PyPI Package: olostep | Requirements: Python 3.11+

Installation

pip install olostep

Authentication

Get your API key from the Olostep Dashboard.

Quick Start

The SDK provides two client options depending on your use case:

Sync Client (`Olostep`)

Best for: Scripts and simple use cases where you prefer blocking operations.

The sync client provides a simpler, blocking interface that’s easier to get started with if you’re new to async/await.

Async Client (`AsyncOlostep`)

Best for: Production applications, and handling many concurrent requests.

The async client provides non-blocking operations and is the recommended choice for production applications that need high throughput.

Sync Client (Olostep)

The sync client (Olostep) provides a blocking interface that’s perfect for scripts and simple use cases.

from olostep import Olostep

# Provide the API key either via passing in the 'api_key' parameter or
# by setting the OLOSTEP_API_KEY environment variable

# The sync client handles resource management automatically
# No explicit close needed - resources are cleaned up after each operation
client = Olostep(api_key="YOUR_REAL_KEY")
scrape_result = client.scrapes.create(url_to_scrape="https://example.com")

Basic Web Scraping

from olostep import Olostep

client = Olostep(api_key="your-api-key")

# Simple scraping
result = client.scrapes.create(url_to_scrape="https://example.com")
print(f"Scraped {len(result.html_content)} characters")

# Multiple formats
result = client.scrapes.create(
    url_to_scrape="https://example.com",
    formats=["html", "markdown"]
)
print(f"HTML: {len(result.html_content)} chars")
print(f"Markdown: {len(result.markdown_content)} chars")

Batch Processing

from olostep import Olostep

client = Olostep(api_key="your-api-key")

# Process multiple URLs efficiently
batch = client.batches.create(
    urls=[
        "https://www.google.com/search?q=python",
        "https://www.google.com/search?q=javascript",
        "https://www.google.com/search?q=typescript"
    ]
)

# Wait for completion and process results
for item in batch.items():
    content = item.retrieve(["html"])
    print(f"Processed {item.url}: {len(content.html_content)} bytes")

Smart Web Crawling

from olostep import Olostep

client = Olostep(api_key="your-api-key")

# Crawl with intelligent filtering
crawl = client.crawls.create(
    start_url="https://www.bbc.com",
    max_pages=100,
    include_urls=["/articles/**", "/blog/**"],
    exclude_urls=["/admin/**"]
)

for page in crawl.pages():
    content = page.retrieve(["html"])
    print(f"Crawled: {page.url}")

Site Mapping

from olostep import Olostep

client = Olostep(api_key="your-api-key")

# Extract all links from a website
maps = client.maps.create(url="https://example.com")

# Get all discovered URLs
urls = []
for url in maps.urls():
    urls.append(url)
    if len(urls) >= 10:  # Limit for demo
        break

print(f"Found {len(urls)} URLs")

AI-Powered Answers

from olostep import Olostep

client = Olostep(api_key="your-api-key")

# Get answers from web pages using AI
answer = client.answers.create(
    task="What is the main topic of https://example.com?"
)
print(f"Answer: {answer.answer}")

Async Client (AsyncOlostep)

The async client (AsyncOlostep) is the recommended client for high-performance applications, backend services, and when you need to handle many concurrent requests.

from olostep import AsyncOlostep

# Provide the API key either via passing in the 'api_key' parameter or
# by setting the OLOSTEP_API_KEY environment variable

# RESOURCE MANAGEMENT
# ===================
# The SDK supports two usage patterns for resource management:

# 1. Context Manager (Recommended for one-off usage):
#    Automatically handles resource cleanup
async with AsyncOlostep(api_key="YOUR_REAL_KEY") as client:
    scrape_result = await client.scrapes.create(url_to_scrape="https://example.com")
# Transport is automatically closed here

# 2. Explicit Close (For long-lived services):
#    Requires manual resource cleanup
client = AsyncOlostep(api_key="YOUR_REAL_KEY")
try:
    scrape_result = await client.scrapes.create(url_to_scrape="https://example.com")
finally:
    await client.close()  # Manually close the transport

