sridentify is a command-line utility and Python API for quickly
identifying the EPSG Registry Code
from a .prj file typically associated with ESRI
Shapefiles. It ships with a
SQLite database containing mappings of Well-known
Text strings to EPSG
codes, the bulk of which was manually sourced and cleaned from an ESRI
website.
It's not complete, however, and in the event you test it against a WKT
string not in the database it will search the
prj2epsg.org API. If the API returns an exact
match, that code is returned and saved to the SQLite database. Handling
several partial matches is currently planned, but not yet implemented.
sridentify is written in Python, 2- and 3-compatible, and has no external dependencies.
pip install --user sridentify
The --user is important if installing system-wide (i.e., not in a virtualenv), because the
user running sridentify must have write permissions on the SQLite database in the event that
sridentify tries to save a new result fetched from the prj2epsg API to the database.
By default pip install --user will install to $HOME/.local and place the executable script
in $HOME/.local/bin. You should add this to your $PATH if you want to run sridentify
without having to specify the full location to the executable.
$ sridentify seattle_land_use.prj
2285
# Example use in conjunction with the `shp2pgsql` command-line utility
# that ships with PostGIS. Assuming you have a PostGIS-enabled database named `seattle`,
# and you have a shapefile called `seattle_land_use` that you want to import into that database
# but you're not sure what spatial projection the shapefile uses.
$ shp2pgsql -s $(sridentify seattle_land_use.prj) -g the_geom -ID seattle_land_use.shp | psql -d seattle>>> from sridentify import Sridentify
>>> # Read .prj file from the filesystem
>>> ident = Sridentify()
>>> ident.from_file('/path/to/seattle_land_use.prj')
>>> ident.get_epsg()
2285
>>> # Paste in Well-Known Text string directly
>>> ident = Sridentify(prj="""PROJCS["NAD_1983_StatePlane_Washington_North_FIPS_4601_Feet",GEOGCS["GCS_North_American_1983",DATUM["D_North_American_1983",SPHEROID["GRS_1980",6378137.0,298.257222101]],PRIMEM["Greenwich",0.0],UNIT["Degree",0.0174532925199433]],PROJECTION["Lambert_Conformal_Conic"],PARAMETER["False_Easting",1640416.666666667],PARAMETER["False_Northing",0.0],PARAMETER["Central_Meridian",-120.8333333333333],PARAMETER["Standard_Parallel_1",47.5],PARAMETER["Standard_Parallel_2",48.73333333333333],PARAMETER["Latitude_Of_Origin",47.0],UNIT["Foot_US",0.3048006096012192]]""")
>>> ident.get_epsg()
2285More and more governments and organizations are making their GIS data available to the public on open data portals. Local governments typically store and use GIS data in the map projection most appropriate for their location on planet Earth. For the United States, this is typically the State Plane Coordinate System. Other common systems are Universal Transverse Mercator, or a highly localized system that is accurate only within the geographic boundaries of the entity's jusrisdiction.
ESRI Shapefiles are a common format for publishing GIS data, although a "shapefile" with the .shp extension is really just data describing the geometry. Shapefiles are typically bundled with a dBase file ( .dbf extension ) which contains data attributes about the geometry and a small text file describing the spatial reference system of the geomtry in WKT format.
sridentify is not meant to be a full-fledged client library to the actual
EPSG database, for that you're probably looking for something like python-epsg
Rather, sridentify is for those looking to quickly identify the EPSG code
of a shapefile, especially when importing into PostGIS . Of course, you could use ogr2ogr
to convert everything into a web-friendly projection, like:
$ ogr2ogr -f PostgreSQL -t_srs EPSG:4326 PG:dbname=seattle seattle_land_use.shpBut transforming spatial data from one projection to another is a lossy operation and can result in coordinate drift. Ideally, you would store the original data in its original coordinate system and then transform copies as needed.