kgbase 0.25

No project description provided

Kgbase

Installation

pip install kgbase

Authentication

To import library.

from kgbase import Query

To authenticate:

q = Query()
q.login(
    username='Your username',
    password='Your password'
)

You must first signup from kgbase website. Your password must not be shared or exposed via publicly accessible resources (such as browser client-side scripting).

To use a proxy:

proxies = {
  "http": "http://10.10.1.10:3128",
  "https": "http://10.10.1.10:1080",
}

q = Query(
    proxies=proxies
)

To ignore verifying the SSL certficate:

q = Query(
    verify=False
)

To get user state:

q.get_user_state()

To logout:

q.logout()

Projects

To get all public projects:

q.get_public_projects()

To get all team projects:

q.get_team_projects()

To get favorite projects:

q.get_favorite_projects()

To get my projects:

q.get_user_projects()

To create project:

q.create_project(
    name='electronic devices', 
    is_public=True, 
)

The projectId is returned from response. You need to know projectId when you update or detroy project.

{
    "ok": true,
    "project": {
        "uuid": "a0406c6d-122a-4ff2-af7f-976ba5cbca01",
        "graphqlId": "Project/1509",
        "projectId": "ctx-M53lgnjpCkc_plt0lqo",
        "...": "..."
    }
}

To get project state:

q.get_project_state(
    project_id='ctx-M53lgnjpCkc_plt0lqo'
)

To update project:

result = q.update_project(
    project_id='ctx-M53lgnjpCkc_plt0lqo', 
    name='electronic devices',
    is_public=False
)

To destroy project:

q.destroy_project(
    project_id='ctx-M53lgnjpCkc_plt0lqo'
)

Table

To create table:

q.create_table(
    project_id='ctx-M53pLqASSUmxj5yU7LO',
    display_name='Product',
    description='Product Description'
)

The tableId is returned from the response. You need to know tableId when you update or delete table.

{
    "ok": true,
    "tableId": "tab-M53nDu1y1SXMVA_ny97"
}

To update table:

q.update_table(
    project_id='ctx-M53lgnjpCkc_plt0lqo',
    table_id='tab-M53nDu1y1SXMVA_ny97',
    display_name='Smartphones',
    description='Smartphones manufacurer and model name',
)

To delete table:

q.delete_table(
    project_id='ctx-M53lgnjpCkc_plt0lqo',
    table_id='tab-M53nDu1y1SXMVA_ny97'
)

Column

To create column:

q.create_column(
    project_id='ctx-M53lgnjpCkc_plt0lqo',
    table_id='tab-M53nDu1y1SXMVA_ny97'
    display_name='company',
    data_type='text'
)

The data_type should be one of text, number, boolean, url, date, date_added, link_one, link_many.

The columnId is returned from the response. You need to know columnId when you update or delete column.

{
    "ok": true,
    "columnId": "col-1"
}

To create link_one or link_many column:

result = q.create_column(
    project_id='ctx-M53lgnjpCkc_plt0lqo',
    table_id='tab-M53nDu1y1SXMVA_ny97'
    display_name='manufacturer',
    data_type='link_one',
    linked_table='tab-dfj23eijSFdfewf'
)

To update column:

q.update_column(
    project_id='ctx-M53pLqASSUmxj5yU7LO',
    table_id='tab-M53pRIARajeAYTwPIAN',
    column_id='col-1',
    display_name='price',
    data_type='number',
)

To update column to link to table:

q.update_column(
    project_id='ctx-M53pLqASSUmxj5yU7LO',
    table_id='tab-M53pRIARajeAYTwPIAN',
    column_id='col-2',
    display_name='product',
    data_type='link_many',
    linked_table='tab-dfj23eijSFdfewf'
)

To delete column:

q.delete_column(
    project_id='ctx-M53pLqASSUmxj5yU7LO',
    table_id='tab-M53pRIARajeAYTwPIAN',
    column_id='col-1',
)

Vertex

To create vertex:

q.create_vertex(
    project_id='ctx-M53pLqASSUmxj5yU7LO',
    table_id='tab-M53pRIARajeAYTwPIAN',
    values={
        'col-0': 'Google',                  # text
        'col-1': 10.5,                      # number
        'col-2': False,                     # boolean
        'col-3': 'https://google.com',      # url
        'col-4': datetime.datetime.today()  # date
        # 'col-5'     # date_added
    },
    edges=[]
)

Note that you don't have to set date_added type column in creating vertex. Since the value is automatically set by today. Also column_ids and values lists should have same size, and each value in list should match with each data type of columns that you defined before.

