sodapy 2.2.0

Python library for the Socrata Open Data API

sodapy - UNMAINTAINED

:rotating_light: NOTICE: sodapy still works well, but is unmaintained as of Aug 31, 2022. No new features or bugfixes will be added. Use at your own risk.

sodapy is a python client for the Socrata Open Data API.

Installation

You can install with pip install sodapy.

If you want to install from source, then clone this repository and run python setup.py install from the project root.

Requirements

At its core, this library depends heavily on the Requests package. All other requirements can be found in requirements.txt. sodapy is currently compatible with Python 3.5, 3.6, 3.7, 3.8, 3.9, and 3.10.

Documentation

The official Socrata Open Data API docs provide thorough documentation of the available methods, as well as other client libraries. A quick list of eligible domains to use with this API is available via the Socrata Discovery API or Socrata's Open Data Network.

This library supports writing directly to datasets with the Socrata Open Data API. For write operations that use data transformations in the Socrata Data Management Experience (the user interface for creating datasets), use the Socrata Data Management API. For more details on when to use SODA vs the Data Management API, see the Data Management API documentation. A Python SDK for the Socrata Data Management API can be found at socrata-py.

Examples

There are some jupyter notebooks in the examples directory with usage examples of sodapy in action.

Interface

Table of Contents

• client
• datasets
• get
• get_all
• get_metadata
• update_metadata
• download_attachments
• create
• publish
• set_permission
• upsert
• replace
• create_non_data_file
• replace_non_data_file
• delete
• close

client

Import the library and set up a connection to get started.

>>> from sodapy import Socrata
>>> client = Socrata(
        "sandbox.demo.socrata.com",
        "FakeAppToken",
        username="[e-mail]",
        password="mypassword",
        timeout=10
    )

username and password are only required for creating or modifying data. An application token isn't strictly required (can be None), but queries executed from a client without an application token will be subjected to strict throttling limits. You may want to increase the timeout seconds when making large requests. To create a bare-bones client:

>>> client = Socrata("sandbox.demo.socrata.com", None)

A client can also be created with a context manager to obviate the need for teardown:

>>> with Socrata("sandbox.demo.socrata.com", None) as client:
>>>    # do some stuff

The client, by default, makes requests over HTTPS. To modify this behavior, or to make requests through a proxy, take a look here.

datasets(limit=0, offset=0)

Retrieve datasets associated with a particular domain. The optional limit and offset keyword args can be used to retrieve a subset of the datasets. By default, all datasets are returned.

>>> client.datasets()
[{"resource" : {"name" : "Approved Building Permits", "id" : "msk6-43c6", "parent_fxf" : null, "description" : "Data of approved building/construction permits",...}, {resource : {...}}, ...]

get(dataset_identifier, content_type="json", **kwargs)

Retrieve data from the requested resources. Filter and query data by field name, id, or using SoQL keywords.

>>> client.get("nimj-3ivp", limit=2)
[{u'geolocation': {u'latitude': u'41.1085', u'needs_recoding': False, u'longitude': u'-117.6135'}, u'version': u'9', u'source': u'nn', u'region': u'Nevada', u'occurred_at': u'2012-09-14T22:38:01', u'number_of_stations': u'15', u'depth': u'7.60', u'magnitude': u'2.7', u'earthquake_id': u'00388610'}, {...}]

>>> client.get("nimj-3ivp", where="depth > 300", order="magnitude DESC", exclude_system_fields=False)
[{u'geolocation': {u'latitude': u'-15.563', u'needs_recoding': False, u'longitude': u'-175.6104'}, u'version': u'9', u':updated_at': 1348778988, u'number_of_stations': u'275', u'region': u'Tonga', u':created_meta': u'21484', u'occurred_at': u'2012-09-13T21:16:43', u':id': 132, u'source': u'us', u'depth': u'328.30', u'magnitude': u'4.8', u':meta': u'{\n}', u':updated_meta': u'21484', u'earthquake_id': u'c000cnb5', u':created_at': 1348778988}, {...}]

>>> client.get("nimj-3ivp/193", exclude_system_fields=False)
{u'geolocation': {u'latitude': u'21.6711', u'needs_recoding': False, u'longitude': u'142.9236'}, u'version': u'C', u':updated_at': 1348778988, u'number_of_stations': u'136', u'region': u'Mariana Islands region', u':created_meta': u'21484', u'occurred_at': u'2012-09-13T11:19:07', u':id': 193, u'source': u'us', u'depth': u'300.70', u'magnitude': u'4.4', u':meta': u'{\n}', u':updated_meta': u'21484', u':position': 193, u'earthquake_id': u'c000cmsq', u':created_at': 1348778988}

