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
Team up Django and Web mapping - DjangoCon Euro...
Search
Makina Corpus
May 20, 2014
Technology
3
870
Team up Django and Web mapping - DjangoCon Europe 2014
● Web mapping should be simple
● Google Maps should become unusual
● Demystify cartography !
Makina Corpus
May 20, 2014
Tweet
Share
More Decks by Makina Corpus
See All by Makina Corpus
Publier vos données sur le Web - Forum TIC de l'ATEN 2014
makinacorpus
0
740
Créez votre propre fond de plan à partir de données OSM en utilisant TileMill
makinacorpus
0
130
Petit déjeuner "Les bases de la cartographie sur le Web"
makinacorpus
0
420
Petit déjeuner "Développer sur le cloud, ou comment tout construire à partir de rien" le 11 février - Toulouse
makinacorpus
0
270
CoDe, le programme de développement d'applications mobiles de Makina Corpus
makinacorpus
0
110
Petit déjeuner "Alternatives libres à GoogleMaps" du 11 février 2014 - Nantes - Sylvain Beorchia
makinacorpus
0
660
Petit déjeuner "Les nouveautés de la cartographie en ligne" du 12 décembre
makinacorpus
1
390
Tests carto avec Mocha
makinacorpus
0
810
Alternatives libres à Google Maps
makinacorpus
2
3k
Other Decks in Technology
See All in Technology
Findy Freelance 利用シーン別AI活用例
ness
0
390
Lambda management with ecspresso and Terraform
ijin
2
150
20250807_Kiroと私の反省会
riz3f7
0
190
S3 Glacier のデータを Athena からクエリしようとしたらどうなるのか/try-to-query-s3-glacier-from-athena
emiki
0
210
2時間で300+テーブルをデータ基盤に連携するためのAI活用 / FukuokaDataEngineer
sansan_randd
0
140
マルチモーダル基盤モデルに基づく動画と音の解析技術
lycorptech_jp
PRO
5
570
【OptimizationNight】数理最適化のラストワンマイルとしてのUIUX
brainpadpr
1
430
VLMサービスを用いた請求書データ化検証 / SaaSxML_Session_1
sansan_randd
0
240
反脆弱性(アンチフラジャイル)とデータ基盤構築
cuebic9bic
3
170
アカデミーキャンプ 2025 SuuuuuuMMeR「燃えろ!!ロボコン」 / Academy Camp 2025 SuuuuuuMMeR "Burn the Spirit, Robocon!!" DAY 1
ks91
PRO
0
130
LLMで構造化出力の成功率をグンと上げる方法
keisuketakiguchi
0
650
GMOペパボのデータ基盤とデータ活用の現在地 / Current State of GMO Pepabo's Data Infrastructure and Data Utilization
zaimy
3
210
Featured
See All Featured
KATA
mclloyd
32
14k
The Art of Programming - Codeland 2020
erikaheidi
54
13k
Documentation Writing (for coders)
carmenintech
73
5k
It's Worth the Effort
3n
185
28k
Mobile First: as difficult as doing things right
swwweet
223
9.9k
XXLCSS - How to scale CSS and keep your sanity
sugarenia
248
1.3M
The Success of Rails: Ensuring Growth for the Next 100 Years
eileencodes
46
7.6k
I Don’t Have Time: Getting Over the Fear to Launch Your Podcast
jcasabona
33
2.4k
Why Our Code Smells
bkeepers
PRO
337
57k
A designer walks into a library…
pauljervisheath
207
24k
RailsConf & Balkan Ruby 2019: The Past, Present, and Future of Rails at GitHub
eileencodes
139
34k
Product Roadmaps are Hard
iamctodd
PRO
54
11k
Transcript
Team up Django and Web mapping DjangoCon Europe 2014 Mathieu
Leplatre @leplatrem www.makina-corpus.com
Main goals... • Web mapping should be simple • Google
Maps should become unusual • Demystify cartography !
Main focus... • Fundamentals of cartography (projections, PostGIS) • Web
mapping (GeoJSON, Leaflet) • Django toolbox (a.k.a. GeoDjango)
Fundamentals of cartography
Coordinates : position on Earth • Longitude (x) – Latitude
(y) • Decimal degrees (-180 +180, -90 +90) → → • GPS
Geodesy : shape of the Earth Ellipsoid (GPS, WGS 84)
Illustration: http://mapschool.io
Projections • Equations (lat/lng pixels) ↔ • Representation on a
plane ( compromises) → • Spatial reference transformation Illustration: http://mapschool.io
Spatial References • Coordinate system (cartesian) • Units (degrees, meters)
• Main axis (equator, Greenwich) • Ellipsoid, geoid (WGS 84) • Conical, cylindrical, conformal, ... • … Store WGS 84 → (GPS, srid=4326) Display →Mercator (Google Maps, srid=3857)
Vector data • Point (x, y, z) • Line (ordered
list of points) • Polygon (enveloppe + holes) Illustration: http://mapschool.io
Raster data • ~ Images • Map backgrounds (satellite, plan,
…) • Atimetry data (DEM) Illustration: http://mapschool.io
A PostGIS database • Column types (Point, LineString, Polygon, …Raster...)