The vertexId is returned from the result. You need to know vertexId when you update or delete vertex.

{
    "ok": true,
    "vertex": {
        "id": "row-M53tT_jTviJ50qyzgsL",
        "label": "tab-M53pRIARajeAYTwPIAN",
        "values": [
            {
                "key": "col-1",
                "value": "Google"
            },
            {
                "key": "col-2",
                "value": "Pixel 3A"
            }
        ],
        "contextId": null
    }
}

To create vertex linking to another table's vertex:

q.create_vertex(
    project_id='ctx-M57S8onUVXwdNMRgHPf',
    table_id='tab-M57wxxBqH0D7aKgYhhH',
    values={
        'col-0': 'Apple7',
        'col-1': True,
        'col-2': ''
    },
    edges=[
        ("column3", "row-M587jZETRpuCBIXUfw6"),
        ("column3", "row-M58Ac6n8CjPhqp8-u7M")
    ]
)

The edges is a list of tuples which includes Column Label and a vertexId that is linked to.

To update vertex:

q.update_vertex(
    project_id='ctx-M53pLqASSUmxj5yU7LO',
    table_id='tab-M53pRIARajeAYTwPIAN',
    vertex_id='row-M53tT_jTviJ50qyzgsL',
    values={
        'col-1': 'Google', 
        'col-2': 'Pixel 3'
    },
    edges= []
)

To create edge to existing vertex:

result = q.update_vertex(
    project_id='ctx-M57S8onUVXwdNMRgHPf',
    table_id='tab-M57wxxBqH0D7aKgYhhH',
    vertex_id='row-M588L2W5S9WZJ5t2Spn',
    values={
        'col-0': 'Google',
        'col-1': True,
        'col-2': ''
    },
    edges=[
        ("column3", "row-M587jZETRpuCBIXUfw6"),
        ("column3", "row-M58A84uaVMbCC16pSUA")
    ]
)

To delete vertex:

q.delete_vertex(
    project_id='ctx-M53pLqASSUmxj5yU7LO',
    table_id='tab-M53pRIARajeAYTwPIAN',
    vertex_id='row-M53tT_jTviJ50qyzgsL',
)

Schema

To get schema of project:

q.get_schema(
    project_id='ctx-M53lgnjpCkc_plt0lqo'
)

You can see tables and columns belonging to the project.

{
    "tables": [
        {
            "tableId": "tab-M53lgnpTXtQhMqMBXHH",
            "displayName": "Table 1",
            "columns": [],
            "config": null
        }
    ],
    "links": []
}

Graph

To get vertices:

q.get_graph(
    project_id='ctx-M53zulrvmCoNnP7PMEU',
    table_id='tab-M53zumOvT3xKmqkIB_X',
    filters=[],
    offset=1,
    limit=50
)

To add filtering:

q.get_graph(
    project_id='ctx-M53zulrvmCoNnP7PMEU',
    table_id='tab-M53zumOvT3xKmqkIB_X',
    filters=[
        {
            "property": "col-2",
            "predicate": "=",
            "value": "apple"
        }
    ],
    offset=1,
    limit=50
)

To add group by and aggregations:

q.summarize_graph(
    project_id='ctx-M53zulrvmCoNnP7PMEU',
    table_id='tab-M53zumOvT3xKmqkIB_X',
    filters=[
        {
            "property": "col-2",
            "predicate": "=",
            "value": "apple"
        }
    ],
    groups=[
        {
            "property": "col-2"
        }
    ],
    aggregations=[
        {
            "property": "col-1",
            "function": "sum"
        }
    ],
    offset=1,
    limit=50
)

The function should be one of count, sum, mean, max, min.

Bulk Upload

To upload csv file:

q.bulk_upload(
    project_id='ctx-M57S8onUVXwdNMRgHPf',
    table_id='tab-MK09gSNWBtZ-g2iLW8-',
    filepath='/Users/sangwonseo/Downloads/company1.csv',
    column_ids=['col-0', 'col-1', 'col-2'],
    configs={
        'countSkipRows': 0,
        'hasHeader': False,
        'dropEmpty': False
    }
)

Note that csv files should follow Amazon neptune csv format. For more details, please refer to the documentation. https://docs.aws.amazon.com/neptune/latest/userguide/bulk-load-tutorial-format-gremlin.html

For more details about Kgbase

Please visit https://kgbase.com/pages/learn

If you are interested in Kgbase demo

Please request demo at https://kgbase.com/pages/request-demo

If you have any questions

Please email at [e-mail]

License

MIT