Skip to content

Latest commit

 

History

History
348 lines (306 loc) · 9.65 KB

README.md

File metadata and controls

348 lines (306 loc) · 9.65 KB
Raw

FastGeostats

FastGeostats is a geospatial api to perform math and statistics on tables from a mulititude of PostgreSQL databases. FastGeostats is written in Python using the FastAPI web framework.

Source Code: https://github.com/mkeller3/FastGeostats

Requirements

FastGeostats requires PostGIS >= 2.4.0.

Configuration

In order for the api to work you will need to edit the config.py file with your database connections.

Example

DATABASES = {
    "data": {
        "host": "localhost", # Hostname of the server
        "database": "data", # Name of the database
        "username": "postgres", # Name of the user, ideally only SELECT rights
        "password": "postgres", # Password of the user
        "port": 5432, # Port number for PostgreSQL
    }
}

Usage

Running Locally

To run the app locally uvicorn main:app --reload

Production

Build Dockerfile into a docker image to deploy to the cloud.

API

Method URL Description
POST /api/v1/tables/statistics Statistics
POST /api/v1/tables/bins Bins
POST /api/v1/tables/numeric_breaks Numeric Breaks
POST /api/v1/tables/custom_break_values Custom Break Values
GET /api/v1/health_check Server health check: returns 200 OK

Endpoint Description's

Statistics

Description

The statistics endpoints allows you to perform a multitude of common math statistics on your table such as 'distinct', 'avg', 'count', 'sum', 'max', 'min'.

Parameters

  • coordinates=coords - a list of coordinates used to filter the response
  • geometry_type=geom_type - The type of geometry from the coordinates parameters. Options: 'POINT', 'LINESTRING', 'POLYGON'
  • spatial_relationship=relationship - The type of spatial query to perform with the coordinates parameters. Options: 'ST_Intersects', 'ST_Crosses', 'ST_Within', 'ST_Contains', 'ST_Overlaps', 'ST_Disjoint', 'ST_Touches'
  • filter=cql-expr - filters features via a CQL expression.

Example: In the example below we will be searching for the number of parcels, average deed ac, and distinct first names filtered by last name of DOOLEY.

Example Input

{
    "database": "data",
    "table": "mclean_county_parcels",
    "aggregate_columns": [
        {
            "type": "count",
            "column": "gid"
        },
        {
            "type": "avg",
            "column": "deed_ac"
        },
        {
            "type": "distinct",
            "column": "first_name",
            "group_column": "first_name",
            "group_method": "count"
        }
    ],
    "filter": "last_name LIKE '%DOOLEY%'"
}

Example Output

{
    "results": {
        "count_gid": 19,
        "avg_deed_ac": 64.28666666666666,
        "distinct_first_name_count_first_name": [
            {
                "first_name": "",
                "count": 3
            },
            {
                "first_name": "COLE",
                "count": 3
            },
            {
                "first_name": "% BAS",
                "count": 2
            },
            {
                "first_name": "%FIRST MID AG SERVICES ",
                "count": 2
            },
            {
                "first_name": "COLE & WENDY",
                "count": 1
            },
            {
                "first_name": "EDITH",
                "count": 1
            },
            {
                "first_name": "JAMES R & TERESA",
                "count": 1
            },
            {
                "first_name": "KENNETH",
                "count": 1
            },
            {
                "first_name": "KEVIN",
                "count": 1
            },
            {
                "first_name": "LUCAS",
                "count": 1
            },
            {
                "first_name": "MCCALLA O & DEANA J",
                "count": 1
            },
            {
                "first_name": "THOMAS",
                "count": 1
            },
            {
                "first_name": "WENDY",
                "count": 1
            }
        ]
    },
    "status": "SUCCESS"
}

Bins

Description

The bins endpoints allows you to help visualize the spread of a data for a numerical column.

Parameters

  • coordinates=coords - a list of coordinates used to filter the response
  • geometry_type=geom_type - The type of geometry from the coordinates parameters. Options: 'POINT', 'LINESTRING', 'POLYGON'
  • spatial_relationship=relationship - The type of spatial query to perform with the coordinates parameters. Options: 'ST_Intersects', 'ST_Crosses', 'ST_Within', 'ST_Contains', 'ST_Overlaps', 'ST_Disjoint', 'ST_Touches'
  • filter=cql-expr - filters features via a CQL expression.