• Geographic functions (distance, area, extents ...) • Spatial Indexes (rectangles trees...) $ sudo apt-get install postgis $ psql -d mydatabase > CREATE EXTENSION postgis;
Examples (1/2) CREATE TABLE island ( label VARCHAR(128), code INTEGER,
geom geometry(Polygon, 4326) )
Examples (1/2) CREATE TABLE island ( label VARCHAR(128), code INTEGER,
geom geometry(Polygon, 4326) ) Usual table Usual columns Vector geometry column Geometry type Spatial Reference
Examples (2/2) SELECT room.id FROM room, island WHERE ST_Intersects(room.geom, island.geom)
Spatial join INSERT INTO room (geom) VALUES ( 'SRID=4326;POINT(3.12, 43.1)'::geometry ) EWKT format PostgreSQL cast
Web mapping
Web map anatomy... • JavaScript + DOM • Initialization of
a <div> • « Layers »
Web map anatomy... • JavaScript + DOM • Initialization of
a <div> • « Layers » • <img> (Backgrounds) • Vector SVG → • Lat/Long pixels → (CSS) • DOM Events (interactions)
Leaflet example <script src="leaflet.js"></script> <link rel="stylesheet" href="leaflet.css" /> <div id="simplemap"></div>
<script> var map = L.map('simplemap') .fitWorld(); L.tileLayer('http://tile.osm.org/{z}/{x}/{y}.png') .addTo(map); L.marker([43.07, -5.78]) .addTo(map); </script> http://leafletjs.com
None
See also... D3 • SVG • Cool dataviz • Exotic
map projections OpenLayers 3 • Canvas • GIS-oriented (advanced use-cases)
Formats (1/2) • Raster images (PNG ou JPG) • z/x/y.png
• Projection 3857 (Mercator, Google Maps) →Cold data (tile provider) … → or warm (rendering engine)
Formats (2/2) • GeoJSON for vector data • Projection 4326
(GPS, WGS84) • Interactive ? • Volume ? • Frequency ? →Dynamic (from database) SELECT ST_AsGeoJSON(geom) FROM ...
Nothing new... Web Server Browser PostGIS Django Rendering Engine Vector
overlay Raster overlay Raster background Images z/x/y.png GeoJSON z/x/y.png, UTFGrid, ...
GeoDjango
Django Ecosystem for Cartography from django.contrib.gis import ... • Models
fields (syncdb) • GeoQuerySet (spatial queries) • Form fields (Django 1.6) • System libraries deps (GEOS, GDAL) • Widgets & AdminSite (OpenLayers 2)
Examples (1/2) from django.contrib.gis.db import ( models as gismodels )
class Island(gismodels.Model): label = models.CharField(max_length=128), code = models.IntegerField(), geom = gismodels.PolygonField(srid=4326) objects = gismodels.GeoManager()
Examples (2/2) from django.contrib.gis.geos import Polygon embiez = Island(label='Embiez', geom=Polygon(((0.34,
0.44), (0.36, 0.42), …))) ~ from django.contrib.gis.geos import fromstr myroom = fromstr('SRID=4326;POINT(3.12, 43.1)') Island.objects.filter(geom__intersects=myroom)
• Views (Class-based) • Generalization & approximation (less details and
decimals) • Serialization (dumpdata, loaddata) • Model fields (text, no GIS !) • Vector tiles (huge datasets) django-geojson from djgeojson.views import GeoJSONLayerView urlpatterns = patterns('', url(r'^buildings.geojson$', GeoJSONLayerView.as_view(model=Building)) )
django-leaflet • Quick-start kit {% leafletmap "djangocon" %} • Assets
(collecstatic) • Easy plugins integration • Global configuration (settings.py, plugins, ...) • Leaflet projections machinery (reprojection)
django-leaflet (view) {% load leaflet_tags %} ... {% leaflet_js %}
{% leaflet_css %} </head> <body> {% leafletmap "buildingmaps" callback="initMap" %} <script type="text/javascript"> function initMap(map) { // Ajax Load $.getJSON("{% url app:layer %}", function (data) { // Add GeoJSON to the map L.geoJSON(data).addTo(map); }); } GeoJSON view
LAYER
django-leaflet (edit) Forms • Fields (Django 1.6+ API) • Widgets
(hidden <textarea> + Leaflet.draw) class BuildingForm(forms.ModelForm): class Meta: model = Building widgets = {'geom': LeafletWidget()} fields = ('code', 'geom') • Leaflet AdminSite admin.site.register(Building, LeafletGeoAdmin)
SCREENSHOT FORM
Conclusion
Anyone can do Web mapping... Cartography is simple. • Just
some special data types • Follows Django conventions • Few lines of code Illustration: Chef Gusteau, Pixar Inc.
… with appropriate tools and strategies ! Cartography is hard.
• Performance (Web) • Volume (precision) • Data refresh rate (caching) • User stories too far from data (modeling) Illustration: Anton Ego, Pixar Inc.
Makina Corpus - Free Software | Cartography | Mobile Questions
? ...and answers ! PostGIS - Leaflet – JavaScript http://makina-corpus.com