hammad-python 0.0.20

No project description provided

hammad-python

Happily Accelerated Micro-Modules (for) Application Development

Introduction

The hammad-python library, is a mix of a love letter and collection of mixed resources for developing Python applications. This library is meant to be used for rapid prototyping and development, and is focused on providing styled placeholder tools for common patterns, tasks and workflows.

The package is currently built into the following structures:

• hammad-python : Contains most core functionality and resources.
• hammad-python[ai] : Contains easy to use resources for Generative AI related tasks such as generating completions with language models, or creating embeddings.
• hammad-python[serve] : Contains FastAPI / Uvicorn based resources for serving and running applications.

Installation

You can install the package using pip or uv:

pip install hammad-python

# or install the `ai` extension
# pip install 'hammad-python[ai]'

# or install the `serve` extension
# pip install 'hammad-python[serve]'

uv pip install hammad-python

# or install the `ai` extension
# uv pip install 'hammad-python[ai]'

# or install the `serve` extension
# uv pip install 'hammad-python[serve]'

Basic Usage

Data Structures, Databases and Other Data Related Resources

Collections

Using hammad.data.collections is a simple way to create searchable collections of data using both bm25 and vector based search.

from hammad.data.collections import create_collection

# Create either a `vector` or `searchable` collection
col = create_collection(type = "searchable")

# add anything to the collection
col.add("This is some text")
col.add(5)
col.add({'text' : "this is a dictionary"})

# search the collection
print(col.query("text"))

Databases

Any collection can be either used as a standalone database, or can be added as one of the collections within a database. Databases provide a unified interface for handling both Searchable and Vector based collections.

from hammad.data.collections import create_collection
from hammad.data.databases import Database

# Create either a `vector` or `searchable` collection
col = create_collection(type = "searchable")

col.add("This is some text")

# Databases can either be created on memory or using a path
db = Database(location = "memory")

db.add_collection(col)

# search globally or within a single collection
print(db.query("text"))

Styling / Introspection Resources

The hammad-python package contains a variety of components that can be used to easily style, or run introspection (logging) on your code.

from hammad.cli import print, input, animate

# Use extended `rich` styling easily
print("Hello, World", bg_settings = {"title" : "This is a title"})

# Easily collect various forms of input in a single function
class User(BaseModel):
    name : str
    age : int

# TIP:
# you can style this the same way with `print`
user = input("Enter some information about yourself: ", schema = User)

# easily run a collection of prebuilt animations
animate("This is a rainbow!", type = "rainbow", duration = 2, refresh_rate = 20)

Using the various hammad.logging resources, you can both create custom & styled loggers, as well as easily inspect various aspects of your code during runtime.

from hammad.logging import Logger

# create standard / file based loggers easily
logger = Logger("hammad", level = "info", rich = Trues)

file_logger = Logger("hammad-file", level = "info", file = "hammad.log")

# log to the console
logger.info("This is an info message")

# Use the various `trace_` methods to run various introspection tasks
from hammad.logging import (
   trace,
   trace_cls,
   trace_function,
   trace_http
)