info-needed-comments.py

import subprocess
import json
from collections import Counter
from nltk.corpus import stopwords
from nltk import pos_tag
from spellchecker import SpellChecker

# List of triagers whose comments will be included in the analysis
TRIAGERS = ["karrtikr","karthiknadig","paulacamargo25","eleanorjboyd", "brettcannon", "ericsnowcurrently", "DonJayamanne"]

# Get the list of issue numbers
result = subprocess.run(["gh", "issue", "list", "--state=closed", "--label=info-needed", "-R", "microsoft/vscode-python", "--json", "number"], capture_output=True)
if result.returncode != 0:
    print(result.stderr.decode())
    exit(1)

issues = json.loads(result.stdout.decode())

# Loop through the issue numbers and retrieve the second last comment for each issue
all_comments = []
for issue in issues:
    result = subprocess.run(["gh", "issue", "view", str(issue["number"]), "--comments", "-R", "microsoft/vscode-python", "--json", "comments"], capture_output=True)
    if result.returncode != 0:
        print(result.stderr.decode())
        exit(1)

    data = json.loads(result.stdout.decode())
    if "comments" not in data:
        print(f"Error: comments not found in JSON output for issue {issue['number']}")
        continue

    triager_comments = []
    for comment in data["comments"]:
        if comment["author"]["login"] in TRIAGERS:
            triager_comments.append(comment)

    if len(triager_comments) >= 2:
        comment = triager_comments[-2]
        print(comment["body"])
        print('--------------------')
        all_comments.append(comment["body"])

# Ignore stop words, nouns, and pronouns, and count the frequency of the remaining words in the second last comment across all issues
stop_words = set(stopwords.words('english'))
spell = SpellChecker(language='en', distance=1)
all_words = []
for comment in all_comments:
    words = comment.split()
    if len(words) >= 2:
        # filter out stop words, non-words, pronouns, and nouns
        cleaned_words = [word for word, pos in pos_tag(words) if word not in stop_words and word.isalpha() and not spell.unknown([word]) and pos != 'NN' and pos != 'NNS' and pos != 'NNP' and pos != 'NNPS' and pos != 'PRP' and pos != 'PRP$']
        all_words.extend(cleaned_words)

word_counts = Counter(all_words)

# Print the top 10 most frequent words
print("Top 10 most frequent words in the second last comment across all issues (excluding stop words, nouns, and pronouns):")
for word, count in word_counts.most_common(10):
    print(f"{word}: {count} occurrences")