The aim of this tool is to convert an ESRI file Geodatabase to a PostGIS database maintaining data, domains, subtypes and relationships.
The tool will copy over the feature classes as postgis layers and the tables as standard tables. The domains and subtypes will be converted to PostgreSQL lookup tables.
The tool will then create all necessary indexes and constraints to maintain the required relates between the layers, tables and lookup tables.
The tool creates materialized views of the tables including descriptions of the lookup tables suitable for web publishing.
To recreate the same experience of the domains and subtypes in QGIS using the output data, please install the plugin Data Manager <https://github.com/cartologic/qgis-datamanager-plugin>
_.
Now you can have domain experience in QGIS that is stored in the database and not in the QGIS project.
This library requires GDAL/OGR libraries and ESRI ArcGIS to be installed in the system.
This package should be installed only on windows systems because of ArcGIS (Arcpy) limitation.
This tool has been tested with PostgreSQL 11 , PostGIS 2.5, Arcgis Desktop 10.6.1 and gdal 2.4.0
Install required packages::
pip install psycopg2>=2.6.2
pip install pyyaml>=3.12
pip install archook==1.1.0
Install fgdb2postgis::
git clone https://github.com/dersteppenwolf/fgdb2postgis.git
cd fgdb2postgis
python setup.py install
Note:
-
This tool requires to have GDAL/OGR libraries and ArcGIS 10.3 or later installed.
-
ESRI Python packages usually under C:\Python27\ArcGIS10.* might not have pip included make sure to
- Install pip if not already installed
- Setup ESRI python and GDAL/OGR in windows path environment variable
usage: fgdb2postgis [-h] [-v] [-yml] [--fgdb [FGDB]] [--database [DATABASE]]
[--host [HOST]] [--port [PORT]] [--user [USER]]
[--password [PASSWORD]] [--include_empty [INCLUDE_EMPTY]]
[--lookup_tables_schema [LOOKUP_TABLES_SCHEMA]]
[--a_srs [A_SRS]] [--t_srs [T_SRS]]
Convert a Filegeodatabase to Postgis.
optional arguments:
-h, --help show this help message and exit
-v, --version Program version
-yml, --yml Create .yml and exit
--fgdb [FGDB] Name of the filegeodatabase for conversion
--database [DATABASE]
Name of the postgis database to be created for
conversion
--host [HOST] Database host
--port [PORT] Postgresql port
--user [USER] database user
--password [PASSWORD]
database password
--include_empty [INCLUDE_EMPTY]
Include empty tables and features. Default False
--lookup_tables_schema [LOOKUP_TABLES_SCHEMA]
Name of the schema for lookup tables.
Default:lookup_tables
--a_srs [A_SRS] Assign an output SRS.
--t_srs [T_SRS] Reproject/transform to this SRS on output.
Command line options::
Show help:
fgdb2postgis -h
Generate the yml file and exit:
fgdb2postgis -yml --fgdb sample.gdb
Convert fgdb:
fgdb2postgis --fgdb mygdb.gdb --database=migratetdb --host=localhost --port=5432 --user=user_migrate --password=user_migrate --a_srs=EPSG:4686 --t_srs=EPSG:4686 --include_empty=False --lookup_tables_schema=mylookuptableschema
Create a yaml file mapping the file geodatabase's feature datasets, feature classes and tables to postgresql's schemas. It is required that the yaml file have the same name with the file geodatabase with the extension .yaml
Example::
filegdb: sample.gdb
yaml: sample.gdb.yaml
.. note:: The Yaml file should be located in the same folder with the file geodatabase. If run without the yaml file will convert the full database and load it into the public schema. The schema lookup_tables will always be created regardless of the yaml file.
Yaml file example::
Schemas:
- Administrative
- Epidemiology
- Radioactivity
- Seismic
FeatureDatasets:
Epidemiology:
- Epidemiology
Radioactivity:
- Radioactivity
Seismic:
- Seismic
FeatureClasses:
Administrative:
- sectors
- governorates
- sub_sectors
Tables:
Epidemiology:
- EpidemiologyTS
- EpidemiologyTST
Radioactivity:
- RadiationTS
- RadiationTST
Seismic:
- EarthquakeTS
- SeismicTST
Schemas: The schemas to be created in the target postgis database.
FeatureDatasets: Mapping of the geodatabase's feature datasets to the schemas of the target postgis database
FeatureClasses: Mapping of the geodatabase's feature classes that do not belong to any feature dataset to the schemas of the target postgis database
Tables: Mapping of the geodatabase's tables to the schemas of target postgis database
Tips:
-
Currently the tool support only Latin Name fields and suptypes, domain values can be in any language, make sure to set the corresponding windows domain
-
DB user must be superuser :
CREATE USER user_migrate WITH PASSWORD 'xxxxx' LOGIN SUPERUSER INHERIT CREATEDB CREATEROLE NOREPLICATION;
- if you want to drop the migration user use:
revoke all ON SCHEMA information_schema from user_migrate ;
revoke all ON ALL TABLES IN SCHEMA information_schema from user_migrate;
drop user user_migrate;
Warnings:
-
DO NOT apply this tool in a production postgis database!. Insted use a staging database.
-
The target postgis database is created by the program
-
The tool only includes Simple features (Polygons, polylines, and points representing objects or places. See https://desktop.arcgis.com/es/arcmap/latest/analyze/arcpy-functions/featureclass-properties.htm )
-
At this moment only the following geometry type conversions are supported:
Esri Type | GDAL Type |
---|---|
Polygon | MULTIPOLYGON |
Polyline | MULTILINESTRING |
Point | POINT |
Multipoint | MULTIPOINT |
-
If you do not use the python distribution included in Arcgis Desktop you must have into account the ArcGIS Desktop and Numpy compability (see https://support.esri.com/en/technical-article/000013224):
- 10.7.1 - Python 2.7.16 and NumPy 1.9.3
- 10.7 - Python 2.7.15 and NumPy 1.9.3
- 10.6.1 - Python 2.7.14 and NumPy 1.9.3
- 10.6 - Python 2.7.14 and NumPy 1.9.3
- 10.5.1 - Python 2.7.13 and NumPy 1.9.3
- 10.5 - Python 2.7.12 and NumPy 1.9.3
The tool creates a materialized view for each postgis table including the descriptions (label) of the related lookup tables. Such materialized view can be used for web mapping using software like Geoserver.
Example:
-- Feature
CREATE MATERIALIZED VIEW cartografia_100k.hito_limite_mv AS
select t.* , ft0.description as ruleid_label , ft1.description as symbol_label
from cartografia_100k.hito_limite as t left join cartografia_100k.lut_hito_limite_rep_rules as ft0 on ( t.ruleid = ft0.code ) left join cartografia_100k.lut_dom_gen_plts as ft1 on ( t.symbol = ft1.code )
WITH DATA;
Credit goes to James Ramm <ramshacklerecording@gmail.com>
_ who kindly developed and shared the archook package.
Refactored by Juan Carlos Méndez https://github.com/dersteppenwolf
GNU Public License (GPL) Version 3