxError, with helpful messages, forany syntax error.
  • The TemplateSyntaxError exceptions use the tag_name variable.Don’t hard-code the tag’s name in your error messages, because thatcouples the tag’s name to your function. token.contents.split()[0]will ‘’always’’ be the name of your tag – even when the tag has noarguments.
  • The function returns a CurrentTimeNode with everything the node needsto know about this tag. In this case, it passes the argument –"%Y-%m-%d %I:%M %p". The leading and trailing quotes from thetemplate tag are removed in format_string[1:-1].
  • The parsing is very low-level. The Django developers have experimentedwith writing small frameworks on top of this parsing system, usingtechniques such as EBNF grammars, but those experiments made the templateengine too slow. It’s low-level because that’s fastest.

Writing the renderer

The second step in writing custom tags is to define a Node subclass thathas a render() method.

Continuing the above example, we need to define CurrentTimeNode:

import datetime
from django import template

class CurrentTimeNode(template.Node):
    def __init__(self, format_string):
        self.format_string = format_string

    def render(self, context):
        return datetime.datetime.now().strftime(self.format_string)

Notes:

  • init() gets the formatstring from docurrent_time().Always pass any options/parameters/arguments to a Node via its__init().
  • The render() method is where the work actually happens.
  • render() should generally fail silently, particularly in a productionenvironment. In some cases however, particularly ifcontext.template.engine.debug is True, this method may raise anexception to make debugging easier. For example, several core tags raisedjango.template.TemplateSyntaxError if they receive the wrong number ortype of arguments.Ultimately, this decoupling of compilation and rendering results in anefficient template system, because a template can render multiple contextswithout having to be parsed multiple times.

Auto-escaping considerations

The output from template tags is not automatically run through theauto-escaping filters (with the exception ofsimple_tag() as described above). However, thereare still a couple of things you should keep in mind when writing a templatetag.

If the render() function of your template stores the result in a contextvariable (rather than returning the result in a string), it should take careto call mark_safe() if appropriate. When the variable is ultimatelyrendered, it will be affected by the auto-escape setting in effect at thetime, so content that should be safe from further escaping needs to be markedas such.

Also, if your template tag creates a new context for performing somesub-rendering, set the auto-escape attribute to the current context’s value.The init method for the Context class takes a parameter calledautoescape that you can use for this purpose. For example:

from django.template import Context

def render(self, context):
    # ...
    new_context = Context({'var': obj}, autoescape=context.autoescape)
    # ... Do something with new_context ...

This is not a very common situation, but it’s useful if you’re rendering atemplate yourself. For example:

def render(self, context):
    t = context.template.engine.get_template('small_fragment.html')
    return t.render(Context({'var': obj}, autoescape=context.autoescape))

If we had neglected to pass in the current context.autoescape value to ournew Context in this example, the results would have always beenautomatically escaped, which may not be the desired behavior if the templatetag is used inside a {% autoescape off %} block.

Thread-safety considerations

Once a node is parsed, its render method may be called any number of times.Since Django is sometimes run in multi-threaded environments, a single node maybe simultaneously rendering with different contexts in response to two separaterequests. Therefore, it’s important to make sure your template tags are threadsafe.

To make sure your template tags are thread safe, you should never store stateinformation on the node itself. For example, Django provides a builtincycle template tag that cycles among a list of given strings each timeit’s rendered:

{% for o in some_list %}
    <tr class="{% cycle 'row1' 'row2' %}">
        ...
    </tr>
{% endfor %}

A naive implementation of CycleNode might look something like this:

import itertools
from django import template

class CycleNode(template.Node):
    def __init__(self, cyclevars):
        self.cycle_iter = itertools.cycle(cyclevars)

    def render(self, context):
        return next(self.cycle_iter)

But, suppose we have two templates rendering the template snippet from above atthe same time:

  • Thread 1 performs its first loop iteration, CycleNode.render()returns ‘row1’
  • Thread 2 performs its first loop iteration, CycleNode.render()returns ‘row2’
  • Thread 1 performs its second loop iteration, CycleNode.render()returns ‘row1’
  • Thread 2 performs its second loop iteration, CycleNode.render()returns ‘row2’The CycleNode is iterating, but it’s iterating globally. As far as Thread 1and Thread 2 are concerned, it’s always returning the same value. This isnot what we want!

To address this problem, Django provides a render_context that’s associatedwith the context of the template that is currently being rendered. Therender_context behaves like a Python dictionary, and should be used tostore Node state between invocations of the render method.

