Skip to main content

SQL Generator and Query Result Retriever

Project description

TextualOrm

TextualOrm

This tool generates SQL queries from natural language and retrieves query results. This orm generates sql queries from your input and the specifics of your connected database and can also run them to retrieve records.

Find the code here

Currently, this ORM only supports retrieval queries. Queries that perform delete, create, or other CRUD (Create, Read, Update, Delete) actions will not be executed on your database..

The core of this application is the use of Large Language Models to generate the sql queries. This orm currently supports two LLMs:

  1. SQL Generator LLM - (free and the default llm used).
  2. OpenAI (requires subscription to OpenAI)

You can find more information about the default SQL Generator LLM by going to the link here. This has been fine-trained from the flan-t5-base model.

Requirements

  • Python
  • Postgres
  • Redis Note: Support for additional databases such as MySQL will be added in future updates.

Installation

pip install textual_orm

Usage Instructions

  1. Initialize the Orm

    from textual_orm import TextualOrm
    textual_orm = TextualOrm(connection_string="postgresql://user:password@host:port/db_name",
                llm_type=LLMType.DEFAULT, redis_host="localhost",
                redis_port=6379)
    

    To use the OpenAI implementation, add your api-key to the arguments:

    from textual_orm import TextualOrm
    textual_orm = TextualOrm(connection_string="postgresql://user:password@host:port/db_name",
                llm_type=LLMType.DEFAULT, redis_host="localhost",
                redis_port=6379, api-key="")
    
    
  2. Call the setup method: await textual_orm.setup()

  3. Generate the SQL query

    sql_query = await textual_orm.make_sql_request("List of settings", ["setting"])
    print(sql_query)
    

    This method takes three arguments:

    • question: Input question
    • tables: List of tables as reference
    • request_data: A boolean to indicate if it should query the database or return only the sql_query (default value is False)

    By default, the make_sql_request does not actually query the database. It returns back the generated sql query which you can look at and verify. To get an alternative sql query, please modify your input question.

    To query the database with the generated sql query. Call the method passing in request_data=True. This will return a response in this format:

    {
      'query': 'SELECT * FROM setting ORDER BY created_at DESC LIMIT 5;',
      'data': [<records>]
    }
    

    Where data is a list of records from the query response. data will be None if request_data=False as in the default case.

Note that first time run may take a little time.

For better performance, speed and caching, redis is required.

Additional Arguments

This orm uses a default postgres max pool of 10. You can modify it if needed by passing your value to the max_pool argument.

Below is a list of other supported arguments to Orm:

  • min_size=1 minimum size of pool
  • max_size=10 maxiumum size of pool
  • api_key="" api key for the given llm

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

textual_orm-0.1.3.tar.gz (10.0 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

textual_orm-0.1.3-py3-none-any.whl (11.9 kB view details)

Uploaded Python 3

File details

Details for the file textual_orm-0.1.3.tar.gz.

File metadata

  • Download URL: textual_orm-0.1.3.tar.gz
  • Upload date:
  • Size: 10.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.12.4

File hashes

Hashes for textual_orm-0.1.3.tar.gz
Algorithm Hash digest
SHA256 7464a75ad9f2bf2a9e640ae2d6301f0f0d9bc65659e77381c65339c4edc678f4
MD5 f0ea0ae0d19b471d2829e840ca0cc861
BLAKE2b-256 b40627b65a97d6cf965ee17113730d3b98c5825df49ad328971d4a3c01c0be9c

See more details on using hashes here.

File details

Details for the file textual_orm-0.1.3-py3-none-any.whl.

File metadata

  • Download URL: textual_orm-0.1.3-py3-none-any.whl
  • Upload date:
  • Size: 11.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.12.4

File hashes

Hashes for textual_orm-0.1.3-py3-none-any.whl
Algorithm Hash digest
SHA256 25bb11ed18bfa90f1d452581d9761efe6ef514c7248d84262d8d522422d999fe
MD5 29e0cc9a328008c1e219da3f253b8b85
BLAKE2b-256 77d19dd5a0ea61971f7b7b1c57a112792001d1c74469b72be2b30a4d50fddaf5

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page