Basic Web Scraping

import asyncio
from olostep import AsyncOlostep

async def main():
    async with AsyncOlostep(api_key="your-api-key") as client:
        # Simple scraping
        result = await client.scrapes.create(url_to_scrape="https://example.com")
        print(f"Scraped {len(result.html_content)} characters")

        # Multiple formats
        result = await client.scrapes.create(
            url_to_scrape="https://example.com",
            formats=["html", "markdown"]
        )
        print(f"HTML: {len(result.html_content)} chars")
        print(f"Markdown: {len(result.markdown_content)} chars")

asyncio.run(main())

Batch Processing

import asyncio
from olostep import AsyncOlostep

async def main():
    async with AsyncOlostep(api_key="your-api-key") as client:
        # Process multiple URLs efficiently
        batch = await client.batches.create(
            urls=[
                "https://www.google.com/search?q=python",
                "https://www.google.com/search?q=javascript",
                "https://www.google.com/search?q=typescript"
            ]
        )

        # Wait for completion and process results
        async for item in batch.items():
            content = await item.retrieve(["html"])
            print(f"Processed {item.url}: {len(content.html_content)} bytes")

asyncio.run(main())

Smart Web Crawling

import asyncio
from olostep import AsyncOlostep

async def main():
    async with AsyncOlostep(api_key="your-api-key") as client:
        # Crawl with intelligent filtering
        crawl = await client.crawls.create(
            start_url="https://www.bbc.com",
            max_pages=100,
            include_urls=["/articles/**", "/blog/**"],
            exclude_urls=["/admin/**"]
        )

        async for page in crawl.pages():
            content = await page.retrieve(["html"])
            print(f"Crawled: {page.url}")

asyncio.run(main())

Site Mapping

import asyncio
from olostep import AsyncOlostep

async def main():
    async with AsyncOlostep(api_key="your-api-key") as client:
        # Extract all links from a website
        maps = await client.maps.create(url="https://example.com")

        # Get all discovered URLs
        urls = []
        async for url in maps.urls():
            urls.append(url)
            if len(urls) >= 10:  # Limit for demo
                break

        print(f"Found {len(urls)} URLs")

asyncio.run(main())

AI-Powered Answers

import asyncio
from olostep import AsyncOlostep

async def main():
    async with AsyncOlostep(api_key="your-api-key") as client:
        # Get answers from web pages using AI
        answer = await client.answers.create(
            task="What is the main topic of https://example.com?"
        )
        print(f"Answer: {answer.answer}")

asyncio.run(main())

SDK Reference

Method Structure

Both SDK clients provide the same clean, pythonic interface organized into logical namespaces:

Namespace	Purpose	Key Methods
`scrapes`	Single URL extraction	`create()`, `get()`
`batches`	Multi-URL processing	`create()`, `info()`, `items()`
`crawls`	Website traversal	`create()`, `info()`, `pages()`
`maps`	Link extraction	`create()`, `urls()`
`answers`	AI-powered extraction	`create()`, `get()`
`retrieve`	Content retrieval	`get()`

Each operation returns stateful objects with ergonomic methods for follow-up operations.

Error Handling

Catch all SDK errors using the base exception class:

from olostep import Olostep, Olostep_BaseError

client = Olostep(api_key="your-api-key")

try:
    result = client.scrapes.create(url_to_scrape="https://example.com")
except Olostep_BaseError as e:
    print(f"Error has occurred: {type(e).__name__}")
    print(f"Error message: {e}")

For detailed error handling information, including the full exception hierarchy and granular error handling options, see Detailed Error Handling.

Automatic Retries

The SDK automatically retries on transient errors (network issues, temporary server problems) based on the RetryStrategy configuration. You can customize the retry behavior by passing a RetryStrategy instance when creating the client:

from olostep import Olostep, RetryStrategy

retry_strategy = RetryStrategy(
    max_retries=3,
    initial_delay=1.0,
    jitter_min=0.2,
    jitter_max=0.8
)

client = Olostep(api_key="your-api-key", retry_strategy=retry_strategy)
result = client.scrapes.create("https://example.com")

For detailed retry configuration options and best practices, see Retry Strategy.