Let’s refactor our CycleNode implementation to use the render_context:

class CycleNode(template.Node):
    def __init__(self, cyclevars):
        self.cyclevars = cyclevars

    def render(self, context):
        if self not in context.render_context:
            context.render_context[self] = itertools.cycle(self.cyclevars)
        cycle_iter = context.render_context[self]
        return next(cycle_iter)

Note that it’s perfectly safe to store global information that will not changethroughout the life of the Node as an attribute. In the case ofCycleNode, the cyclevars argument doesn’t change after the Node isinstantiated, so we don’t need to put it in the render_context. But stateinformation that is specific to the template that is currently being rendered,like the current iteration of the CycleNode, should be stored in therender_context.

Note

Notice how we used self to scope the CycleNode specific informationwithin the render_context. There may be multiple CycleNodes in agiven template, so we need to be careful not to clobber another node’sstate information. The easiest way to do this is to always use self asthe key into render_context. If you’re keeping track of several statevariables, make render_context[self] a dictionary.

Registering the tag

Finally, register the tag with your module’s Library instance, as explainedin writing custom template filtersabove. Example:

register.tag('current_time', do_current_time)

The tag() method takes two arguments:

  • The name of the template tag – a string. If this is left out, thename of the compilation function will be used.
  • The compilation function – a Python function (not the name of thefunction as a string).As with filter registration, it is also possible to use this as a decorator:

@register.tag(name="current_time")def do_current_time(parser, token):    …

@register.tagdef shout(parser, token): …

If you leave off the name argument, as in the second example above, Djangowill use the function’s name as the tag name.

Passing template variables to the tag

Although you can pass any number of arguments to a template tag usingtoken.split_contents(), the arguments are all unpacked asstring literals. A little more work is required in order to pass dynamiccontent (a template variable) to a template tag as an argument.

While the previous examples have formatted the current time into a string andreturned the string, suppose you wanted to pass in aDateTimeField from an object and have the templatetag format that date-time:

<p>This post was last updated at {% format_time blog_entry.date_updated "%Y-%m-%d %I:%M %p" %}.</p>

Initially, token.split_contents() will return three values:

  • The tag name format_time.
  • The string 'blog_entry.date_updated' (without the surroundingquotes).
  • The formatting string '"%Y-%m-%d %I:%M %p"'. The return value fromsplit_contents() will include the leading and trailing quotes forstring literals like this.Now your tag should begin to look like this:
from django import template

def do_format_time(parser, token):
    try:
        # split_contents() knows not to split quoted strings.
        tag_name, date_to_be_formatted, format_string = token.split_contents()
    except ValueError:
        raise template.TemplateSyntaxError(
            "%r tag requires exactly two arguments" % token.contents.split()[0]
        )
    if not (format_string[0] == format_string[-1] and format_string[0] in ('"', "'")):
        raise template.TemplateSyntaxError(
            "%r tag's argument should be in quotes" % tag_name
        )
    return FormatTimeNode(date_to_be_formatted, format_string[1:-1])

You also have to change the renderer to retrieve the actual contents of thedate_updated property of the blog_entry object. This can beaccomplished by using the Variable() class in django.template.

To use the Variable class, instantiate it with the name of the variable tobe resolved, and then call variable.resolve(context). So, for example:

class FormatTimeNode(template.Node):
    def __init__(self, date_to_be_formatted, format_string):
        self.date_to_be_formatted = template.Variable(date_to_be_formatted)
        self.format_string = format_string

    def render(self, context):
        try:
            actual_date = self.date_to_be_formatted.resolve(context)
            return actual_date.strftime(self.format_string)
        except template.VariableDoesNotExist:
            return ''

Variable resolution will throw a VariableDoesNotExist exception if itcannot resolve the string passed to it in the current context of the page.

Setting a variable in the context

The above examples output a value. Generally, it’s more flexible if yourtemplate tags set template variables instead of outputting values. That way,template authors can reuse the values that your template tags create.

To set a variable in the context, use dictionary assignment on the contextobject in the render() method. Here’s an updated version ofCurrentTimeNode that sets a template variable current_time instead ofoutputting it:

import datetime
from django import template

class CurrentTimeNode2(template.Node):
    def __init__(self, format_string):
        self.format_string = format_string
    def render(self, context):
        context['current_time'] = datetime.datetime.now().strftime(self.format_string)
        return ''

Note that render() returns the empty string. render() should alwaysreturn string output. If all the template tag does is set a variable,render() should return the empty string.

