Add approval stats page

This commit is contained in:
rubenwardy 2024-05-17 18:52:55 +01:00
parent a920854796
commit 04b87a4e74
6 changed files with 442 additions and 1 deletions

@ -19,4 +19,4 @@ from flask import Blueprint
bp = Blueprint("admin", __name__)
from . import admin, audit, licenseseditor, tagseditor, versioneditor, warningseditor, languageseditor, email
from . import admin, audit, licenseseditor, tagseditor, versioneditor, warningseditor, languageseditor, email, approval_stats

@ -0,0 +1,77 @@
# ContentDB
# Copyright (C) 2024 rubenwardy
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with this program. If not, see <https://www.gnu.org/licenses/>.
import datetime
from flask import render_template, request, abort, redirect, url_for, jsonify
from . import bp
from app.logic.approval_stats import get_approval_statistics
from app.models import UserRank
from app.utils import rank_required
@bp.route("/admin/approval_stats/")
@rank_required(UserRank.APPROVER)
def approval_stats():
start = request.args.get("start")
end = request.args.get("end")
if start and end:
try:
start = datetime.datetime.fromisoformat(start)
end = datetime.datetime.fromisoformat(end)
except ValueError:
abort(400)
elif start:
return redirect(url_for("admin.approval_stats", start=start, end=datetime.datetime.utcnow().date().isoformat()))
elif end:
return redirect(url_for("admin.approval_stats", start="2020-07-01", end=end))
else:
end = datetime.datetime.utcnow()
start = end - datetime.timedelta(days=365)
stats = get_approval_statistics(start, end)
return render_template("admin/approval_stats.html", stats=stats, start=start, end=end)
@bp.route("/admin/approval_stats.json")
@rank_required(UserRank.APPROVER)
def approval_stats_json():
start = request.args.get("start")
end = request.args.get("end")
if start and end:
try:
start = datetime.datetime.fromisoformat(start)
end = datetime.datetime.fromisoformat(end)
except ValueError:
abort(400)
else:
end = datetime.datetime.utcnow()
start = end - datetime.timedelta(days=365)
stats = get_approval_statistics(start, end)
for key, value in stats.packages_info.items():
stats.packages_info[key] = value.__dict__()
return jsonify({
"start": start.isoformat(),
"end": end.isoformat(),
"editor_approvals": stats.editor_approvals,
"packages_info": stats.packages_info,
"turnaround_time": {
"avg": stats.avg_turnaround_time,
"max": stats.max_turnaround_time,
},
})