Advanced Features

Smart Input Coercion

The SDK intelligently handles various input formats for maximum convenience:

from olostep import Olostep, Country

client = Olostep(api_key="your-api-key")

# Formats: string, list, or enum
client.scrapes.create(url_to_scrape="https://example.com", formats="html")
client.scrapes.create(url_to_scrape="https://example.com", formats=["html", "markdown"])

# Countries: case-insensitive strings or enums
client.scrapes.create(url_to_scrape="https://example.com", country="us")
client.scrapes.create(url_to_scrape="https://example.com", country=Country.US)

# Lists: single values or lists
client.batches.create(urls="https://example.com")    # Single URL
client.batches.create(urls=["https://a.com", "https://b.com"])  # Multiple URLs

Advanced Scraping Options

from olostep import Olostep, Format, Country, WaitAction, FillInputAction

client = Olostep(api_key="your-api-key")

# Full control over scraping behavior
result = client.scrapes.create(
    url_to_scrape="https://news.google.com/",
    wait_before_scraping=3000,
    formats=[Format.HTML, Format.MARKDOWN],
    remove_css_selectors=["script", ".popup"],
    actions=[
        WaitAction(milliseconds=1500),
        FillInputAction(selector="searchbox", value="olostep")
    ],
    parser="@olostep/google-news",
    country=Country.US,
    remove_images=True
)

Batch Processing with Custom IDs

from olostep import Olostep, Country

client = Olostep(api_key="your-api-key")

batch = client.batches.create([
    {"url": "https://www.google.com/search?q=python", "custom_id": "search_1"},
    {"url": "https://www.google.com/search?q=javascript", "custom_id": "search_2"},
    {"url": "https://www.google.com/search?q=typescript", "custom_id": "search_3"}
],
country=Country.US,
parser="@olostep/google-search"
)

# Process results by custom ID
# When using a parser, retrieve JSON content instead of HTML
for item in batch.items():
    if item.custom_id == "search_2":
        content = item.retrieve(["json"])
        print(f"Search result: {content.json_content}")

Intelligent Crawling

from olostep import Olostep

client = Olostep(api_key="your-api-key")

# Crawl with intelligent filtering
crawl = client.crawls.create(
    start_url="https://www.bbc.com",
    max_pages=1000,
    max_depth=3,
    include_urls=["/articles/**", "/news/**"],
    exclude_urls=["/ads/**", "/tracking/**"],
    include_external=False,
    include_subdomain=True,
)

for page in crawl.pages():
    content = page.retrieve(["html"])
    print(f"Crawled: {page.url}")

Site Mapping with Filters

from olostep import Olostep

client = Olostep(api_key="your-api-key")

# Extract all links with advanced filtering
maps = client.maps.create(
    url="https://www.bbc.com",
    include_subdomain=True,
    include_urls=["/articles/**", "/news/**"],
    exclude_urls=["/ads/**", "/tracking/**"]
)

# Get filtered URLs
urls = []
for url in maps.urls():
    urls.append(url)

print(f"Found {len(urls)} relevant URLs")

Answers Retrieval

from olostep import Olostep

client = Olostep(api_key="your-api-key")

# First create an answer
created_answer = client.answers.create(
    task="What is the main topic of https://example.com?"
)

# Then retrieve it using the ID
answer = client.answers.get(answer_id=created_answer.id)
print(f"Answer: {answer.answer}")

Content Retrieval

from olostep import Olostep

client = Olostep(api_key="your-api-key")

# Get content by retrieve ID
result = client.retrieve.get(retrieve_id="ret_123")

# Get multiple formats
result = client.retrieve.get(retrieve_id="ret_123", formats=["html", "markdown", "text", "json"])

Logging

Enable logging to debug issues:

import logging

logging.basicConfig(level=logging.INFO)
logger = logging.getLogger("olostep")
logger.setLevel(logging.INFO)  # Use DEBUG for verbose output

Log Levels: INFO (recommended), DEBUG (verbose), WARNING, ERROR

Retry Strategy Configuration

The RetryStrategy class controls how the Olostep SDK handles transient API errors through automatic retries with exponential backoff and jitter. This helps ensure reliable operation in production environments where temporary network issues, rate limits, and server overload can cause intermittent failures.

