A library to make web service calls to map service REST APIs easier. Currently supported:
- ArcGIS (version 10.1 and greater)
- THREDDS
- WMS / NcWMS (versions 1.1.1 and 1.3.0)
- ScienceBase
Each leverages the restle library to represent queried map service data as Python objects. Each also provides some default functionality for rendering projected map service data as images, which may be overridden per class as needed.
Beyond this are some utilities for working with images (PIL) and extents (mostly Geographic, Web Mercator and other proj4 compatible projections).
Install with pip install mapservice-clientlib
.
Below are some examples of each supported map service web API standard:
ArcGIS Map, Feature and Image services may be queried.
from clients.arcgis import MapServerResource, ArcGISSecureResource
from clients.arcgis import FeatureLayerResource, FeatureServerResource, ImageServerResource
from clients.utils.geometry import Extent
# Query the feature service, or an individual layer (lazy=False: query executed right away)
client = FeatureServerResource.get(service_url, lazy=False)
layer = FeatureLayerResource.get(service_url + "/0", lazy=False)
# Query an image service lazily (default behavior: executes query on property reference)
client = ImageServerResource.get(service_url, lazy=True)
client.extent # Query executes here
# Query a map service and generate an image
arcgis_image = MapServerResource.get(service_url).get_image(
extent, width=400, height=200,
layers="show:0",
layer_defs="<arcgis_layer_defs>",
time="<arcgis_time_val>",
custom_renderers={} # Renderer JSON
)
# Query a secure map service (generates token from URL and credentials)
client = MapServerResource.get(service_url, username="user", password="pass")
# Query a secure map service with existing token
token_obj = ArcGISSecureResource.generate_token(
service_url, "user", "pass", duration=15
)
client = MapServerResource.get(service_url, token=token_obj.token)
# Reproject an ArcGIS extent to Web Mercator
old_extent = Extent(
{'xmin': -180.0000, 'xmax': 180.0000, 'ymin': -90.0000, 'ymax': 90.0000},
spatial_reference={'wkid': 4326}
)
geometry_url = 'http://tasks.arcgisonline.com/arcgis/rest/services/Geometry/GeometryServer'
client = GeometryServiceClient(geometry_url)
extent = client.project_extent(old_extent, {'wkid': 3857}).limit_to_global_extent()
WMS services may be queried, with support for NcWMS
from clients.wms import WMSResource
# Query a secure WMS service
client = WMSResource.get(
url=wms_url, token="token", token_id="josso", version="1.3.0", spatial_ref="EPSG:3857"
)
# Query a public WMS service and generate an image (supports NcWMS as well)
wms_image = WMSResource.get(
wms_url
).get_image(
extent, width=400, height=200,
layer_ids=[...],
style_ids=[...],
time_range="<wms_time_val>",
params={...}, # Additional image params
image_format="png"
)
THREDDS resources may be queried, with metadata from related WMS endpoint:
from clients.thredds import ThreddsResource
client = ThreddsResource.get(url)
# See gis-metadata-parser for more
metadata = client._metadata_parser
metadata.data_credits
metadata.use_constraints
# Makes a WMS image request
thredds_image = client.get_image(
extent, width, height,
layer_ids=[...],
style_ids=[...],
time_range="<wms_time_val>",
params={...}, # Additional image params
image_format="png"
)
Public and private ScienceBase items may be queried:
from clients.sciencebase import ScienceBaseResource, ScienceBaseSession
# Query a public ScienceBase item
client = ScienceBaseResource.get(service_url, lazy=False)
client.summary
# Query a private WMS-backed ScienceBase item
sb_session = ScienceBaseSession(
josso_session_id="token",
username="sciencebase_user"
)
sb_session.login("sciencebase_user", "pass")
client = ScienceBaseResource.get(
url=service_url,
lazy=False,
session=sb_session,
# Same token for WMS as for base item
token=sb_session._jossosessionid
)
client.service_client.full_extent # WMSResource.full_extent
# Query a private ArcGIS-backed ScienceBase item
sb_session = ScienceBaseSession(
josso_session_id="token",
username="sciencebase_user"
)
sb_session.login("sciencebase_user", "pass")
client = ScienceBaseResource.get(
url=service_url,
lazy=False,
session=sb_session,
token=sb_session._jossosessionid,
# Separate credentials for ArcGIS service
arcgis_credentials={
# Or just use "token": "existing_token"
"username": "arcgis_user",
"password": "arcgis_pass"
}
)
client.service_client.full_extent # ArcGISResource.full_extent
Extent objects have a number of useful methods. Here are some examples that support projection:
from clients.utils.geometry import Extent
extent_from_dict = Extent({
"xmin": -180.0, "ymin": -90.0, "xmax": 180.0, "ymax": 90.0,
"spatial_reference": {"wkid": 4326}
})
web_mercator_extent = extent_from_dict.project_to_web_mercator()
extent_from_list = Extent(
# In order of xmin, ymin, xmax, ymax
[-20037508.342789244, -20037471.205137067, 20037508.342789244, 20037471.20513706],
spatial_reference="EPSG:3857"
)
geographic_extent = extent_from_dict.project_to_geographic()