How to use onetick.query with onetick.py

onetick.py is a wrapper around OneTick’s python interface library onetick.query.

But unlike onetick.query, onetick.py is developed and supported separately, so it doesn’t cover 100% of functionality of onetick.query.

Support for new OneTick functionality is being constantly added in the new releases of onetick.py, but in case of need it is possible to bypass some of the onetick.py interfaces to use onetick.query classes and functions directly.

In all examples below this code is used to import onetick.query library:

import onetick.query as otq

Implementing EP that is the source of ticks

Some OneTick event processors are the source of ticks and they have a special interface in onetick.py: onetick.py.Source.

Let’s use the simplest source of ticks otq.TickGenerator as an example. For already implemented and more powerful version see class onetick.py.Tick.

>>> data = otp.Source(otq.TickGenerator(fields='long A = 1'))
>>> data.schema
{}

When using onetick.query the user is expected to set the schema manually. That will let onetick.py internal logic know about the fields added or deleted from the tick schema by the onetick.query classes and functions.

>>> node = otq.TickGenerator(fields='long A = 1', bucket_interval=0, bucket_time='BUCKET_START')
>>> data = otp.Source(node, A=otp.long)
>>> data.schema
{'A': <class 'onetick.py.types.long'>}
>>> data['B'] = data['A'] + 1
>>> otp.run(data, symbols=f'{otp.config.default_db}::')
        Time  A  B
0 2003-12-01  1  2

Implementing EP that can be sinked

If OneTick’s event processor is not a source of ticks then it can be sinked.

We can use method onetick.py.Source.sink() to do this. Let’s use event processor otq.AddField as an example. For already implemented and more powerful version see method onetick.py.Source.__setitem__().

Do not forget to update the schema if needed.

>>> data = otp.Tick(A=1)
>>> data.sink(otq.AddField('B', '2'))
>>> data.schema['B'] = otp.long
>>> otp.run(data, symbols=f'{otp.config.default_db}::')
        Time  A  B
0 2003-12-01  1  2

Implementing OneTick built-in functions

Another thing that can be inserted directly is OneTick built-in functions. These functions are used in expressions when adding or updating fields. These functions do not have special representation in onetick.query, so we can just use strings.

Let’s use OneTick’s function REPLACE that replaces some substring in a string. For already implemented and more powerful version see method onetick.py.Operation.str.replace.

Special class otp.raw can be used to represent arbitrary OneTick expression. You only need to pass parameter dtype to let onetick.py internal logic know what is the returned type of this function and you will need to check that the types of arguments passed to this function are correct.

>>> data = otp.Tick(A='Hello world!')
>>> data['B'] = otp.raw('REPLACE(A, "Hello", "Hi")', dtype=otp.string[64])
>>> otp.run(data)
        Time             A          B
0 2003-12-01  Hello world!  Hi world!