>>> client.get("nimj-3ivp", region="Kansas")
[{u'geolocation': {u'latitude': u'38.10', u'needs_recoding': False, u'longitude': u'-100.6135'}, u'version': u'9', u'source': u'nn', u'region': u'Kansas', u'occurred_at': u'2010-09-19T20:52:09', u'number_of_stations': u'15', u'depth': u'300.0', u'magnitude': u'1.9', u'earthquake_id': u'00189621'}, {...}]

get_all(dataset_identifier, content_type="json", **kwargs)

Read data from the requested resource, paginating over all results. Accepts the same arguments as get(). Returns a generator.

>>> client.get_all("nimj-3ivp")
<generator object Socrata.get_all at 0x7fa0dc8be7b0>

>>> for item in client.get_all("nimj-3ivp"):
...     print(item)
...
{'geolocation': {'latitude': '-15.563', 'needs_recoding': False, 'longitude': '-175.6104'}, 'version': '9', ':updated_at': 1348778988, 'number_of_stations': '275', 'region': 'Tonga', ':created_meta': '21484', 'occurred_at': '2012-09-13T21:16:43', ':id': 132, 'source': 'us', 'depth': '328.30', 'magnitude': '4.8', ':meta': '{\n}', ':updated_meta': '21484', 'earthquake_id': 'c000cnb5', ':created_at': 1348778988}
...

>>> import itertools
>>> items = client.get_all("nimj-3ivp")
>>> first_five = list(itertools.islice(items, 5))
>>> len(first_five)
5

get_metadata(dataset_identifier, content_type="json")

Retrieve the metadata associated with a particular dataset.

>>> client.get_metadata("nimj-3ivp")
{"newBackend": false, "licenseId": "CC0_10", "publicationDate": 1436655117, "viewLastModified": 1451289003, "owner": {"roleName": "administrator", "rights": [], "displayName": "Brett", "id": "cdqe-xcn5", "screenName": "Brett"}, "query": {}, "id": "songs", "createdAt": 1398014181, "category": "Public Safety", "publicationAppendEnabled": true, "publicationStage": "published", "rowsUpdatedBy": "cdqe-xcn5", "publicationGroup": 1552205, "displayType": "table", "state": "normal", "attributionLink": "http://foo.bar.com", "tableId": 3523378, "columns": [], "metadata": {"rdfSubject": "0", "renderTypeConfig": {"visible": {"table": true}}, "availableDisplayTypes": ["table", "fatrow", "page"], "attachments": ... }}

update_metadata(dataset_identifier, update_fields, content_type="json")

Update the metadata for a particular dataset. update_fields should be a dictionary containing only the metadata keys that you wish to overwrite.

Note: Invalid payloads to this method could corrupt the dataset or visualization. See this comment for more information.

>>> client.update_metadata("nimj-3ivp", {"attributionLink": "https://anothertest.com"})
{"newBackend": false, "licenseId": "CC0_10", "publicationDate": 1436655117, "viewLastModified": 1451289003, "owner": {"roleName": "administrator", "rights": [], "displayName": "Brett", "id": "cdqe-xcn5", "screenName": "Brett"}, "query": {}, "id": "songs", "createdAt": 1398014181, "category": "Public Safety", "publicationAppendEnabled": true, "publicationStage": "published", "rowsUpdatedBy": "cdqe-xcn5", "publicationGroup": 1552205, "displayType": "table", "state": "normal", "attributionLink": "https://anothertest.com", "tableId": 3523378, "columns": [], "metadata": {"rdfSubject": "0", "renderTypeConfig": {"visible": {"table": true}}, "availableDisplayTypes": ["table", "fatrow", "page"], "attachments": ... }}

download_attachments(dataset_identifier, content_type="json", download_dir="~/sodapy_downloads")

Download all attachments associated with a dataset. Return a list of paths to the downloaded files.

