typedb-query-builder

typedb-query-builder💪

Description

It's a simple and lightweight python library for creating TypeDB queries in an Object-Oriented manner, which abstracts string preprocessing from the user.

The main advantages of this solutions are:

• removing boilerplate repetitive code and string preprocessing with very readable API for building TypeDB queries
• easy integration with any changes of typedb-client API. As a matter of fact, these changes should be handled by typedb-query-builder, so that you don't have to worry about your code
• removing unnecessary issues related with typos that appeared during typing verbose queries

Usage

The main compoenent of this packages is typedb_query_builder.typedb_query_builder.TypeDBQueryBuilder.

It provides you the following methods:

• match_entity(entity_name: str, variable_name: str) - used for adding and an entity in match queries.
• match_relationship(relationship_name: str, variable_name: str)) - used for adding and a relationship in match queries.
• insert_entity(entity_name: str, variable_name: str) - used for adding and entity in insert queries.
• insert_relationship(relationship_name: str, variable_name: str)) - used for adding a relationship in match queries.

entity_name and relationship_name are the actual names of the concepts in schema, while variable_name is just a variable that you want to assign to it - in the same way as you would assign a concept to a variable in a TypeDB query (note that within one query, one TypeDBQueryBuilder instance, the same variable name cannot be used for several concepts).

Each of these functions return an object of one of the following types: TypeDBEntityQuery, TypeDBRelationshipQuery. You can use these object to assign certain information to them, specifically:

TypeDBEntityQuery and TypeDBRelationshipQuery have methods:

• has(attribute_name: str, attribute_value: str, attribute_type: str = 'string') - which assign an attribute to an Entity or Relationship. The attribute type is provided by attribute_name, it's actual value by attribute_value. Optionally you can specify a type of the attribute according to it's value in a schema, so that the TypeDBQueryBuilder will be able to create the right query which will be in line with a given schema. If not provided TypeDBQueryBuilder will assume that the attribute is of type string.

Additionally, TypeDBRelationshipQuery has method for adding related objects:

• relates(role: str, thing: Union[TypeDBEntityQuery, "TypeDBRelationshipQuery"]) - which allows you to add a Thing (Entity/Relathinship) to a given relation under a certain role provided by role argument. thing specifies what concept should be added to a relationship. It has to be already instantiated object of type TypeDBEntityQuery or TypeDBRelationshipQuery.

Example

from typedb_query_builder.typedb_query_builder import TypeDBQueryBuilder

tqb = TypeDBQueryBuilder()

p1 = tqb.match_entity('protein', 'p1')    # Add an entity of type 'protein' to a match statement with a variable 'p1'.
p1.has('protein_name', 'ACE')             # Assign an attribute of type 'protein_name' with value 'ACE'
p1.has('protein_id', 'Q1')

p2 = tqb.match_entity('protein', 'p2')    # Add a second entity of type 'protein' to a match statement with a variable 'p2'.
p2.has('protein_name', 'ACE2')            # Assign an attribute of type 'protein_name' with value 'ACE2'
p2.has('protein_id', 'Q2')
p2.has('external-id', 1, 'double')        # This entity has an attribute 'external-id' which is of type double.

pi1 = tqb.insert_relationship('protein_interaction', 'pi')  # Add a relationship of type 'protein_interaction' to insert query.
pi1.relates('associated_protein', p1)                       # Add related entities
pi1.relates('associated_protein', p2)
pi1.has('pi_id', 'PI1')                                     # Add an attribute to a relationship

tqb.compile_query()                                         # Compile query
query = tqb.get_query()                                     # Get query

print(query)

Result

match $p1 isa protein, has protein_name "ACE", has protein_id "Q1"; $p2 isa protein, has protein_name "ACE2", has protein_id "Q2", has external-id 1; insert $pi (associated_protein: $p1, associated_protein: $p2) isa protein_interaction, has pi_id "PI1";