Skip to main content

Pagination

The SDK uses cursor-based pagination for list operations. This page covers how to work with paginated results effectively.

Basic Pagination​

All list() methods return a paginated response:

from credoai import CredoAI

client = CredoAI()

# Get the first page
response = client.use_cases.list(page_limit=10)

print(f"Items: {len(response.items)}")
print(f"Has more: {response.pagination.has_more}")
print(f"Next cursor: {response.pagination.next_cursor}")

Pagination Parameters​

ParameterTypeDescription
page_limitintNumber of items per page (default: 50, max: 100)
page_afterstrCursor for the next page (from pagination.next_cursor)

Manual Pagination​

Iterate through pages manually using the cursor:

all_use_cases = []
cursor = None

while True:
    response = client.use_cases.list(
        page_limit=50,
        page_after=cursor
    )

    all_use_cases.extend(response.items)

    if not response.pagination.has_more:
        break

    cursor = response.pagination.next_cursor

print(f"Total use cases: {len(all_use_cases)}")

Automatic Pagination with list_all()​

The list_all() method handles pagination automatically and returns a generator:

# Iterate through all items
for use_case in client.use_cases.list_all(page_size=50):
    print(f"  - {use_case.name}")

Collecting All Items​

# Collect all items into a list
all_use_cases = list(client.use_cases.list_all(page_size=50))
print(f"Total: {len(all_use_cases)}")

Memory-Efficient Processing​

Using list_all() as a generator is memory-efficient for large datasets:

# Process items one at a time without loading all into memory
for use_case in client.use_cases.list_all(page_size=100):
    process_use_case(use_case)

Async Pagination​

The async client supports async iteration:

import asyncio
from credoai import AsyncCredoAI

async def main():
    async with AsyncCredoAI() as client:
        # Async iteration
        async for use_case in client.use_cases.list_all(page_size=50):
            print(f"  - {use_case.name}")

        # Or collect all
        all_cases = [uc async for uc in client.use_cases.list_all()]
        print(f"Total: {len(all_cases)}")

asyncio.run(main())

Pagination for Relationships​

Relationship resources also support pagination:

# List all models for a use case
for model in client.use_case_models.list_all(use_case_id="uc_abc123"):
    print(f"  - {model.name}")

# List all vendors for a model
for vendor in client.model_vendors.list_all(model_id="model_xyz789"):
    print(f"  - {vendor.name}")

Best Practices​

Choose an Appropriate Page Size​

# For quick overviews, use smaller pages
response = client.use_cases.list(page_limit=10)

# For bulk operations, use larger pages
for uc in client.use_cases.list_all(page_size=100):
    process(uc)

Handle Empty Results​

response = client.use_cases.list()

if not response.items:
    print("No use cases found")
else:
    for uc in response.items:
        print(uc.name)

Early Exit​

Break out of iteration when you've found what you need:

target_name = "Production Model"

for model in client.models.list_all():
    if model.name == target_name:
        print(f"Found: {model.id}")
        break

Response Structure​

The paginated response contains:

response = client.use_cases.list(page_limit=10)

# Items for this page
response.items  # List[UseCaseResponse]

# Pagination metadata
response.pagination.has_more      # bool
response.pagination.next_cursor   # str | None
response.pagination.total_count   # int | None (if available)

Next Steps​