π Modules and Bundles#
Fetchez comes builtin with 70+ different modules to access geospatial data from various remote apis and local file-systems.
Data Modules#
Fetchez includes a powerful Module System that allows you to access various geospatial data sources locally or from around the world.
Modules come with their own arguments to set different data types, modify outputs, set credentials, etc. Hooks can be used to modify modules before, during or after fetching; allowing for full ETL processing workflows.
Module Bundles#
Bundles are YAML configuration files that define a group of Modules and their arguments and hooks. You can make your own, or use a pre-configured Bundle from fetchez or itβs extensions.
Example#
Define your bundle
Put this in your ~/.fetchez/bundles/ plugin folder
name: grav_and_bath
description: >
Some fast bathymetry data sources.
modules:
- module: margrav
args:
weight: .01
- module: nos_hydro
args:
datatype: "xyz"
weight: .35
hooks:
- name: unzip
- name: set_datatype
args:
data_type: "nos_xyz"
- module: charts
args:
weight: .15
hooks:
- name: unzip
- name: filename_filter
args:
match: ".000"
stage: "file"
- name: set_datatype
args:
data_type: "charts_000"
Use the bundle in a recipe
Add the bundle to your recipe!
project:
name: "my_harbor"
region: [-120.5, -120.0, 34.0, 34.5]
modules:
- bundle: grav_and_bath
args: {weight: 1.0}
"""
Extending Bunldes (Plugins and Extensions)#
Fetchez is generic. If you are building a custom tool and want to bundle your own modules, you can register your own bundles either in your project or in the .fetchez configuration directory and they will be discoverable with the fetchez.registry.BundleRegistry
To create an extension where your bundles can be installed and used by fetchez, make a directory called βbundlesβ in your project; add any YAML module bundles to that directory and register them with fetchez in your pyproject.toml:
[project.entry-points."fetchez.bundles"]
my_project_bundles = "my_project.bundles"