Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
Hunting For Treasure In Django
Search
Seb
May 28, 2015
Technology
0
180
Hunting For Treasure In Django
Seb
May 28, 2015
Tweet
Share
More Decks by Seb
See All by Seb
Double Click - Continue Building Better CLIs
elbaschid
0
460
I Can Be A Speaker, So Can You
elbaschid
0
320
Click - PyCaribbean 2017 - Puerto Rico
elbaschid
0
440
Conferencing - Engineering Meeting
elbaschid
1
45
Show & Tell - PyCon US 2016 Summary
elbaschid
1
110
Click: A Pleasure To Write, A Pleasure To Use
elbaschid
0
650
Hunting for Treasure in Django
elbaschid
1
700
Moby & The Beanstalk
elbaschid
1
510
Docker In Production - A War Story
elbaschid
1
320
Other Decks in Technology
See All in Technology
業務の煩悩を祓うAI活用術108選 / AI 108 Usages
smartbank
9
18k
Agentic AIが変革するAWSの開発・運用・セキュリティ ~Frontier Agentsを試してみた~ / Agentic AI transforms AWS development, operations, and security I tried Frontier Agents
yuj1osm
0
160
TED_modeki_共創ラボ_20251203.pdf
iotcomjpadmin
0
170
意外と知らない状態遷移テストの世界
nihonbuson
PRO
1
350
テストセンター受験、オンライン受験、どっちなんだい?
yama3133
0
200
ペアーズにおけるAIエージェント 基盤とText to SQLツールの紹介
hisamouna
2
1.9k
SES向け、生成AI時代におけるエンジニアリングとセキュリティ
longbowxxx
0
240
202512_AIoT.pdf
iotcomjpadmin
0
160
Cloud WAN MCP Serverから考える新しいネットワーク運用 / 20251228 Masaki Okuda
shift_evolve
PRO
0
130
AgentCoreとStrandsで社内d払いナレッジボットを作った話
motojimayu
1
1.2k
半年で、AIゼロ知識から AI中心開発組織の変革担当に至るまで
rfdnxbro
0
160
Snowflake Industry Days 2025 Nowcast
takumimukaiyama
0
150
Featured
See All Featured
Design of three-dimensional binary manipulators for pick-and-place task avoiding obstacles (IECON2024)
konakalab
0
320
Building AI with AI
inesmontani
PRO
1
590
How to make the Groovebox
asonas
2
1.9k
Creating an realtime collaboration tool: Agile Flush - .NET Oxford
marcduiker
35
2.3k
Leveraging Curiosity to Care for An Aging Population
cassininazir
1
140
Leadership Guide Workshop - DevTernity 2021
reverentgeek
1
170
Utilizing Notion as your number one productivity tool
mfonobong
2
190
Design and Strategy: How to Deal with People Who Don’t "Get" Design
morganepeng
132
19k
The Straight Up "How To Draw Better" Workshop
denniskardys
239
140k
Evolving SEO for Evolving Search Engines
ryanjones
0
86
YesSQL, Process and Tooling at Scale
rocio
174
15k
個人開発の失敗を避けるイケてる考え方 / tips for indie hackers
panda_program
122
21k
Transcript
Hunting for Treasure in Django Sebastian Vetter @elbaschid
Who Am I?
Sebastian • Django & Python Developer • Backend Engineer @
Mobify • github/twitter: elbaschid
What's The Treasure?
Awesome Django Features • Forms, • Views, • Models, •
the ORM, or • other commonly used APIs.
But They Are Boring
Real Treasure
What Does That Mean? • Useful pieces of Django code.
• Considered public API. • Documentation is available (sort of). • Mainly used within Django itself.
My Hunting Strategy • Digging through the Django source. •
Hanging out with Funkybob. • Learning from other great people.
What I'll Do • Show a few "hidden" treasures. •
Explain what they do. • Look at examples.
cached_property
Where is it useful? • Time or compute heavy properties
on a class. • Synchronous calls to remote servers. • Used more than once, e.g. code & template.
What does it do? • It's a decorator. • Caches
the return value. • Lives as long as the instance.
It looks like this class MyObject(object): @cached_property def compute_heavy_method(self): ...
return result
Imagine A Color API class Color(object): def __init__(self, hex): self.hex
= hex def _request_colour_name(self, hex): print "Requesting #{}".format(hex) rsp = requests.get(API_ENDPOINT.format(hex)) return rsp.json()[0].get("title") @property def name(self): return self._request_colour_name(self.hex)
Here's the problem • Using the name attribute will call
the API • Every time!
Here's the problem >>> c = Color('ffffff') >>> c.name Requesting
#ffffff white >>> c.name Requesting #ffffff white
Possible solution @property def name(self): if self._name is None: self._name
= self._request_colour_name(self.hex) return self._name
Or you can use cached_property from django.utils.functional import cached_property @cached_property
def name(self): return self._request_colour_name(self.hex)
Using the cached property >>> c = Color('ffffff') >>> c.name
Requesting #ffffff white >>> c.name white
Isn't That Great
All you Need To Know from django.utils.functional import cached_property •
Only cached for the lifetime of the instance. • Be careful with querysets. • Django docs • Source
import_string
Where is it useful? • Make a class or function
configurable. • Allow loading class/function from string.
What does it do? • Takes dotted path to a
class or function. • Loads it. • Returns the class or function object.
It looks like this from django.utils.module_loading import import_string get_func =
import_string('requests.get') print get_func # <function requests.api.get> get_func('https://google.ca') # <Response [200]>
# settings.py UPLOAD_VALIDATION_PIPELINE = [ 'my_project.uploads.validators.is_tarball', 'my_project.uploads.validators.has_readme_file', 'my_project.uploads.validators.has_no_!']
All you Need To Know from django.utils.module_loading import import_string •
Imports a class or function from a dotted path. • Django docs • Source
lazy and lazy_property
Where is it useful? • Accessing settings at parse time,
e.g. class attributes. • Translating strings outside of a view. • Translations in the settings module.
Here's a problem class UserSignupView(CreateView): ... success_url = reverse('signup-confirmed')
How can we fix it? from django.utils.functional import lazy class
UserSignupView(CreateView): ... success_url = lazy(reverse('signup-confirmed'), unicode)
Lazy Django • The Settings object is lazy. • Several
helpers have lazy siblings: • reverse_lazy • ugettext_lazy • Not sure what lazy_property is useful for.
All you Need To Know from django.utils.functional import lazy from
django.utils.functional import SimpleLazyObject • Imports a class or function from a dotted path. • Django docs • Source
RequestFactory
Where Is It Useful? • Testing request related code. •
Mocking will be too much work. • Using the test client doesn't make sense.
Create GET Request from django.test import RequestFactory request = RequestFactory().get('/some/path')
# with a query string query_params = {"token": "secret-token"} request = RequestFactory().get('/some/path', data=query_params)
Create POST Request from django.test import RequestFactory post_params = {'username':'testuser',
'password':'supersecret'} request = RequestFactory().post('/login', data=post_params)
All you Need To Know from django.test import RequestFactory •
Creates a fake request for given URL. • Can handle all HTTP methods. • Will save you some mocking work. • Django docs • Source
The Treasure Is Yours
Thanks! Questions? • www.roadsi.de • @elbaschid • github.com/elbaschid Slides: https://speakerdeck.com/elbaschid