148
app/logic/approval_stats.py Normal file

@ -0,0 +1,148 @@
# ContentDB
# Copyright (C) 2024 rubenwardy
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with this program. If not, see <https://www.gnu.org/licenses/>.
import datetime
from collections import namedtuple, defaultdict
from typing import Dict, Optional
from sqlalchemy import or_
from app.models import AuditLogEntry, db, PackageState
class PackageInfo:
state: Optional[PackageState]
first_submitted: Optional[datetime.datetime]
last_change: Optional[datetime.datetime]
approved_at: Optional[datetime.datetime]
wait_time: int
total_approval_time: int
is_in_range: bool
events: list[tuple[str, str, str]]
def __init__(self):
self.state = None
self.first_submitted = None
self.last_change = None
self.approved_at = None
self.wait_time = 0
self.total_approval_time = -1
self.is_in_range = False
self.events = []
def __lt__(self, other):
return self.wait_time < other.wait_time
def __dict__(self):
return {
"first_submitted": self.first_submitted.isoformat(),
"last_change": self.last_change.isoformat(),
"approved_at": self.approved_at.isoformat() if self.approved_at else None,
"wait_time": self.wait_time,
"total_approval_time": self.total_approval_time if self.total_approval_time >= 0 else None,
"events": [ { "date": x[0], "by": x[1], "title": x[2] } for x in self.events ],
}
def add_event(self, created_at: datetime.datetime, causer: str, title: str):
self.events.append((created_at.isoformat(), causer, title))
def get_state(title: str):
if title.startswith("Approved "):
return PackageState.APPROVED
assert title.startswith("Marked ")
for state in PackageState:
if state.value in title:
return state
if "Work in Progress" in title:
return PackageState.WIP
raise Exception(f"Unable to get state for title {title}")
Result = namedtuple("Result", "editor_approvals packages_info avg_turnaround_time max_turnaround_time")
def _get_approval_statistics(entries: list[AuditLogEntry], start_date: Optional[datetime.datetime] = None, end_date: Optional[datetime.datetime] = None) -> Result:
editor_approvals = defaultdict(int)
package_info: Dict[str, PackageInfo] = {}
ignored_packages = set()
turnaround_times: list[int] = []
for entry in entries:
package_id = str(entry.package.get_id())
if package_id in ignored_packages:
continue
info = package_info.get(package_id, PackageInfo())
package_info[package_id] = info
is_in_range = (((start_date is None or entry.created_at >= start_date) and
(end_date is None or entry.created_at <= end_date)))
info.is_in_range = info.is_in_range or is_in_range
new_state = get_state(entry.title)
if new_state == info.state:
continue
info.add_event(entry.created_at, entry.causer.username if entry.causer else None, new_state.value)
if info.state == PackageState.READY_FOR_REVIEW:
seconds = int((entry.created_at - info.last_change).total_seconds())
info.wait_time += seconds
if is_in_range:
turnaround_times.append(seconds)
if new_state == PackageState.APPROVED:
ignored_packages.add(package_id)
info.approved_at = entry.created_at
if is_in_range:
editor_approvals[entry.causer.username] += 1
if info.first_submitted is not None:
info.total_approval_time = int((entry.created_at - info.first_submitted).total_seconds())
elif new_state == PackageState.READY_FOR_REVIEW:
if info.first_submitted is None:
info.first_submitted = entry.created_at
info.state = new_state
info.last_change = entry.created_at
packages_info_2 = {}
package_count = 0
for package_id, info in package_info.items():
if info.first_submitted and info.is_in_range:
package_count += 1
packages_info_2[package_id] = info
if len(turnaround_times) > 0:
avg_turnaround_time = sum(turnaround_times) / len(turnaround_times)
max_turnaround_time = max(turnaround_times)
else:
avg_turnaround_time = 0
max_turnaround_time = 0
return Result(editor_approvals, packages_info_2, avg_turnaround_time, max_turnaround_time)
def get_approval_statistics(start_date: Optional[datetime.datetime] = None, end_date: Optional[datetime.datetime] = None) -> Result:
entries = AuditLogEntry.query.filter(AuditLogEntry.package).filter(or_(
AuditLogEntry.title.like("Approved %"),
AuditLogEntry.title.like("Marked %"))
).order_by(db.asc(AuditLogEntry.created_at)).all()
return _get_approval_statistics(entries, start_date, end_date)

@ -0,0 +1,116 @@
{% extends "base.html" %}
{% block title %}
Approval Stats
{% endblock %}
{% block content %}
<h1>{{ self.title() }}</h1>
<p class="float-end">
<a href="{{ url_for('admin.approval_stats', start='2020-07-01') }}">Since Aug 2020</a> |
<a href="{{ url_for('admin.approval_stats') }}">Last 365 days</a>
</p>
{% if start or end %}
<p>
From {{ start.date() }} to {{ end.date() }}.
</p>
{% endif %}
<div class="row mb-5">
<div class="col-md-4">
<div class="card h-100">
<div class="card-body align-items-center text-center">
<div class="mt-0 mb-3">
Total packages submitted
</div>
<div class="my-0 h4">
{{ stats.packages_info | length }}
</div>
</div>
</div>
</div>
<div class="col-md-4">
<div class="card h-100">
<div class="card-body align-items-center text-center">
<div class="mt-0 mb-3">
Average turnaround time
</div>
<div class="my-0 h4">
{{ (stats.avg_turnaround_time / (60*60*24)) | round(1) }} days
</div>
</div>
</div>
</div>
<div class="col-md-4">
<div class="card h-100">
<div class="card-body align-items-center text-center">
<div class="mt-0 mb-3">
Max turnaround time
</div>
<div class="my-0 h4">
{{ (stats.max_turnaround_time / (60*60*24)) | round(1) }} days
</div>
</div>
</div>
</div>
</div>
<p class="text-muted">
Turnaround: how long after clicking "Ready for Review" does the package receive a response. A package can be marked
as ready for review multiple times.
</p>
<h2>Editor Approvals</h2>
<table class="table">
<tr>
<th>Name</th>
<th>Count</th>
</tr>
{% for name, count in stats.editor_approvals.items() | sort(attribute=1, reverse=True) %}
<tr>
<td>{{ name }}</td>
<td>{{ count }}</td>
</tr>
{% endfor %}
</table>
<h2>Packages</h2>
<table class="table">
<tr>
<th>Name</th>
<th>First submitted</th>
<th>Approved at</th>
<th>Time waiting for review</th>
<th>Time to approve</th>
</tr>
{% for name, info in stats.packages_info.items() | sort(attribute="1", reverse=True) %}
{% set parts = name.split("/") %}
<tr>
<td><a href="{{ url_for('packages.audit', author=parts[0], name=parts[1]) }}">{{ name }}</a></td>
<td>{{ info.first_submitted.date() }}</td>
<td>{% if info.approved_at %}{{ info.approved_at.date() }}{% endif %}</td>
<td>{{ (info.wait_time / (60*60*24)) | round(1) }}</td>
<td>
{% set approval_time = info.total_approval_time %}
{% if approval_time >= 0 %}
{{ (approval_time / (60*60*24)) | round(1) }}
{% endif %}
</td>
</tr>
{% endfor %}
</table>
<h2>Export</h2>
<p>
<a href="{{ url_for('admin.approval_stats_json') }}">approval_stats.json</a>
</p>
{% endblock %}

