Reads a CDB for coverages, mosaics them into temporary files, imports them into GeoServer, and sets up a WCS.
I recommend using QGIS as a testing client to preview rasters generated with cdb-mosaic, and to preview rasters from the GeoServer WCS.
These scripts require Ruby 2.4.0 or newer to run.
Uses GDAL to add overviews to GeoTIFFs or VRTs in given directory. Run this after cdb-mosaic.
$ ruby -Ilib bin/cdb-addo path/to/CDB_temp/final
Overviews will make the GeoTIFFs/VRTs load faster in GeoServer WCS. VRTs will have external .ovr files created, TIFFs will have their overviews embedded.
$ ruby -Ilib bin/cdb-list-components path/to/CDB
List datasets in a CDB, grouped by component selectors. Counts the files found for each component selector.
$ ruby -Ilib bin/cdb-list-datasets path/to/CDB
List datasets in a CDB, grouped by geocells. Counts the GeoTIFFs and JP2s found.
$ ruby -Ilib bin/cdb-list-geocells path/to/CDB
List geocells in a CDB, and the sub-collections for each geocell.
$ ruby -Ilib bin/cdb-mosaic path/to/CDB
Creates a new directory at path/to/CDB_temp and scans the coverages in the CDB and merges them using gdal_merge.py. First merges in UREFs under the highest LODs, then merges UREFs, then merges GeoCells to leave datasets. Different component selectors are merged separately. All output files are GeoTIFFs.
Output GeoTIFFs will have overviews added using GDAL. Lossy rasters will have JPEG overviews, lossless rasters will have LZW compressed overviews.
$ ruby -Ilib bin/cdb-mosaic-vrt path/to/CDB
Creates a new directory at path/to/CDB_temp and scans the coverages in the CDB and merges them using virtual datasets (VRTs). Files are selected from 001_Elevation and 004_Imagery under the highest LODs, then grouped by Component Selectors, GeoCells, and then UREFs. They are then combined into VRTs in the reverse order.
All output files are VRTs. Very fast, but compatibility may be limited.
$ ruby -Ilib bin/cdb-verify path/to/CDB
Checks that the CDB has a Tiles directory.
$ ruby -Ilib bin/geoserver-setup path/to/CDB_temp/final http://user:password@geoserver:port/geoserver
Logs into GeoServer, creates a new workspace, creates a coverage store for every raster in the final directory, then creates coverage layers for each raster. If there is an error, it quits and prints an error message. Will delete the workspace if import fails.
Set the environment variable HTTPS to false to disable HTTPS connections to GeoServer during configuration.
$ ruby -Ilib bin/geoserver-setup-vrts path/to/CDB_temp/final http://user:password@geoserver:port/geoserver
Logs into GeoServer, creates a new workspace, creates a coverage store for every VRT in the "final" directory, then creates coverage layers for each raster. If there is an error, it quits and prints an error message. Will delete the workspace if import fails.
Set the environment variable HTTPS to false to disable HTTPS connections to GeoServer during configuration.
GeoServer requires the GDAL plugin for VRTs to work! For installation instructions on Linux and Windows, see the GeoServer Docs. For MacOS installation, see this GeoServer mailing list post. (I put a copy of that post in the docs directory if sourceforge is unavailable.)
For Mac it is also important to edit /usr/local/bin/geoserver that was installed by Mac Homebrew, and change to:
#!/bin/sh
export PATH="/usr/local/opt/gdal2/bin:$PATH"
export GDAL_DATA=/usr/local/Cellar/gdal2/2.2.0/share/gdal
export DYLD_LIBRARY_PATH="/usr/local/lib"
export LD_LIBRARY_PATH="/usr/local/lib"
if [ -z "$1" ]; then
echo "Usage: $ geoserver path/to/data/dir"
else
cd "/usr/local/Cellar/geoserver/2.11.1/libexec" && java -Djava.library.path=/usr/local/lib/ -DGEOSERVER_DATA_DIR=$1 -jar start.jar
fiThis makes sure the GDAL libraries are loaded by GeoServer. The log output from GeoServer will specify if GDAL loaded properly or not.
Copyright GeoServerWeb Lab 2017, All Rights Reserved.
James Badger