:hedgehog: PostHog is the leading platform for building self-driving products. Our developer tools – AI observability, analytics, session replay, flags, experiments, error tracking, logs, and more – capture all the context agents need to diagnose problems, uncover opportunities, and ship fixes. Steer it all from Slack, web, desktop, or the MCP.
Python ab-testingai-analyticsanalyticscdpdata-warehouseexperimentsfeature-flagsjavascriptproduct-analyticspythonreactsession-replaysurveystypescriptweb-analytics
"""Delivery orchestration: a scope's delivery summary, its pull request timelines, and an author's
comparison with their team.
The summary and the timelines take one scope (see ``delivery_scope``) and compare it with the
repository. The author page is the first caller; a team page reads the same summary with a GitHub
team, and a pull request page reads one timeline and the comparison.
"""
from products.engineering_analytics.backend.facade.contracts import (
DeliveryComparison,
DeliveryScopeKind,
DeliverySummary,
PullRequestTimelines,
)
from products.engineering_analytics.backend.logic._shared import _DEFAULT_WINDOW, _parse_window
from products.engineering_analytics.backend.logic.delivery_scope import DeliveryScope, SummaryScope
from products.engineering_analytics.backend.logic.queries._curated import CuratedGitHubSource
from products.engineering_analytics.backend.logic.queries.delivery_comparison import query_delivery_comparison
from products.engineering_analytics.backend.logic.queries.delivery_summary import query_delivery_summary
from products.engineering_analytics.backend.logic.queries.pull_request_timelines import query_pull_request_timelines
def build_delivery_summary(
*,
curated: CuratedGitHubSource,
scope: SummaryScope,
date_from: str | None = None,
date_to: str | None = None,
) -> DeliverySummary:
parsed_from, parsed_to = _parse_window(curated.team, date_from, date_to, default=_DEFAULT_WINDOW)
return query_delivery_summary(curated=curated, scope=scope, date_from=parsed_from, date_to=parsed_to)
def build_pull_request_timelines(
*,
curated: CuratedGitHubSource,
scope: DeliveryScope,
date_from: str | None = None,
date_to: str | None = None,
) -> PullRequestTimelines:
# A single pull request is shown whatever its age, so neither end of the window applies to it.
single_pr = scope.kind == DeliveryScopeKind.PULL_REQUEST
window_from = None if single_pr else date_from
window_to = None if single_pr else date_to
parsed_from, parsed_to = _parse_window(curated.team, window_from, window_to, default=_DEFAULT_WINDOW)
return query_pull_request_timelines(curated=curated, scope=scope, date_from=parsed_from, date_to=parsed_to)
def build_delivery_comparison(
*,
curated: CuratedGitHubSource,
author: str,
focus_pr: int | None = None,
date_from: str | None = None,
date_to: str | None = None,
) -> DeliveryComparison:
parsed_from, parsed_to = _parse_window(curated.team, date_from, date_to, default=_DEFAULT_WINDOW)
return query_delivery_comparison(
curated=curated, author=author, focus_pr=focus_pr, date_from=parsed_from, date_to=parsed_to
)