A Python library for reading FlatGeobuf. Ported from the official TypeScript implementation.
- Minimal dependencies
- Simple API
- Supports cloud-optimized bounding box filtering
- Works on JupyterLite (Pyodide)
pip install flatgeobuf
import flatgeobuf as fgb
# All features
with open("example.fgb", "rb") as f:
data = fgb.load(f)
# ...or features within a bounding box
with open("example.fgb", "rb") as f:
data = fgb.load(f, bbox=(-26.5699, 63.1191, -12.1087, 67.0137))
print(data)
# { "type": "FeatureCollection", "features": [...] }
import flatgeobuf as fgb
# All features
data = fgb.load_http("https://raw.githubusercontent.com/flatgeobuf/flatgeobuf/master/test/data/countries.fgb")
# ...or features within a bounding box
data = fgb.load_http(
"https://raw.githubusercontent.com/flatgeobuf/flatgeobuf/master/test/data/countries.fgb",
bbox=(-26.5699, 63.1191, -12.1087, 67.0137)
)
print(data)
# { "type": "FeatureCollection", "features": [...] }
NOTE: At the moment, load_http_async()
is not truly asynchronous.
import flatgeobuf as fgb
# All features
data = await fgb.load_http_async("https://raw.githubusercontent.com/flatgeobuf/flatgeobuf/master/test/data/countries.fgb")
# ...or features within a bounding box
data = await fgb.load_http_async(
"https://raw.githubusercontent.com/flatgeobuf/flatgeobuf/master/test/data/countries.fgb",
bbox=(-26.5699, 63.1191, -12.1087, 67.0137)
)
print(data)
# { "type": "FeatureCollection", "features": [...] }
import flatgeobuf as fgb
# All features
with open("example.fgb", "rb") as f:
reader = fgb.Reader(f)
for feature in reader:
print(feature)
# { "type": "Feature", "properties": {...}, "geometry": {...} }
# ...or features within a bounding box
with open("example.fgb", "rb") as f:
reader = fgb.Reader(f, bbox=(-26.5699, 63.1191, -12.1087, 67.0137))
for feature in reader:
print(feature)
# { "type": "Feature", "properties": {...}, "geometry": {...} }
import flatgeobuf as fgb
# All features
reader = fgb.HTTPReader("https://raw.githubusercontent.com/flatgeobuf/flatgeobuf/master/test/data/countries.fgb")
for feature in reader:
print(feature)
# { "type": "Feature", "properties": {...}, "geometry": {...} }
# ...or features within a bounding box
reader = fgb.HTTPReader(
"https://raw.githubusercontent.com/flatgeobuf/flatgeobuf/master/test/data/countries.fgb",
bbox=(-26.5699, 63.1191, -12.1087, 67.0137)
)
for feature in reader:
print(feature)
# { "type": "Feature", "properties": {...}, "geometry": {...} }
NOTE: At the moment, HTTPReader
is not truly asynchronous.
import flatgeobuf as fgb
# All features
reader = fgb.HTTPReader("https://raw.githubusercontent.com/flatgeobuf/flatgeobuf/master/test/data/countries.fgb")
async for feature in reader:
print(feature)
# { "type": "Feature", "properties": {...}, "geometry": {...} }
# ...or features within a bounding box
reader = fgb.HTTPReader(
"https://raw.githubusercontent.com/flatgeobuf/flatgeobuf/master/test/data/countries.fgb",
bbox=(-26.5699, 63.1191, -12.1087, 67.0137)
)
async for feature in reader:
print(feature)
# { "type": "Feature", "properties": {...}, "geometry": {...} }
1. Install flatgeobuf
on JupyterLite:
%pip install flatgeobuf
# ...or
import micropip
await micropip.install("flatgeobuf")
2. Enable HTTP requests in Pyodide via pyodide_http
import pyodide_http
pyodide_http.patch_all()
3. Run as usual!
import flatgeobuf as fgb
data = fgb.load_http(
"https://raw.githubusercontent.com/flatgeobuf/flatgeobuf/master/test/data/countries.fgb",
bbox=(-26.5699, 63.1191, -12.1087, 67.0137)
)
print(data)
# { "type": "FeatureCollection", "features": [...] }
- Read FlatGeobuf
- Read top-level (
FeatureCollection
) properties
- Read top-level (
- Write FlatGeobuf
- Deploy JuptyerLite examples
- Rewrite some parts in Rust? (parcked R-tree, geometry intersection)
MIT