Skip to content

Python Querying API.

The following documentation defines a Python API to be used when querying the kdb Insights and kdb Insights Enterprise Offerings.

Classes:

  • Query – Python API for querying kdb Insights and kdb Insights Enterprise.

Query

Bases: ApiClient

Python API for querying kdb Insights and kdb Insights Enterprise.

See Connection Parameters for shared constructor arguments.

Attributes:

Functions:

fetch_custom_apis

fetch_custom_apis()

Query the database for custom APIs and attach them as additional attributes.

get_data

get_data(table, start_time=lambda: datetime.now(tz=(timezone.utc)) - timedelta(minutes=15), end_time=lambda: datetime.now(tz=(timezone.utc)), *, input_timezone=None, output_timezone=None, filter=None, group_by=None, aggregations=None, fill=None, temporality=None, slice=None, sort_columns=None, labels=None, **kwargs)

Run a getData.

API call on kdb Insights Enterprise.

Parameters:

  • table (str) – The name of the table from which to retrieve data
  • start_time (str | datetime | None) – The earliest time data is to be queried from, by default this is 15 minutes prior to the time that the get_data call is made.
  • end_time (str | datetime | None) – The latest time data is to be queried from, by default this is the time that the get_data function is called.
  • input_timezone (Optional[str]) – The timezones of start_time and end_time, default is UTC if not specified.
  • output_timezone (Optional[str]) – The timezone of output timestamp columns, default is UTC if not specified.
  • filter (Optional[List[list]]) – A list of triadic Python lists defining filter conditions in the format [['function', 'column name', 'parameter'], ... ]
  • group_by (Optional[str]) – A list of Python str items defining the columns to group aggregations by
  • aggregations (Optional[List[list]]) – A list of Python str items or a list of lists of str items defining. The columns to be returned from the table across DAPs if defined as a list of str. If provided as a list of triadic Python lists as follows [['assignName', 'agg', 'column'], ...] then assignName defines the name to be assigned to the return of the aggregation, agg defines the aggregation function to be used and column defines the column on which that aggregation is applied.
  • fill (Optional[str]) – This defines how to handle the presence of null values. This should be supplied as a single str item with value forward to define the use of the last non-null value in a column or zero to replace all null values with the 0 value of that column.
  • temporality (Optional[str]) – This determines the way that the start_time and end_time arguments are interpreted. The two support input types are 'slice' and 'snapshot'.
  • slice (Optional[str]) – This should be defined as a Python list of 'times' which denote the time range to grab between each date in the start_time and end_time range when a temporality of 'slice' is used.
  • sort_columns (Optional[List[str]]) – This should be a Python list of str type objects denoting the columns which are to be sorted on.
  • labels (Optional[Dict[str, str]]) – labels defined in the assembly file of the DAPs can be used to only select data from DAPs in assemblies matching that label.
  • *kwargs* – Other possible arguments.

Returns:

  • Table – A pykx.Table or pykx.KeyedTable returned from a query to kdb Insights Enterprise

Examples:

Connect to a running instance of kdb Insights and query using get_data with default values

import kxi.query
conn = kxi.query.Query()
conn.get_data('my_table')

Query the instance providing arguments to various get_data calls.

conn.get_data('my_table', group_by = ['sensorID', 'qual'])
conn.get_data('my_table', filter = [['within', 'qual', [0, 2]]])
conn.get_data('my_table', start_time='2000.05.26',end_time='2000.05.27')

get_meta

get_meta()

Run a getMeta.

API call on kdb Insights Enterprise.

Returns:

  • Dictionary – A pykx.Dictionary containing metadata information about the various processes and schemas which describe a database.

Examples:

Connect to a running instance of kdb Insights and retrieve metadata information about the database, schemas and assemblies

import kxi.query
conn = kxi.query.Query()
conn.get_meta()

get_table_schemas

get_table_schemas(assembly=None)

Get table schemas in the format of {"table": {"col": pykxtype}}.

Parameters:

  • assembly (str | None) – Optional assembly name to filter for.

Returns:

Example 
query.get_table_schemas()
{
    "tableName": {
        "col1": kxi.util.QType(-12, pykx.TimestampAtom, "timestamp", "p")
    }
}

ping

ping(labels=None)

Run a ping.

This API is intended to run on a data access process and returns true if it reaches a target

Parameters:

  • labels (Optional[dict]) – labels defined in the assembly file of the DAPs can be used to only select data from DAPs in assemblies matching that label.

Returns:

preview

preview(table, start_time=None, end_time=None, limit=None)

Run a preview.

The preview API is a lightweight data retrieval API which is able to fetch small samples of a table using minimal time and resources, for you to review or to use in testing software compatibility with a schema.

Parameters:

  • table (str) – The name of the table from which to retrieve data
  • start_time (str | datetime | None) – The earliest time data is to be queried from, by default this is 15 minutes prior to the time that the get_data call is made.
  • end_time (str | datetime | None) – The latest time data is to be queried from, by default this is the time that the get_data function is called.
  • limit (int | None) – The maximum number of rows to return.

qsql

qsql(query, assembly, target, use_qe=False)

Run a qSQL.

API call on kdb Insights Enterprise.

Parameters:

  • query (str) – The SQL query which is to be run against the query environment.
  • assembly (str) – The Assembly name to query, must support query environment.
  • target (str) – The DAP instance to query.
  • use_qe (bool) – Indicate whether to use the QE or PROD environment.

Returns:

  • K – A pykx.K returned from a query to kdb Insights Enterprise

Examples:

Connect to a running instance of kdb Insights and query using qsql

import kxi.query
conn = kxi.query.Query()
conn.qsql('my_table', "assembly-qe", "rdb")

refresh_udas

refresh_udas()

Re-fetch UDAs from getMeta and return the updated namespace.

Call this if the deployment has changed since the last access.

Returns:

  • 'UdaNamespace' – The refreshed UdaNamespace.
Example 
conn.refresh_udas()
conn.udas.example.daAPI(table="trade", column="price")

sql

sql(query, use_qe=True, **kwargs)

Run an SQL.

API call on kdb Insights Enterprise.

Parameters:

  • query (str) – The SQL query which is to be run against the query environment
  • use_qe (bool) – Indicate whether to use the QE or PROD environment.
  • *kwargs* – Additional parameters.

Returns:

  • Table – A pykx.Table or pykx.KeyedTable returned from a query to kdb Insights Enterprise

Examples:

Connect to a running instance of kdb Insights and query using sql

import kxi.query
conn = kxi.query.Query()
conn.sql('SELECT * FROM my_table')

udas

udas: 'UdaNamespace'

Lazy UDA namespace — tab-complete across q namespaces.

Back to top