ado-workflows

Azure DevOps workflow automation library — PR review analysis, comment lifecycle, repository discovery, and Python SDK wrappers.

Install

pip install ado-workflows

Quick Start

from ado_workflows import ConnectionFactory, AdoClient

# Authenticate (uses DefaultAzureCredential — Azure CLI, managed identity, etc.)
factory = ConnectionFactory()
connection = factory.get_connection("https://dev.azure.com/MyOrg")
client = AdoClient(connection)

PR Review Status

from ado_workflows import get_review_status

status = get_review_status(client, "MyProject", "MyRepo", pr_id=12345)
print(status.approval_status)    # ApprovalStatus.APPROVED / PENDING / REJECTED
print(status.total_reviewers)    # 3
print(status.required_approvals) # 2
for vs in status.vote_statuses:
    print(f"{vs.name}: {vs.vote_text} (stale: {vs.vote_invalidated})")

Pending Review Analysis

from ado_workflows import analyze_pending_reviews

result = analyze_pending_reviews(
    client, "MyProject", "MyRepo",
    max_days_old=14,
    creator_filter="alice",
)
for pr in result.pending_prs:  # sorted by days_open descending
    print(f"PR #{pr.pr_id} ({pr.days_open}d): {pr.title}")
    print(f"  Needs {pr.needs_approvals_count} more approvals")
    for r in pr.pending_reviewers:
        print(f"  - {r.display_name}")

Comment Operations

from ado_workflows import (
    analyze_pr_comments,
    post_comment,
    reply_to_comment,
    resolve_comments,
)

# Analyze threads
analysis = analyze_pr_comments(client, "MyProject", "MyRepo", pr_id=12345)
print(f"{analysis.summary.active_count} active, {analysis.summary.resolved_count} resolved")

# Post, reply, resolve
thread_id = post_comment(client, "MyProject", "MyRepo", pr_id=12345, content="LGTM")
reply_to_comment(client, "MyProject", "MyRepo", pr_id=12345, thread_id=thread_id, content="Thanks!")
result = resolve_comments(client, "MyProject", "MyRepo", pr_id=12345, thread_ids=[thread_id])
print(f"Resolved: {result.resolved}, Failed: {result.failed}")

Repository Discovery

from ado_workflows import discover_repositories, parse_ado_url

# Parse any Azure DevOps URL
org, project, repo, pr_id = parse_ado_url(
    "https://dev.azure.com/MyOrg/MyProject/_git/MyRepo/pullrequest/42"
)

# Discover git repos in a workspace
repos = discover_repositories("/path/to/workspace")
for r in repos:
    print(f"{r['name']} → {r['remote_url']}")

PR Context

from ado_workflows import establish_pr_context

# From a URL
ctx = establish_pr_context("https://dev.azure.com/MyOrg/MyProject/_git/MyRepo/pullrequest/42")

# From a PR ID (requires RepositoryContext to be set)
ctx = establish_pr_context(42, working_directory="/path/to/repo")

print(ctx.org_url)    # https://dev.azure.com/MyOrg
print(ctx.pr_id)      # 42

Architecture

Three-layer API:

Layer	Purpose	State	SDK dependency
1 — Primitives	URL parsing, git inspection, date parsing	None	None
2 — Context	Repository context caching	Thread-safe	None
3 — PR Operations	Review, comments, lifecycle, pending analysis	Composes L1+L2	azure-devops

See docs/ARCHITECTURE.md for full design rationale.

Error Handling

All errors are ActionableError instances (from actionable-errors) with structured context:

from actionable_errors import ActionableError

try:
    status = get_review_status(client, "MyProject", "MyRepo", pr_id=99999)
except ActionableError as e:
    print(e.error_type)   # ErrorType.CONNECTION
    print(e.service)      # "AzureDevOps"
    print(e.suggestion)   # "Verify the PR ID exists and you have read access..."

Development

uv sync --extra dev
task check          # lint + type-check + test

See CONTRIBUTING.md for full development setup and coding standards.

License

MIT