Here’s how you’d use this new version of the tag:

{% current_time "%Y-%m-%d %I:%M %p" %}<p>The time is {{ current_time }}.</p>

Variable scope in context

Any variable set in the context will only be available in the sameblock of the template in which it was assigned. This behavior isintentional; it provides a scope for variables so that they don’t conflictwith context in other blocks.

But, there’s a problem with CurrentTimeNode2: The variable namecurrent_time is hard-coded. This means you’ll need to make sure yourtemplate doesn’t use {{ current_time }} anywhere else, because the{% current_time %} will blindly overwrite that variable’s value. A cleanersolution is to make the template tag specify the name of the output variable,like so:

{% current_time "%Y-%m-%d %I:%M %p" as my_current_time %}
<p>The current time is {{ my_current_time }}.</p>

To do that, you’ll need to refactor both the compilation function and Nodeclass, like so:

import re

class CurrentTimeNode3(template.Node):
    def __init__(self, format_string, var_name):
        self.format_string = format_string
        self.var_name = var_name
    def render(self, context):
        context[self.var_name] = datetime.datetime.now().strftime(self.format_string)
        return ''

def do_current_time(parser, token):
    # This version uses a regular expression to parse tag contents.
    try:
        # Splitting by None == splitting by spaces.
        tag_name, arg = token.contents.split(None, 1)
    except ValueError:
        raise template.TemplateSyntaxError(
            "%r tag requires arguments" % token.contents.split()[0]
        )
    m = re.search(r'(.*?) as (\w+)', arg)
    if not m:
        raise template.TemplateSyntaxError("%r tag had invalid arguments" % tag_name)
    format_string, var_name = m.groups()
    if not (format_string[0] == format_string[-1] and format_string[0] in ('"', "'")):
        raise template.TemplateSyntaxError(
            "%r tag's argument should be in quotes" % tag_name
        )
    return CurrentTimeNode3(format_string[1:-1], var_name)

The difference here is that do_current_time() grabs the format string andthe variable name, passing both to CurrentTimeNode3.

Finally, if you only need to have a simple syntax for your customcontext-updating template tag, consider using thesimple_tag() shortcut, which supports assigningthe tag results to a template variable.

Parsing until another block tag

Template tags can work in tandem. For instance, the standard{% comment %} tag hides everything until {% endcomment %}.To create a template tag such as this, use parser.parse() in yourcompilation function.

Here’s how a simplified {% comment %} tag might be implemented:

def do_comment(parser, token):
    nodelist = parser.parse(('endcomment',))
    parser.delete_first_token()
    return CommentNode()

class CommentNode(template.Node):
    def render(self, context):
        return ''

Note

The actual implementation of {% comment %} is slightlydifferent in that it allows broken template tags to appear between{% comment %} and {% endcomment %}. It does so by callingparser.skip_past('endcomment') instead of parser.parse(('endcomment',))followed by parser.delete_first_token(), thus avoiding the generation of anode list.

parser.parse() takes a tuple of names of block tags ‘’to parse until’‘. Itreturns an instance of django.template.NodeList, which is a list ofall Node objects that the parser encountered ‘’before’’ it encounteredany of the tags named in the tuple.

In "nodelist = parser.parse(('endcomment',))" in the above example,nodelist is a list of all nodes between the {% comment %} and{% endcomment %}, not counting {% comment %} and {% endcomment %}themselves.

After parser.parse() is called, the parser hasn’t yet “consumed” the{% endcomment %} tag, so the code needs to explicitly callparser.delete_first_token().

CommentNode.render() returns an empty string. Anything between{% comment %} and {% endcomment %} is ignored.

Parsing until another block tag, and saving contents

In the previous example, do_comment() discarded everything between{% comment %} and {% endcomment %}. Instead of doing that, it’spossible to do something with the code between block tags.

For example, here’s a custom template tag, {% upper %}, that capitalizeseverything between itself and {% endupper %}.

Usage:

{% upper %}This will appear in uppercase, {{ your_name }}.{% endupper %}

As in the previous example, we’ll use parser.parse(). But this time, wepass the resulting nodelist to the Node:

def do_upper(parser, token):
    nodelist = parser.parse(('endupper',))
    parser.delete_first_token()
    return UpperNode(nodelist)

class UpperNode(template.Node):
    def __init__(self, nodelist):
        self.nodelist = nodelist
    def render(self, context):
        output = self.nodelist.render(context)
        return output.upper()

The only new concept here is the self.nodelist.render(context) inUpperNode.render().

For more examples of complex rendering, see the source code of{% for %} in django/template/defaulttags.py and{% if %} in django/template/smartif.py.