---
id: dbs3-client
version: "4.0.19"
license: Apache License, Version 2.0
license_treatment: permissive
maintenance: dormant
---

# dbs3-client

License: permissive · Maintenance: dormant · Downloads: 485.7K/mo

## What it is and what it does

DBS3-Client is a Python wrapper around the Data Bookkeeping Service 3 REST API used by the CMS experiment at CERN. It allows Python scripts to query and manage event dataset metadata—including processing history, associated files, runs, and data tiers—without directly calling HTTP endpoints. The library abstracts the REST layer and JSON serialization, providing a Pythonic interface to a CMS-internal data catalog.

The package depends on dbs3-pycurl for HTTP communication and requires libcurl to be installed on the system. Setup involves installing system libraries, configuring X.509 certificates for authentication, and exporting environment variables. It is designed for CMS physicists and data managers working with Monte Carlo simulations and recorded collision data.

Use it for:

- Query CMS dataset metadata (tiers, files, runs) from Python scripts in analysis workflows.
- Automate dataset discovery and validation during data processing pipelines.
- Retrieve processing history and file locations for a given dataset in batch jobs.
- Validate dataset availability and metadata consistency in Monte Carlo production.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Python client for querying and managing datasets in the Data Bookkeeping Service 3 (DBS3), a CMS experiment event data catalog that tracks datasets, processing history, files, and runs via REST API.

Yes, if you are a CMS physicist or data manager needing programmatic access to DBS3 from Python. No, if you are outside the CMS collaboration or need a general-purpose REST client—the package is CMS-specific, dormant (last release 697 days ago), and requires non-trivial system setup (libcurl, X.509 certs). The low install friction and zero known vulnerabilities are offset by operational complexity and maintenance uncertainty.

## Install

pip install dbs3-client
uv add dbs3-client
poetry add dbs3-client

## Installing dbs3-client

Before you install: Low install friction with a single runtime dependency (dbs3-pycurl). However, the package is dormant—last release was 697 days ago—and requires libcurl system libraries and X.509 certificate setup, which adds operational complexity despite straightforward pip installation.

License in practice: Licensed under Apache License, Version 2.0 (permissive), allowing commercial and private use with minimal restrictions; suitable for most deployment contexts.

Quickstart:

pip install dbs3-client

from dbs3_client import DbsApi

url = "https://cmsweb-testbed.cern.ch/dbs/prod/global/DBSReader/"
dbs3api = DbsApi(url=url)
print(dbs3api.listDataTiers(data_tier_name='AOD'))

Requires libcurl system library (install via yum or apt-get), X.509 certificate and key files for authentication, and environment variables DBS3_CLIENT_ROOT, X509_USER_CERT, X509_USER_KEY to be set.

Verify before relying:

- Whether the package works with Python versions beyond 3.8.2 despite documentation recommending 3.8.2 specifically.
- Current compatibility with modern pycurl versions or if pycurl==7.43.0.6 remains a hard requirement.
- Whether the package is actively maintained or if dormancy signals end-of-life for this CMS-internal tool.
- Exact API surface and available methods beyond listDataTiers() shown in documentation.

## Package facts

- License: Apache License, Version 2.0 (permissive)
- Python support: unspecified
- Install friction: low
- Maintenance: dormant
- Downloads: 485.7K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

CMS dataset catalog client, DBS3 Python API, event data bookkeeping service, CMS data management, dataset metadata query, REST API data catalog, cms-specific, data-catalog

[View on SkillFed](https://skillfed.io/packages/dbs3-client) · [View on PyPI](https://pypi.org/project/dbs3-client/)