@ -41,6 +41,10 @@
<i class="fas fa-search me-2"></i>
Zip grep
</a>
<a class="list-group-item list-group-item-action" href="{{ url_for('admin.approval_stats') }}">
<i class="fas fa-chart-line me-2"></i>
Approval stats
</a>
</div>
<h3>Types</h3>

@ -0,0 +1,96 @@
# ContentDB
# Copyright (C) 2024 rubenwardy
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with this program. If not, see <https://www.gnu.org/licenses/>.
import datetime
from app.logic.approval_stats import _get_approval_statistics
from app.models import AuditLogEntry, User
class MockPackage:
def __init__(self, id_: str):
self.id_ = id_
def get_id(self):
return self.id_
class MockEntry:
def __init__(self, date: str, package_id: str, username: str, title: str):
causer = User()
causer.username = username
self.causer = causer
self.created_at = datetime.datetime.strptime(date, "%Y-%m-%dT%H:%M:%S%z")
self.title = title
self.package = MockPackage(package_id)
# noinspection PyTypeChecker
def make_entry(date: str, package_id: str, username: str, title: str) -> AuditLogEntry:
return MockEntry(date, package_id, username, title)
def test_empty():
stats = _get_approval_statistics([])
assert not stats.editor_approvals
assert not stats.packages_info
assert stats.avg_turnaround_time == 0
def test_package_simple():
package_1 = "user1/one"
stats = _get_approval_statistics([
make_entry("2023-04-03T12:32:00Z", package_1, "user1", "Marked Title as Ready for Review"),
make_entry("2023-04-06T12:32:00Z", package_1, "reviewer", "Marked Title as Changes Needed"),
make_entry("2023-04-07T12:32:00Z", package_1, "user1", "Marked Title as Ready for Review"),
make_entry("2023-04-08T12:32:00Z", package_1, "reviewer", "Approved Title"),
])
assert stats.packages_info[package_1].total_approval_time == 5*60*60*24
assert stats.packages_info[package_1].wait_time == 4*60*60*24
def test_average_turnaround():
package_1 = "user1/one"
package_2 = "user2/two"
stats = _get_approval_statistics([
make_entry("2023-04-03T12:00:00Z", package_1, "user1", "Marked Title as Ready for Review"),
make_entry("2023-04-03T18:00:00Z", package_2, "user2", "Marked Title as Ready for Review"),
make_entry("2023-04-06T12:00:00Z", package_1, "reviewer", "Marked Title as Changes Needed"),
make_entry("2023-04-07T12:00:00Z", package_1, "user1", "Marked Title as Ready for Review"),
make_entry("2023-04-08T12:00:00Z", package_1, "reviewer", "Approved Title"),
make_entry("2023-04-10T18:00:00Z", package_2, "reviewer", "Approved Title"),
])
assert stats.avg_turnaround_time == 316800
def test_editor_counts():
package_1 = "user1/one"
package_2 = "user1/two"
package_3 = "user1/three"
stats = _get_approval_statistics([
make_entry("2023-04-03T12:00:00Z", package_1, "user1", "Marked Title as Ready for Review"),
make_entry("2023-04-03T12:00:00Z", package_2, "user1", "Marked Title as Ready for Review"),
make_entry("2023-04-03T12:00:00Z", package_3, "user1", "Marked Title as Ready for Review"),
make_entry("2023-04-08T12:00:00Z", package_1, "reviewer", "Approved Title"),
make_entry("2023-04-10T18:00:00Z", package_2, "reviewer", "Approved Title"),
make_entry("2023-04-11T18:00:00Z", package_3, "reviewer2", "Approved Title"),
])
assert len(stats.editor_approvals) == 2
assert stats.editor_approvals["reviewer"] == 2
assert stats.editor_approvals["reviewer2"] == 1