Default Behavior

By default, the SDK uses the following retry configuration:

Max retries: 5 attempts
Initial delay: 2 seconds
Backoff: Exponential (2^attempt)
Jitter: 10-90% of delay (randomized)

This means:

Attempt 1: Immediate
Attempt 2: ~2-3.6s delay
Attempt 3: ~4-7.2s delay
Attempt 4: ~8-14.4s delay
Attempt 5: ~16-28.8s delay

Maximum duration: ~57 seconds for all retries (worst case)

Custom Configuration

from olostep import AsyncOlostep, RetryStrategy

# Create custom retry strategy
retry_strategy = RetryStrategy(
    max_retries=3,
    initial_delay=1.0,
    jitter_min=0.2,  # 20% minimum jitter
    jitter_max=0.8,  # 80% maximum jitter
)

# Use with client
async with AsyncOlostep(
    api_key="your-api-key",
    retry_strategy=retry_strategy
) as client:
    result = await client.scrapes.create("https://example.com")

When Retries Happen

The SDK automatically retries on:

Temporary server issues (OlostepServerError_TemporaryIssue)
Timeout responses (OlostepServerError_NoResultInResponse)

Other errors (authentication, validation, resource not found, etc.) fail immediately without retry.

Transport vs Caller Retries

The SDK has two retry layers:

Transport layer: Handles network-level connection failures (DNS, timeouts, etc.)
Caller layer: Handles API-level transient errors (controlled by RetryStrategy)

Both layers are independent and have separate configuration. The total maximum duration is the sum of both layers.

Calculating Max Duration

retry_strategy = RetryStrategy(max_retries=5, initial_delay=2.0)
max_duration = retry_strategy.max_duration()
print(f"Max call duration: {max_duration:.2f}s")

Configuration Examples

Here are some examples of how to configure the retry strategy for different use cases.

Conservative Strategy

# Fewer retries, shorter delays
retry_strategy = RetryStrategy(
    max_retries=3,
    initial_delay=1.0,
    jitter_min=0.2,
    jitter_max=0.8
)
# Max duration: ~12.6s

Aggressive Strategy

# More retries for critical operations
retry_strategy = RetryStrategy(
    max_retries=10,
    initial_delay=0.5
)
# Max duration: ~969.75s

No Retries (Fail Fast)

# Disable retries for immediate failure feedback
retry_strategy = RetryStrategy(max_retries=0)

client = AsyncOlostep(api_key="your-api-key", retry_strategy=retry_strategy)

High-Throughput Strategy

# Optimized for high-volume operations
retry_strategy = RetryStrategy(
    max_retries=2,
    initial_delay=0.5,
    jitter_min=0.1,
    jitter_max=0.3  # Lower jitter for more predictable timing
)
# Max duration: ~1.95s

Understanding Jitter

Jitter adds randomization to prevent “thundering herd” problems when many clients retry simultaneously. The jitter is calculated as:

base_delay = initial_delay * (2 ** attempt)
jitter_range = base_delay * (jitter_max - jitter_min)
jitter = random.uniform(base_delay * jitter_min, base_delay * jitter_min + jitter_range)
final_delay = base_delay + jitter

For example, with initial_delay=2.0, jitter_min=0.1, jitter_max=0.9:

Attempt 0: base=2.0s, jitter=0.2-1.8s, final=2.2-3.8s
Attempt 1: base=4.0s, jitter=0.4-3.6s, final=4.4-7.6s
Attempt 2: base=8.0s, jitter=0.8-7.2s, final=8.8-15.2s

