NDOJ/judge/views/stats.py

201 lines
6.6 KiB
Python
Raw Normal View History

2020-01-21 06:35:58 +00:00
from itertools import chain, repeat
from operator import itemgetter
2022-09-22 22:33:14 +00:00
import json
2022-09-23 00:23:41 +00:00
import pandas as pd
2020-01-21 06:35:58 +00:00
from django.conf import settings
from django.db.models import Case, Count, FloatField, IntegerField, Value, When
from django.db.models.expressions import CombinedExpression
from django.http import JsonResponse
from django.shortcuts import render
from django.utils.translation import gettext as _
2022-09-22 22:33:14 +00:00
from django.http import Http404
from django.views.generic import TemplateView
from django.utils.safestring import mark_safe
2022-09-23 00:23:41 +00:00
from django.db import connection
2020-01-21 06:35:58 +00:00
from judge.models import Language, Submission
2022-05-14 17:57:27 +00:00
from judge.utils.stats import (
chart_colors,
get_bar_chart,
get_pie_chart,
highlight_colors,
)
2020-01-21 06:35:58 +00:00
2022-09-22 22:33:14 +00:00
class StatViewBase(TemplateView):
def get(self, request, *args, **kwargs):
if not request.user.is_superuser:
raise Http404
return super().get(request, *args, **kwargs)
2020-01-21 06:35:58 +00:00
2022-09-22 22:33:14 +00:00
class StatLanguage(StatViewBase):
template_name = "stats/language.html"
ac_count = Count(
Case(When(submission__result="AC", then=Value(1)), output_field=IntegerField())
2022-05-14 17:57:27 +00:00
)
2020-01-21 06:35:58 +00:00
2022-09-22 22:33:14 +00:00
def repeat_chain(iterable):
return chain.from_iterable(repeat(iterable))
2020-01-21 06:35:58 +00:00
2022-09-22 22:33:14 +00:00
def language_data(
self, language_count=Language.objects.annotate(count=Count("submission"))
):
languages = (
language_count.filter(count__gt=0)
.values("key", "name", "count")
2022-05-14 17:57:27 +00:00
.order_by("-count")
)
2022-09-22 22:33:14 +00:00
num_languages = min(len(languages), settings.DMOJ_STATS_LANGUAGE_THRESHOLD)
other_count = sum(map(itemgetter("count"), languages[num_languages:]))
return {
2022-09-23 00:23:41 +00:00
"labels": list(map(itemgetter("name"), languages[:num_languages]))
+ ["Other"],
"datasets": [
{
"backgroundColor": chart_colors[:num_languages] + ["#FDB45C"],
"highlightBackgroundColor": highlight_colors[:num_languages]
+ ["#FFC870"],
"data": list(map(itemgetter("count"), languages[:num_languages]))
+ [other_count],
},
],
}
2022-09-22 22:33:14 +00:00
def ac_language_data(self):
return self.language_data(Language.objects.annotate(count=self.ac_count))
def status_data(self, statuses=None):
if not statuses:
statuses = (
Submission.objects.values("result")
.annotate(count=Count("result"))
.values("result", "count")
.order_by("-count")
)
data = []
for status in statuses:
res = status["result"]
if not res:
continue
count = status["count"]
data.append((str(Submission.USER_DISPLAY_CODES[res]), count))
return get_pie_chart(data)
def ac_rate(self):
rate = CombinedExpression(
2022-09-23 00:23:41 +00:00
self.ac_count / Count("submission"),
"*",
Value(100.0),
output_field=FloatField(),
2022-09-22 22:33:14 +00:00
)
data = (
Language.objects.annotate(total=Count("submission"), ac_rate=rate)
.filter(total__gt=0)
.order_by("total")
.values_list("name", "ac_rate")
)
return get_bar_chart(list(data))
def get_context_data(self, **kwargs):
context = super().get_context_data(**kwargs)
context["title"] = _("Language statistics")
context["tab"] = "language"
context["data_all"] = mark_safe(json.dumps(self.language_data()))
context["lang_ac"] = mark_safe(json.dumps(self.ac_language_data()))
context["status_counts"] = mark_safe(json.dumps(self.status_data()))
context["ac_rate"] = mark_safe(json.dumps(self.ac_rate()))
2022-09-23 00:23:41 +00:00
return context
def group_data_by_period(dates, data, period="1W"):
df = pd.DataFrame()
df["dates"] = pd.to_datetime(dates)
df["data"] = data
df = df.groupby([pd.Grouper(key="dates", freq=period)])["data"].mean().reset_index()
return list(df["dates"]), list(df["data"])
class StatSite(StatViewBase):
template_name = "stats/site.html"
def create_dataset(self, query, label):
labels = []
data = []
with connection.cursor() as cursor:
cursor.execute(query)
for row in cursor.fetchall():
data.append(row[0])
labels.append(row[1])
labels, data = group_data_by_period(labels, data, self.period)
res = {
"labels": labels,
"datasets": [
{
"label": label,
"data": data,
"borderColor": "rgb(75, 192, 192)",
"pointRadius": 1,
"borderWidth": 2,
}
],
}
return mark_safe(json.dumps(res, default=str))
def get_submissions(self):
query = """
SELECT COUNT(*), CAST(date AS Date) as day from judge_submission GROUP BY day;
"""
return self.create_dataset(query, _("Submissions"))
def get_comments(self):
query = """
SELECT COUNT(*), CAST(time AS Date) as day from judge_comment GROUP BY day;
"""
return self.create_dataset(query, _("Comments"))
def get_new_users(self):
query = """
SELECT COUNT(*), CAST(date_joined AS Date) as day from auth_user GROUP BY day;
"""
return self.create_dataset(query, _("New users"))
def get_chat_messages(self):
query = """
SELECT COUNT(*), CAST(time AS Date) as day from chat_box_message GROUP BY day;
"""
return self.create_dataset(query, _("Chat messages"))
def get_contests(self):
query = """
SELECT COUNT(*), CAST(start_time AS Date) as day from judge_contest GROUP BY day;
"""
return self.create_dataset(query, _("Contests"))
def get_groups(self):
query = """
SELECT COUNT(*), CAST(creation_date AS Date) as day from judge_organization GROUP BY day;
"""
return self.create_dataset(query, _("Groups"))
def get(self, request, *args, **kwargs):
self.period = request.GET.get("period", "1W")
return super().get(request, *args, **kwargs)
def get_context_data(self, **kwargs):
context = super().get_context_data(**kwargs)
context["tab"] = "site"
context["title"] = _("Site statistics")
context["submissions"] = self.get_submissions()
context["comments"] = self.get_comments()
context["new_users"] = self.get_new_users()
context["chat_messages"] = self.get_chat_messages()
context["contests"] = self.get_contests()
context["groups"] = self.get_groups()
return context