gedih3.raster.export

Raster Export Module

This module provides utilities for exporting raster data to various formats, with support for GeoTIFF compression, tiling, and batch operations.

Attributes

logger

Functions

export_raster(→ str)	Export xarray Dataset to GeoTIFF file.
export_raster_partition(→ str)	Export raster partition(s) to file(s).
rasterize_and_export_partitions(→ List[str])	Rasterize and export GeoDataFrame partitions to individual files.
build_vrt(tif_files, vrt_path)	Build a GDAL VRT file mosaicking a list of GeoTIFF tiles.
build_vrt_xml(tif_files, vrt_path)	Write a VRT mosaic as XML using rasterio only — no osgeo bindings.
build_vrt_safe(tif_files, vrt_path)	Build a VRT mosaic, downgrading any failure to a warning.
merge_and_export_rasters(→ str)	Rasterize all partitions and merge into a single output file.
compute_raster_stats(→ Dict[str, Dict[str, float]])	Compute basic statistics for each variable in a raster dataset.

Module Contents

gedih3.raster.export.logger

gedih3.raster.export.export_raster(xras: xarray.Dataset, output_path: str, compress: str = 'LZW', tiled: bool = True, blocksize: int = 256, bigtiff: bool = True) → str

Export xarray Dataset to GeoTIFF file.

Parameters:

xrasxr.Dataset

Raster dataset to export

output_pathstr

Output file path

compressstr

Compression method (‘LZW’, ‘ZSTD’, ‘DEFLATE’, ‘NONE’)

tiledbool

Use tiled output format

blocksizeint

Tile block size in pixels

bigtiffbool

Use BigTIFF format for large files

Returns:

str

Output file path

gedih3.raster.export.export_raster_partition(data: pandas.Series | xarray.Dataset, output_dir: str, fmt: str = 'tif', compress: str = 'LZW', partition_id_attr: str | None = None) → str

Export raster partition(s) to file(s).

This function handles the case where a Series may contain multiple rasters from different spatial tiles. Each raster is exported to its own file based on its tile ID attribute.

Parameters:

datapd.Series or xr.Dataset

Raster data (Series of DataArrays/Datasets or single Dataset)

output_dirstr

Output directory

fmtstr

Output format (‘tif’, ‘nc’)

compressstr

Compression method for GeoTIFF

partition_id_attrstr, optional

Attribute name containing partition ID for filename

Returns:

str

Output file path(s), comma-separated if multiple files written

gedih3.raster.export.rasterize_and_export_partitions(gdf: geopandas.GeoDataFrame | dask.dataframe.DataFrame, output_dir: str, rasterize_func, columns: List[str] | None = None, fmt: str = 'tif', compress: str = 'LZW', show_progress: bool = True, **rasterize_kwargs) → List[str]

Rasterize and export GeoDataFrame partitions to individual files.

Parameters:

gdfGeoDataFrame or dask GeoDataFrame

Input spatially-indexed data

output_dirstr

Output directory for raster files

rasterize_funccallable

Function to rasterize each partition (e.g., rasterize_h3_partition)

columnslist of str, optional

Columns to include in rasterization

fmtstr

Output format (‘tif’, ‘nc’)

compressstr

Compression method for GeoTIFF

show_progressbool

Show Dask progress bar

Returns:

list of str

Paths to output files

gedih3.raster.export.build_vrt(tif_files, vrt_path)

Build a GDAL VRT file mosaicking a list of GeoTIFF tiles.

Uses the osgeo.gdal bindings when they are importable, and falls back to build_vrt_xml() otherwise. The GDAL bindings are not pip-installable without a version-matched system libgdal, so pip-only installs take the fallback; conda / HPC installs keep the authoritative backend.

Parameters:

tif_fileslist of str

Paths to input GeoTIFF files

vrt_pathstr

Output VRT file path

Returns:

str

Path to the written VRT

gedih3.raster.export.build_vrt_xml(tif_files, vrt_path)

Write a VRT mosaic as XML using rasterio only — no osgeo bindings.

Handles the narrow case gedih3’s own writers emit: north-up GeoTIFF tiles sharing one CRS and one pixel size. Anything outside that (rotated geotransforms, mixed CRS) raises rather than emitting a subtly wrong mosaic — use the osgeo backend for the general case.

As with build_vrt(), the mosaic resolution is pinned to the first tile so partial edge tiles cannot drag it off the canonical grid.

Parameters:

tif_fileslist of str

Paths to input GeoTIFF files

vrt_pathstr

Output VRT file path

Returns:

str

Path to the written VRT

Raises:

GediRasterizationError

If the tiles are rotated, span multiple CRS, or use a dtype with no GDAL equivalent.

gedih3.raster.export.build_vrt_safe(tif_files, vrt_path)

Build a VRT mosaic, downgrading any failure to a warning.

The .tif tiles are the deliverable; the VRT is a convenience mosaic over them. A missing GDAL backend or an unmosaicable tile set must not discard a completed rasterization job.

Parameters:

tif_fileslist of str

Paths to input GeoTIFF files

vrt_pathstr

Output VRT file path

Returns:

str or None

Path to the written VRT, or None if it could not be built.

gedih3.raster.export.merge_and_export_rasters(gdf: geopandas.GeoDataFrame | dask.dataframe.DataFrame, output_path: str, rasterize_func, columns: List[str] | None = None, compress: str = 'LZW', show_progress: bool = True, **rasterize_kwargs) → str

Rasterize all partitions and merge into a single output file.

Parameters:

gdfGeoDataFrame or dask GeoDataFrame

Input spatially-indexed data

output_pathstr

Output file path

rasterize_funccallable

Function to rasterize each partition

columnslist of str, optional

Columns to include in rasterization

compressstr

Compression method for GeoTIFF

show_progressbool

Show Dask progress bar

Returns:

str

Path to output file

gedih3.raster.export.compute_raster_stats(xras: xarray.Dataset) → Dict[str, Dict[str, float]]

Compute basic statistics for each variable in a raster dataset.

Parameters:

xrasxr.Dataset

Input raster dataset

Returns:

dict

Statistics for each variable: {var_name: {min, max, mean, std, count}}