Best Practices

For Production Applications

# Balanced approach for production
retry_strategy = RetryStrategy(
    max_retries=5,
    initial_delay=2.0,
    jitter_min=0.1,
    jitter_max=0.9
)

For Development/Testing

# Fast feedback for development
retry_strategy = RetryStrategy(
    max_retries=2,
    initial_delay=0.5,
    jitter_min=0.1,
    jitter_max=0.3
)

For Batch Operations

# Conservative for large batch jobs
retry_strategy = RetryStrategy(
    max_retries=3,
    initial_delay=1.0,
    jitter_min=0.2,
    jitter_max=0.8
)

Monitoring and Debugging

The SDK logs retry information at the DEBUG level:

DEBUG: Temporary issue, retrying in 2.34s
DEBUG: No result in response, retrying in 4.67s

Enable debug logging to monitor retry behavior:

import logging
logging.getLogger("olostep").setLevel(logging.DEBUG)

Error Handling

When all retries are exhausted, the original error is raised:

try:
    result = await client.scrapes.create("https://example.com")
except OlostepServerError_TemporaryIssue as e:
    print(f"Failed after all retries: {e}")
    # Handle the permanent failure

Performance Considerations

Memory: Each retry attempt uses additional memory for request/response objects
Time: Total operation time can be significantly longer with retries enabled
API Limits: Retries count against your API usage limits
Network: More network traffic due to retry attempts

Choose your retry strategy based on your application’s requirements for reliability vs. performance.

Detailed Error Handling

Exception Hierarchy

The Olostep SDK provides a comprehensive exception hierarchy for different failure scenarios. All exceptions inherit from Olostep_BaseError. There are three main error types that directly inherit from Olostep_BaseError:

Olostep_APIConnectionError - Network-level connection failures
OlostepServerError_BaseError - Errors raised (sort of) by the API server
OlostepClientError_BaseError - Errors raised by the client SDK

Why Connection Errors Are Separate

Olostep_APIConnectionError is separate from server errors because it represents network-level failures that occur before the API can process the request. These are transport layer issues (DNS or HTTP failures, timeouts, connection refused, etc.) rather than API-level errors. HTTP status codes (4xx, 5xx) are considered API responses and are categorized as server errors, even though they indicate problems.

Olostep_BaseError
├── Olostep_APIConnectionError
├── OlostepServerError_BaseError
│   ├── OlostepServerError_TemporaryIssue
│   │   ├── OlostepServerError_NetworkBusy
│   │   └── OlostepServerError_InternalNetworkIssue
│   ├── OlostepServerError_RequestUnprocessable
│   │   ├── OlostepServerError_ParserNotFound
│   │   └── OlostepServerError_OutOfResources
│   ├── OlostepServerError_BlacklistedDomain
│   ├── OlostepServerError_FeatureApprovalRequired
│   ├── OlostepServerError_AuthFailed
│   ├── OlostepServerError_CreditsExhausted
│   ├── OlostepServerError_InvalidEndpointCalled
│   ├── OlostepServerError_ResourceNotFound
│   ├── OlostepServerError_NoResultInResponse
│   └── OlostepServerError_UnknownIssue
└── OlostepClientError_BaseError
    ├── OlostepClientError_RequestValidationFailed
    ├── OlostepClientError_ResponseValidationFailed
    ├── OlostepClientError_NoAPIKey
    ├── OlostepClientError_AsyncContext
    ├── OlostepClientError_BetaFeatureAccessRequired
    └── OlostepClientError_Timeout

Recommended Error Handling

For most use cases, catch the base error and print the error name:

from olostep import AsyncOlostep, Olostep_BaseError

try:
    result = await client.scrapes.create(url_to_scrape="https://example.com")
except Olostep_BaseError as e:
    print(f"Error has occurred: {type(e).__name__}")
    print(f"Error message: {e}")

This approach catches all SDK errors and provides clear information about what went wrong. The error name (e.g., OlostepServerError_AuthFailed) is descriptive enough to understand the issue.

