Comma separated lists in django templates

First choice: use the existing join template tag.

http://docs.djangoproject.com/en/dev/ref/templates/builtins/#join

Here's their example

{{ value|join:" // " }}

Second choice: do it in the view.

fruits_text = ", ".join( fruits )

Provide fruits_text to the template for rendering.

I would suggest a custom django templating filter rather than a custom tag -- filter is handier and simpler (where appropriate, like here). {{ fruits | joinby:", " }} looks like what I'd want to have for the purpose... with a custom joinby filter:

def joinby(value, arg):
    return arg.join(value)

which as you see is simplicity itself!

Here's the filter I wrote to solve my problem (it doesn't include the Oxford comma)

def join_with_commas(obj_list):
    """Takes a list of objects and returns their string representations,
    separated by commas and with 'and' between the penultimate and final items
    For example, for a list of fruit objects:
    [<Fruit: apples>, <Fruit: oranges>, <Fruit: pears>] -> 'apples, oranges and pears'
    """
    if not obj_list:
        return ""
    l=len(obj_list)
    if l==1:
        return u"%s" % obj_list[0]
    else:    
        return ", ".join(str(obj) for obj in obj_list[:l-1]) \
                + " and " + str(obj_list[l-1])

To use it in the template: {{ fruits|join_with_commas }}

All of the answers here fail one or more of the following:

• They rewrite something (poorly!) that's in the standard template library (ack, top answer!)
• They don't use and for the last item.
• They lack a serial (oxford) comma.
• They use negative indexing, which won't work for django querysets.
• They don't usually handle string sanitation properly.

Here's my entry into this canon. First, the tests:

class TestTextFilters(TestCase):

    def test_oxford_zero_items(self):
        self.assertEqual(oxford_comma([]), '')

    def test_oxford_one_item(self):
        self.assertEqual(oxford_comma(['a']), 'a')

    def test_oxford_two_items(self):
        self.assertEqual(oxford_comma(['a', 'b']), 'a and b')

    def test_oxford_three_items(self):
        self.assertEqual(oxford_comma(['a', 'b', 'c']), 'a, b, and c')

And now the code. Yes, it gets a bit messy, but you'll see that it doesn't use negative indexing:

from django.utils.encoding import force_text
from django.utils.html import conditional_escape
from django.utils.safestring import mark_safe

@register.filter(is_safe=True, needs_autoescape=True)
def oxford_comma(l, autoescape=True):
    """Join together items in a list, separating them with commas or ', and'"""
    l = map(force_text, l)
    if autoescape:
        l = map(conditional_escape, l)

    num_items = len(l)
    if num_items == 0:
        s = ''
    elif num_items == 1:
        s = l[0]
    elif num_items == 2:
        s = l[0] + ' and ' + l[1]
    elif num_items > 2:
        for i, item in enumerate(l):
            if i == 0:
                # First item
                s = item
            elif i == (num_items - 1):
                # Last item.
                s += ', and ' + item
            else:
                # Items in the middle
                s += ', ' + item

    return mark_safe(s)

You can use this in a django template with:

{% load my_filters %}
{{ items|oxford_comma }}

I would simply use ', '.join(['apples', 'oranges', 'pears']) before sending it to the template as a context data.

UPDATE:

data = ['apples', 'oranges', 'pears']
print(', '.join(data[0:-1]) + ' and ' + data[-1])

You will get apples, oranges and pears output.

I think the simplest solution might be:

@register.filter
def comma_list(p_values: Iterable[str]) -> List[str]:
    values = list(p_values)
    if len(values) > 1:
        values[-1] = u'and %s' % values[-1]
    if len(values) > 2:
        return u', '.join(values)
    return u' '.join(values)