>>> client.download_attachments("nimj-3ivp", download_dir="~/Desktop")
    ['/Users/xmunoz/Desktop/nimj-3ivp/FireIncident_Codes.PDF', '/Users/xmunoz/Desktop/nimj-3ivp/AccidentReport.jpg']

create(name, **kwargs)

Create a new dataset. Optionally, specify keyword args such as:

• description description of the dataset
• columns list of fields
• category dataset category (must exist in /admin/metadata)
• tags list of tag strings
• row_identifier field name of primary key
• new_backend whether to create the dataset in the new backend

Example usage:

>>> columns = [{"fieldName": "delegation", "name": "Delegation", "dataTypeName": "text"}, {"fieldName": "members", "name": "Members", "dataTypeName": "number"}]
>>> tags = ["politics", "geography"]
>>> client.create("Delegates", description="List of delegates", columns=columns, row_identifier="delegation", tags=tags, category="Transparency")
{u'id': u'2frc-hyvj', u'name': u'Foo Bar', u'description': u'test dataset', u'publicationStage': u'unpublished', u'columns': [ { u'name': u'Foo', u'dataTypeName': u'text', u'fieldName': u'foo', ... }, { u'name': u'Bar', u'dataTypeName': u'number', u'fieldName': u'bar', ... } ], u'metadata': { u'rowIdentifier': 230641051 }, ... }

publish(dataset_identifier, content_type="json")

Publish a dataset after creating it, i.e. take it out of 'working copy' mode. The dataset id id returned from create will be used to publish.

>>> client.publish("2frc-hyvj")
{u'id': u'2frc-hyvj', u'name': u'Foo Bar', u'description': u'test dataset', u'publicationStage': u'unpublished', u'columns': [ { u'name': u'Foo', u'dataTypeName': u'text', u'fieldName': u'foo', ... }, { u'name': u'Bar', u'dataTypeName': u'number', u'fieldName': u'bar', ... } ], u'metadata': { u'rowIdentifier': 230641051 }, ... }

set_permission(dataset_identifier, permission="private", content_type="json")

Set the permissions of a dataset to public or private.

>>> client.set_permission("2frc-hyvj", "public")
<Response [200]>

upsert(dataset_identifier, payload, content_type="json")

Create a new row in an existing dataset.

>>> data = [{'Delegation': 'AJU', 'Name': 'Alaska', 'Key': 'AL', 'Entity': 'Juneau'}]
>>> client.upsert("eb9n-hr43", data)
{u'Errors': 0, u'Rows Deleted': 0, u'Rows Updated': 0, u'By SID': 0, u'Rows Created': 1, u'By RowIdentifier': 0}

Update/Delete rows in a dataset.

>>> data = [{'Delegation': 'sfa', ':id': 8, 'Name': 'bar', 'Key': 'doo', 'Entity': 'dsfsd'}, {':id': 7, ':deleted': True}]
>>> client.upsert("eb9n-hr43", data)
{u'Errors': 0, u'Rows Deleted': 1, u'Rows Updated': 1, u'By SID': 2, u'Rows Created': 0, u'By RowIdentifier': 0}

upsert's can even be performed with a csv file.

>>> data = open("upsert_test.csv")
>>> client.upsert("eb9n-hr43", data)
{u'Errors': 0, u'Rows Deleted': 0, u'Rows Updated': 1, u'By SID': 1, u'Rows Created': 0, u'By RowIdentifier': 0}

replace(dataset_identifier, payload, content_type="json")

Similar in usage to upsert, but overwrites existing data.

>>> data = open("replace_test.csv")
>>> client.replace("eb9n-hr43", data)
{u'Errors': 0, u'Rows Deleted': 0, u'Rows Updated': 0, u'By SID': 0, u'Rows Created': 12, u'By RowIdentifier': 0}

create_non_data_file(params, file_obj)

Creates a new file-based dataset with the name provided in the files tuple. A valid file input would be:

files = (
    {'file': ("gtfs2", open('myfile.zip', 'rb'))}
)

>>> with open(nondatafile_path, 'rb') as f:
>>>     files = (
>>>         {'file': ("nondatafile.zip", f)}
>>>     )
>>>     response = client.create_non_data_file(params, files)

replace_non_data_file(dataset_identifier, params, file_obj)

Same as create_non_data_file, but replaces a file that already exists in a file-based dataset.

Note: a table-based dataset cannot be replaced by a file-based dataset. Use create_non_data_file in order to replace.

>>>  with open(nondatafile_path, 'rb') as f:
>>>      files = (
>>>          {'file': ("nondatafile.zip", f)}
>>>      )
>>>      response = client.replace_non_data_file(DATASET_IDENTIFIER, {}, files)

delete(dataset_identifier, row_id=None, content_type="json")

Delete an individual row.

>>> client.delete("nimj-3ivp", row_id=2)
<Response [200]>

Delete the entire dataset.

>>> client.delete("nimj-3ivp")
<Response [200]>

close()

Close the session when you're finished.

>>> client.close()

Run tests

$ pytest

Contributing

See CONTRIBUTING.md.

Meta

This package uses semantic versioning.

Source and wheel distributions are available on PyPI. Here is how I create those releases.

python3 setup.py bdist_wheel
python3 setup.py sdist
twine upload dist/*