Makra SDK for Python

The official Python SDK for Makra - web scraping and data extraction made simple.

Installation

pip install makra

Quick Start

from makra import Makra

makra = Makra(api_key="your-api-key")

# Extract data with a query
result = await makra.extract(
    urls=["https://example.com"],
    query="Get all product titles and prices"
)

Features

• Simple API: Easy-to-use async interface for web scraping
• Schema-based Extraction: Define JSON schemas to extract structured data
• Query-based Extraction: Use natural language to describe what to extract
• Multiple Output Formats: Get content as markdown or plain text
• Multiple Providers: Support for various LLM providers (Groq, Google AI)
• Type-safe: Full type hints and Pydantic models
• Crawling: Discover and extract links from pages
• Page Mapping: Get structured maps of page content

API Reference

Initialize Client

from makra import Makra

makra = Makra(api_key="your-api-key")

With custom base URL:

makra = Makra(
    api_key="your-api-key",
    base_url="https://custom-api.example.com",
    api_version="v1"
)

Extract Data

Using a JSON schema:

schema = {
    "type": "object",
    "properties": {
        "title": {"type": "string"},
        "price": {"type": "number"}
    }
}

result = await makra.extract(
    urls=["https://example.com/products"],
    schema=schema
)

Using a natural language query:

result = await makra.extract(
    urls=["https://example.com/products"],
    query="Get all product titles and prices"
)

With actions (navigation, pagination):

from makra import enums

result = await makra.extract(
    urls=["https://example.com"],
    schema=schema,
    actions=[enums.ACTION.NAVIGATION, enums.ACTION.PAGINATION]
)

With extract options:

from makra import types

options = types.ExtractOptions(
    disable_cache=True,
    disable_pagination=False
)

result = await makra.extract(
    urls=["https://example.com"],
    query="Get products",
    options=options
)

Format Content

As markdown:

content = await makra.format(
    urls=["https://example.com"],
    type="markdown"
)

As text:

content = await makra.format(
    urls=["https://example.com"],
    type="text"
)

Using enum:

from makra import enums

content = await makra.format(
    urls=["https://example.com"],
    type=enums.FORMAT_TYPE.MARKDOWN
)

Crawl URLs

links = await makra.crawl(urls=["https://example.com"])

With options:

from makra import types

options = types.CrawlOptions()
links = await makra.crawl(
    urls=["https://example.com"],
    options=options
)

Page Map

page_map = await makra.page_map(urls=["https://example.com"])

Pre-process

result = await makra.pre_process(urls=["https://example.com"])

With options:

from makra import types

options = types.PreprocessOptions()
result = await makra.pre_process(
    urls=["https://example.com"],
    options=options
)

Ping API

status = await makra.ping()

Using Providers

With Groq:

from makra import Makra, providers

groq_provider = providers.Groq(api_key="your-groq-api-key")

makra = Makra(
    api_key="your-makra-api-key",
    provider=groq_provider
)

With Google AI:

from makra import Makra, providers

google_provider = providers.GoogleAI(api_key="your-google-api-key")

makra = Makra(
    api_key="your-makra-api-key",
    provider=google_provider
)

Custom model:

groq_provider = providers.Groq(
    api_key="your-groq-api-key",
    model="llama-3-70b-8192"
)

Types and Enums

Import types:

from makra import types

options = types.ExtractOptions(disable_cache=True)

Import enums:

from makra import enums

# Format types
enums.FORMAT_TYPE.MARKDOWN
enums.FORMAT_TYPE.TEXT

# Actions
enums.ACTION.NAVIGATION
enums.ACTION.PAGINATION

Error Handling

from makra.errors import RequestError

try:
    result = await makra.extract(urls=["https://example.com"], query="...")
except RequestError as e:
    print(f"Error {e.status_code}: {e.message}")

Requirements

• Python 3.12+
• An active Makra API key

License

MIT License - see LICENSE for details.

Links

• Documentation
• Homepage
• GitHub Repository