A light-weight, versatile XYZ tile server, built with Flask and Rasterio 🌍
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
Nicklas Boserup 9bd58730f5
Parallel raster optimisation (#228)
2 weeks ago
.github/workflows replace master with main 4 weeks ago
docs fix docs (#231) 4 weeks ago
terracotta Parallel raster optimisation (#228) 2 weeks ago
tests Parallel raster optimisation (#228) 2 weeks ago
.coveragerc 💄 3 years ago
.gitattributes fix failing cmaps npy files (#124) 3 years ago
.gitignore remove gitignore conf for flask_api template 5 months ago
.travis.yml use travis-ci.com upload token instead 3 months ago
LICENSE Initial commit 4 years ago
README.md replace master with main 4 weeks ago
environment.yml add explicit pip dep to environment.yml 2 years ago
readthedocs.yml convert readme to rst 3 years ago
setup.cfg Add option to skip existing datasets on ingest; update to recent dependencies (#192) 7 months ago
setup.py Parallel raster optimisation (#228) 2 weeks ago
zappa_requirements.txt wip on docs 3 years ago
zappa_settings.toml.in adjust zappa_settings template to new settings 1 year ago
zappa_settings_callback.py add mysql as valid remote driver for zappa 3 years ago

README.md

Build Status Documentation Status codecov GitHub release PyPI release License Python versions

Logo

Terracotta is a pure Python tile server that runs as a WSGI app on a dedicated webserver or as a serverless app on AWS Lambda. It is built on a modern Python 3.6 stack, powered by awesome open-source software such as Flask, Zappa, and Rasterio.

Read the docs | Explore the API | Satlas, powered by Terracotta

Why Terracotta?

  • It is trivial to get going. Got a folder full of cloud-optimized GeoTiffs in different projections you want to have a look at in your browser? terracotta serve -r {name}.tif and terracotta connect localhost:5000 get you there.
  • We make minimal assumptions about your data, so you stay in charge. Keep using the tools you know and love to create and organize your data, Terracotta serves it exactly as it is.
  • Serverless deployment is a first-priority use case, so you don’t have to worry about maintaining or scaling your architecture.
  • Terracotta instances are self-documenting. Everything the frontend needs to know about your data is accessible from only a handful of API endpoints.

The Terracotta workflow

1. Optimize raster files

$ ls -lh
total 1.4G
-rw-r--r-- 1 dimh 1049089 231M Aug 29 16:45 S2A_20160724_135032_27XVB_B02.tif
-rw-r--r-- 1 dimh 1049089 231M Aug 29 16:45 S2A_20160724_135032_27XVB_B03.tif
-rw-r--r-- 1 dimh 1049089 231M Aug 29 16:46 S2A_20160724_135032_27XVB_B04.tif
-rw-r--r-- 1 dimh 1049089 231M Aug 29 16:56 S2A_20170831_171901_25XEL_B02.tif
-rw-r--r-- 1 dimh 1049089 231M Aug 29 16:57 S2A_20170831_171901_25XEL_B03.tif
-rw-r--r-- 1 dimh 1049089 231M Aug 29 16:57 S2A_20170831_171901_25XEL_B04.tif

$ terracotta optimize-rasters *.tif -o optimized/

Optimizing rasters: 100%|██████████████████████████| [05:16<00:00, file=S2A_20170831_...25XEL_B04.tif]

2. Create a database from file name pattern

$ terracotta ingest optimized/S2A_{date}_{}_{tile}_{band}.tif -o greenland.sqlite
Ingesting raster files: 100%|███████████████████████████████████████████| 6/6 [00:49<00:00,  8.54s/it]

3. Serve it up

$ terracotta serve -d greenland.sqlite
 * Serving Flask app "terracotta.server" (lazy loading)
 * Environment: production
   WARNING: Do not use the development server in a production environment.
   Use a production WSGI server instead.
 * Debug mode: off
 * Running on http://localhost:5000/ (Press CTRL+C to quit)

4. Explore the running server

Manually

You can use any HTTP-capable client, such as curl.

$ curl localhost:5000/datasets?tile=25XEL
{"page":0,"limit":100,"datasets":[{"date":"20170831","tile":"25XEL","band":"B02"},{"date":"20170831","tile":"25XEL","band":"B03"},{"date":"20170831","tile":"25XEL","band":"B04"}]}

Modern browsers (e.g. Chrome or Firefox) will render the JSON as a tree.

Interactively

Terracotta also includes a web client. You can start the client (assuming the server is running at http://localhost:5000) using

$ terracotta connect localhost:5000
 * Serving Flask app "terracotta.client" (lazy loading)
 * Environment: production
   WARNING: Do not use the development server in a production environment.
   Use a production WSGI server instead.
 * Debug mode: off
 * Running on http://127.0.0.1:5100/ (Press CTRL+C to quit)

Then open the client page (http://127.0.0.1:5100/ in this case) in your browser.

preview

Development

We gladly accept bug reports and pull requests via GitHub. For your code to be useful, make sure that it is covered by tests and that it satisfies our linting practices (via mypy and flake8).

To run the tests, just install the necessary dependencies via

$ pip install -e .[test]

Then, you can run

$ pytest

from the root of the repository.