Granular Error Handling

If you need more specific error handling, catch the specific error types directly. Avoid using OlostepServerError_BaseError or OlostepClientError_BaseError - these base classes only indicate who raised the error (server vs client), not who’s responsible for fixing it. This is an implementation detail that doesn’t help with error handling logic. Instead, catch specific error types that indicate the actual problem:

from olostep import (
    AsyncOlostep,
    Olostep_BaseError,
    Olostep_APIConnectionError,
    OlostepServerError_AuthFailed,
    OlostepServerError_CreditsExhausted,
    OlostepClientError_NoAPIKey,
)

try:
    result = await client.scrapes.create(url_to_scrape="https://example.com")
except Olostep_APIConnectionError as e:
    print(f"Network error: {type(e).__name__}")
except OlostepServerError_AuthFailed:
    print("Invalid API key")
except OlostepServerError_CreditsExhausted:
    print("Credits exhausted")
except OlostepClientError_NoAPIKey:
    print("API key not provided")
except Olostep_BaseError as e:
    print(f"Error has occurred: {type(e).__name__}")

Configuration

Environment Variables

Variable	Description	Default
`OLOSTEP_API_KEY`	Your API key	Required
`OLOSTEP_BASE_API_URL`	API base URL	`https://api.olostep.com/v1`
`OLOSTEP_API_TIMEOUT`	Request timeout (seconds)	`150`

SDKs

​Installation

​Authentication

​Quick Start

Sync Client (`Olostep`)

Async Client (`AsyncOlostep`)

​Sync Client (Olostep)

​Basic Web Scraping

​Batch Processing

​Smart Web Crawling

​Site Mapping

​AI-Powered Answers

​Async Client (AsyncOlostep)

​Basic Web Scraping

​Batch Processing

​Smart Web Crawling

​Site Mapping

​AI-Powered Answers

​SDK Reference

​Method Structure

​Error Handling

​Automatic Retries

​Advanced Features

​Smart Input Coercion

​Advanced Scraping Options

​Batch Processing with Custom IDs

​Intelligent Crawling

​Site Mapping with Filters

​Answers Retrieval

​Content Retrieval

​Logging

​Retry Strategy Configuration

​Default Behavior

​Custom Configuration

​When Retries Happen

​Transport vs Caller Retries

​Calculating Max Duration

​Configuration Examples

​Conservative Strategy

​Aggressive Strategy

​No Retries (Fail Fast)

​High-Throughput Strategy

​Understanding Jitter

​Best Practices

​For Production Applications

​For Development/Testing

​For Batch Operations

​Monitoring and Debugging

​Error Handling

​Performance Considerations

​Detailed Error Handling

​Exception Hierarchy

​Why Connection Errors Are Separate

​Recommended Error Handling

​Granular Error Handling

​Configuration

​Environment Variables

​Getting Help

​Resources

PyPI Package

Get API Key

Installation

Authentication

Quick Start

Sync Client (Olostep)

Basic Web Scraping

Batch Processing

Smart Web Crawling

Site Mapping

AI-Powered Answers

Async Client (AsyncOlostep)

Basic Web Scraping

Batch Processing

Smart Web Crawling

Site Mapping

AI-Powered Answers

SDK Reference

Method Structure

Error Handling

Automatic Retries

Advanced Features

Smart Input Coercion

Advanced Scraping Options

Batch Processing with Custom IDs

Intelligent Crawling

Site Mapping with Filters

Answers Retrieval

Content Retrieval

Logging

Retry Strategy Configuration

Default Behavior

Custom Configuration

When Retries Happen

Transport vs Caller Retries

Calculating Max Duration

Configuration Examples

Conservative Strategy

Aggressive Strategy

No Retries (Fail Fast)

High-Throughput Strategy

Understanding Jitter

Best Practices

For Production Applications

For Development/Testing

For Batch Operations

Monitoring and Debugging

Error Handling

Performance Considerations

Detailed Error Handling

Exception Hierarchy

Why Connection Errors Are Separate

Recommended Error Handling

Granular Error Handling

Configuration

Environment Variables

Getting Help

Resources