Login

Unique Slugify

Author:
SmileyChris
Posted:
April 8, 2008
Language:
Python
Version:
.96
Tags:
slug
Score:
10 (after 10 ratings)

Automatically create a unique slug for a model.

Note that you don't need to do obj.slug = ... since this method updates the instance's slug field directly. All you usually need is: unique_slugify(obj, obj.title)

A frequent usage pattern is to override the save method of a model and call unique_slugify before the super(...).save() call.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
import re
from django.template.defaultfilters import slugify


def unique_slugify(instance, value, slug_field_name='slug', queryset=None,
                   slug_separator='-'):
    """
    Calculates and stores a unique slug of ``value`` for an instance.

    ``slug_field_name`` should be a string matching the name of the field to
    store the slug in (and the field to check against for uniqueness).

    ``queryset`` usually doesn't need to be explicitly provided - it'll default
    to using the ``.all()`` queryset from the model's default manager.
    """
    slug_field = instance._meta.get_field(slug_field_name)

    slug = getattr(instance, slug_field.attname)
    slug_len = slug_field.max_length

    # Sort out the initial slug, limiting its length if necessary.
    slug = slugify(value)
    if slug_len:
        slug = slug[:slug_len]
    slug = _slug_strip(slug, slug_separator)
    original_slug = slug

    # Create the queryset if one wasn't explicitly provided and exclude the
    # current instance from the queryset.
    if queryset is None:
        queryset = instance.__class__._default_manager.all()
    if instance.pk:
        queryset = queryset.exclude(pk=instance.pk)

    # Find a unique slug. If one matches, at '-2' to the end and try again
    # (then '-3', etc).
    next = 2
    while not slug or queryset.filter(**{slug_field_name: slug}):
        slug = original_slug
        end = '%s%s' % (slug_separator, next)
        if slug_len and len(slug) + len(end) > slug_len:
            slug = slug[:slug_len-len(end)]
            slug = _slug_strip(slug, slug_separator)
        slug = '%s%s' % (slug, end)
        next += 1

    setattr(instance, slug_field.attname, slug)


def _slug_strip(value, separator='-'):
    """
    Cleans up a slug by removing slug separator characters that occur at the
    beginning or end of a slug.

    If an alternate separator is used, it will also replace any instances of
    the default '-' separator with the new separator.
    """
    separator = separator or ''
    if separator == '-' or not separator:
        re_sep = '-'
    else:
        re_sep = '(?:-|%s)' % re.escape(separator)
    # Remove multiple instances and if an alternate separator is provided,
    # replace the default '-' separator.
    if separator != re_sep:
        value = re.sub('%s+' % re_sep, separator, value)
    # Remove separator from the beginning and end of the slug.
    if separator:
        if separator != '-':
            re_sep = re.escape(separator)
        value = re.sub(r'^%s+|%s+$' % (re_sep, re_sep), '', value)
    return value

More like this

  1. Automate unique slugs by taojian 7 years, 8 months ago
  2. Automate unique slug (again) by davidwtbuxton 7 years, 3 months ago
  3. Hierarchical page slugs by baumer1122 8 years ago
  4. YAAS (Yet Another Auto Slug) by carljm 7 years, 3 months ago
  5. uuid model field by newspire 6 years, 8 months ago

Comments

SmileyChris (on April 17, 2008):

Looks pretty similar - my way respects maximum length of the slug field, respects the _default_manager rather than assuming "objects".

Oh, and it also handles the case of changing a slug for an existing instance.

I probably should change mine to using a single query, but since it's only at save time for where I use this method, I didn't need it.

#

SmileyChris (on August 25, 2008):

I just updated the snippet to also allow for changing the slug separator. I needed it to auto-generate a contrib.auth User username which was still editable in admin (doesn't allow the - character so I use slug_separator='_').

#

roderikk (on October 16, 2008):

Hi,

This is an interesting snippet. Does it still work with Django 1.0? Are there any thoughts in post-1.0 to update the slugfield and/or slugify functions to include this functionality?

I was also wondering, if I want to make a unique slug only for a given day would I enter something like this:

self.slug = unique_slugify(self,self.title,queryset=self.__class__._default_manager.filter(datetime_added=self.datetime_added))

#

seanos (on March 28, 2009):

Thanks for this.

One bug I found:

line 29 should be changed to:

if queryset is None:

otherwise if the argument queryset is empty, it will be re-populated with all rows from the model, leading to slug being modified when it should of been left alone.

#

SmileyChris (on June 24, 2009):

Good point, seanos. Updated.

#

eddified (on July 17, 2009):

Good snippet. Maybe add a param that lets the user specify a custom exclude dict in cases where the model doesn't have a pk. But then again, a workaround would be to just specify a query set object that has what you want excluded, excluded. What's this about class.default_manager vs objects? When do I use one vs the other in my own code?

#

SmileyChris (on July 23, 2009):

eddified: Yep, for fringe cases like that, someone can just explicitly provide the queryset object.

Regarding when to use ._default_manager - you should use it when referencing potentially third-party model classes since you can't know if they have a manager named "objects".

#

Thrasher (on March 16, 2012):

Amazing snippet, it really helped me!

#

jdango (on August 22, 2014):

What's the point of adding .exclude(pk=instance.pk) to queryset. If instance.pk is not None, it means that it's an update operation and the field already has a slug. Shouldn't you just return in that case?

#

Please login first before commenting.