thunderbird-metrics-and-reports/sumo-tb-android-create-monthly-csv at main · thunderbird/thunderbird-metrics-and-reports · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
#!/usr/bin/env python3
import sys
import os
import pandas as pd
from datetime import datetime, timezone
import logging

if len(sys.argv) == 3:
    year = int(sys.argv[1])
    month = int(sys.argv[2])
else:
    now = datetime.now(timezone.utc)
    year = now.year
    month = now.month

# Setup logging
logging.basicConfig(level=logging.WARNING, format='%(levelname)s: %(message)s')
logging.warning(f"Year: {year} Month: {month}")


questions_file = f'CONCATENATED_FILES/ANDROID/{year}-{month:02d}-sumo-android-questions.csv'
answers_file = f'CONCATENATED_FILES/ANDROID/{year}-{month:02d}-sumo-android-answers.csv'
trusted_file = 'CONCATENATED_FILES/ANDROID/thunderbird-android-trusted-contributors.csv'
output_file = f'REPORTS/ANDROID/{year}-{month:02d}-sumo-android-report.csv'

if not os.path.exists(questions_file):
    print(f"Questions file {questions_file} does not exist")
    sys.exit(1)

if not os.path.exists(answers_file):
    print(f"Answers file {answers_file} does not exist")
    sys.exit(1)

if not os.path.exists(trusted_file):
    print(f"Trusted contributors file {trusted_file} does not exist")
    sys.exit(1)

questions_df = pd.read_csv(questions_file)
answers_df = pd.read_csv(answers_file)
trusted_df = pd.read_csv(trusted_file)

trusted_creators = set(trusted_df['creator'])

# Parse created times to UTC
answers_df['created_utc'] = pd.to_datetime(answers_df['created'], utc=True)

num_questions = len(questions_df)
num_solved = questions_df['is_solved'].sum()
solved_percentage = round(num_solved / num_questions * 100) if num_questions > 0 else 0

num_ignored = (questions_df['num_answers'] == 0).sum()
ignored_percentage = round(num_ignored / num_questions * 100) if num_questions > 0 else 0

# Algorithm for synthetic_solved_by_random_contributors and synthetic_solved_by_trusted_contributors
# Step 1-2: Get unsolved questions (is_solved == False) with answers > 0
pd.set_option('display.max_columns', None)
unsolved_with_answers = questions_df[(questions_df['is_solved'] == False) & (questions_df['num_answers'] > 0)].copy()
logging.warning(f"unsolved_with_answers: {unsolved_with_answers['id'].tolist()}")
# Step 3: Ignore questions where the last answer is by the creator of the question
# Get last answer for each question
last_answers_df = answers_df.loc[answers_df.groupby('question_id')['created_utc'].idxmax()][['question_id', 'creator']].rename(columns={'creator': 'last_answer_creator'})
logging.warning(f"last_answers_df: {last_answers_df.head()}")
# Merge unsolved questions with their last answer
unsolved_with_last_answer = unsolved_with_answers.merge(last_answers_df, left_on='id', right_on='question_id', how='left')
logging.warning(f"unsolved_with_last_answer: {unsolved_with_last_answer.head()}")
# Filter out questions where last answer is by the creator of the question
filtered_questions = unsolved_with_last_answer[unsolved_with_last_answer['creator'] != unsolved_with_last_answer['last_answer_creator']].copy()
logging.warning(f"filtered_questions: {filtered_questions.head()}")
# Step 4-5: Count synthetic solved by random and trusted contributors
synthetic_solved_by_random = (filtered_questions['last_answer_creator'].notna() & ~filtered_questions['last_answer_creator'].isin(trusted_creators)).sum()
synthetic_solved_by_trusted = (filtered_questions['last_answer_creator'].notna() & filtered_questions['last_answer_creator'].isin(trusted_creators)).sum()

# Percentages: use total number of questions as denominator
synthetic_random_percentage = round(synthetic_solved_by_random / num_questions * 100) if num_questions > 0 else 0
synthetic_trusted_percentage = round(synthetic_solved_by_trusted / num_questions * 100) if num_questions > 0 else 0

# Step 5: Log question IDs
# Log questions with is_solved = True
is_solved_true_ids = questions_df[questions_df['is_solved'] == True]['id'].tolist()
if is_solved_true_ids:
    logging.warning(f"Questions with is_solved=True (ignored): {is_solved_true_ids}")

# Log questions with num_answers = 0
num_answers_zero_ids = questions_df[questions_df['num_answers'] == 0]['id'].tolist()
if num_answers_zero_ids:
    logging.warning(f"Questions with num_answers=0 (ignored): {num_answers_zero_ids}")

# Log questions where last answer is by the creator
last_answer_by_creator_ids = unsolved_with_last_answer[unsolved_with_last_answer['creator'] == unsolved_with_last_answer['last_answer_creator']]['id'].tolist()
if last_answer_by_creator_ids:
    logging.warning(f"Questions where last answer is by creator (ignored): {last_answer_by_creator_ids}")

# Log synthetic solved by random contributors
random_contrib_mask = ~filtered_questions['last_answer_creator'].isin(trusted_creators)
if not filtered_questions[random_contrib_mask].empty:
    random_contrib_ids = filtered_questions[random_contrib_mask]['id'].tolist()
    logging.warning(f"Synthetic solved by random contributors: {random_contrib_ids}")
else:
    random_contrib_ids = []

# Log synthetic solved by trusted contributors
trusted_contrib_mask = filtered_questions['last_answer_creator'].isin(trusted_creators)
if not filtered_questions[trusted_contrib_mask].empty:
    trusted_contrib_ids = filtered_questions[trusted_contrib_mask]['id'].tolist()
    logging.warning(f"Synthetic solved by trusted contributors: {trusted_contrib_ids}")
else:
    trusted_contrib_ids = []

# Calculate synthetic_solved_rate
synthetic_solved_rate = solved_percentage + synthetic_trusted_percentage

# Create semicolon-delimited strings for each question category
question_ids_creator_answered_last = ';'.join(map(str, last_answer_by_creator_ids))
question_ids_trusted_contributor_answered_last = ';'.join(map(str, trusted_contrib_ids))
question_ids_random_contributor_answered_last = ';'.join(map(str, random_contrib_ids))

# Create report
columns = [
    'Date',
    'num_questions',
    'num_solved',
    'solved-percentage',
    'num_ignored',
    'ignored-percentage',
    'synthetic_solved_by_random_contributors',
    'synthetic_solved_by_random_contributors-percentage',
    'synthetic_solved_by_trusted_contributors',
    'synthetic_solved_by_trusted_contributors-percentage',
    'synthetic_solved_rate',
    'question_ids_question_creator_answered_last',
    'question_ids_trusted_contributor_answered_last',
    'question_ids_random_contributor_answered_last'
]

data = [
    f'{year}-{month:02d}',
    num_questions,
    num_solved,
    solved_percentage,
    num_ignored,
    ignored_percentage,
    synthetic_solved_by_random,
    synthetic_random_percentage,
    synthetic_solved_by_trusted,
    synthetic_trusted_percentage,
    synthetic_solved_rate,
    question_ids_creator_answered_last,
    question_ids_trusted_contributor_answered_last,
    question_ids_random_contributor_answered_last
]

report_df = pd.DataFrame([data], columns=columns)
report_df.to_csv(output_file, index=False)

print(f"Created {output_file}")