Example: Calculate 10 bins for the deed_ac column on the mclean_county_parcels table.

Example Input

{
    "database": "data",
    "table": "mclean_county_parcels",
    "column": "deed_ac",
    "bins": 10
}

Example Output

{
    "results": [
        {
            "min": 0.0,
            "max": 145.158,
            "count": 15993
        },
        {
            "min": 145.158,
            "max": 290.316,
            "count": 1088
        },
        {
            "min": 290.316,
            "max": 435.47399999999993,
            "count": 116
        },
        {
            "min": 435.47399999999993,
            "max": 580.632,
            "count": 19
        },
        {
            "min": 580.632,
            "max": 725.79,
            "count": 11
        },
        {
            "min": 725.79,
            "max": 870.9479999999999,
            "count": 1
        },
        {
            "min": 870.9479999999999,
            "max": 1016.1059999999999,
            "count": 0
        },
        {
            "min": 1016.1059999999999,
            "max": 1161.264,
            "count": 0
        },
        {
            "min": 1161.264,
            "max": 1306.4219999999998,
            "count": 0
        },
        {
            "min": 1306.4219999999998,
            "max": 1451.58,
            "count": 1
        }
    ],
    "status": "SUCCESS"
}

Numeric Breaks

Description

Create bins of data based off of different mathmatical break types.

Break Types: equal_interval, head_tail, quantile, jenk.

Parameters

  • coordinates=coords - a list of coordinates used to filter the response
  • geometry_type=geom_type - The type of geometry from the coordinates parameters. Options: 'POINT', 'LINESTRING', 'POLYGON'
  • spatial_relationship=relationship - The type of spatial query to perform with the coordinates parameters. Options: 'ST_Intersects', 'ST_Crosses', 'ST_Within', 'ST_Contains', 'ST_Overlaps', 'ST_Disjoint', 'ST_Touches'
  • filter=cql-expr - filters features via a CQL expression.

Example: Create 3 breaks based off of the column population for the table zip_centroids using a quantile break type.

Example Input

{
    "database": "data",
    "table": "zip_centroids",
    "column": "population",
    "number_of_breaks": 3,
    "break_type": "quantile"
}

Example Output

{
    "results": [
        {
            "min": 0,
            "max": 1470,
            "count": 10301
        },
        {
            "min": 1470,
            "max": 8932,
            "count": 10373
        },
        {
            "min": 8932,
            "max": 133324,
            "count": 10377
        }
    ],
    "status": "SUCCESS"
}

Custom Break Values

Description

Create bins based off of your own min and max ranges and provide a count back for each bin.

Parameters

  • coordinates=coords - a list of coordinates used to filter the response
  • geometry_type=geom_type - The type of geometry from the coordinates parameters. Options: 'POINT', 'LINESTRING', 'POLYGON'
  • spatial_relationship=relationship - The type of spatial query to perform with the coordinates parameters. Options: 'ST_Intersects', 'ST_Crosses', 'ST_Within', 'ST_Contains', 'ST_Overlaps', 'ST_Disjoint', 'ST_Touches'
  • filter=cql-expr - filters features via a CQL expression.

Example: Create 3 custom bins 0 - 1,000, 1,000 - 9,000, and 9,000 - 140,000 based off of the column population for the table zip_centroids using a quantile break type.

Example Input

{
    "database": "data",
    "table": "zip_centroids",
    "column": "population",
    "breaks": [
        {
            "min": 0,
            "max": 1000
        },
        {
            "min": 1000,
            "max": 9000
        },
        {
            "min": 9000,
            "max": 140000
        }
    ]
}

Example Output

{
    "results": [
        {
            "min": 0.0,
            "max": 1000.0,
            "count": 7981
        },
        {
            "min": 1000.0,
            "max": 9000.0,
            "count": 12720
        },
        {
            "min": 9000.0,
            "max": 140000.0,
            "count": 10350
        }
    ],
    "status": "SUCCESS"
}