feat: add multilingual support so Naomi can use Python too

This commit is contained in:
2026-01-23 15:32:02 -08:00
parent 38e7f15d93
commit c0ad74367a
52 changed files with 1305 additions and 46 deletions
+182
View File
@@ -0,0 +1,182 @@
import json
import re
from collections import defaultdict
DAYS = ['Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday', 'Sunday']
UTC_BLOCKS = {
'mornings': (6, 12), # 06:00 - 12:00 UTC
'afternoons': (12, 18), # 12:00 - 18:00 UTC
'evenings': (18, 24), # 18:00 - 00:00 UTC
'nights': (0, 6) # 00:00 - 06:00 UTC
}
def parse_utc_offset(timezone_str: str) -> float:
"""Extract UTC offset from timezone string like 'America/New_York (UTC-5)'"""
match = re.search(r'UTC([+-]?\d+(?::\d+)?)', timezone_str)
if match:
offset_str = match.group(1)
if ':' in offset_str:
parts = offset_str.split(':')
hours = int(parts[0])
minutes = int(parts[1]) if len(parts) > 1 else 0
if hours < 0:
return hours - minutes / 60
return hours + minutes / 60
return float(offset_str)
return 0
def parse_time_slots(time_str: str) -> list[tuple[int, int]]:
"""Parse time slots like '17:00-18:00' or '07:00-08:00; 19:00-20:00'"""
slots = []
if not time_str or time_str.lower() in ['n/a', 'na', '']:
return slots
parts = time_str.split(';')
for part in parts:
part = part.strip()
match = re.search(r'(\d{1,2}):(\d{2})\s*-\s*(\d{1,2}):(\d{2})', part)
if match:
start_hour = int(match.group(1))
end_hour = int(match.group(3))
slots.append((start_hour, end_hour))
return slots
def local_hour_to_utc(local_hour: int, utc_offset: float) -> int:
"""Convert local hour to UTC hour"""
utc_hour = local_hour - utc_offset
return int(utc_hour) % 24
def get_utc_blocks_for_hour(utc_hour: int) -> list[str]:
"""Determine which UTC block(s) an hour falls into"""
blocks = []
for block_name, (start, end) in UTC_BLOCKS.items():
if block_name == 'nights':
if utc_hour >= 0 and utc_hour < 6:
blocks.append(block_name)
elif block_name == 'evenings':
if utc_hour >= 18 and utc_hour < 24:
blocks.append(block_name)
else:
if utc_hour >= start and utc_hour < end:
blocks.append(block_name)
return blocks
def analyze_applicant_availability(timezone_str: str, day_slots: dict) -> dict:
"""Analyze availability for one applicant"""
utc_offset = parse_utc_offset(timezone_str)
block_counts = defaultdict(int)
all_utc_hours = set()
for day in DAYS:
slots = day_slots.get(day, [])
for start_hour, end_hour in slots:
for hour in range(start_hour, end_hour):
utc_hour = local_hour_to_utc(hour, utc_offset)
all_utc_hours.add(utc_hour)
blocks = get_utc_blocks_for_hour(utc_hour)
for block in blocks:
block_counts[block] += 1
available_blocks = []
for block in ['mornings', 'afternoons', 'evenings', 'nights']:
if block_counts[block] >= 3:
available_blocks.append(block)
return {
'utc_offset': utc_offset,
'timezone': timezone_str,
'available_blocks': available_blocks,
'block_counts': dict(block_counts),
'total_unique_utc_hours': len(all_utc_hours)
}
def parse_table_md() -> list[dict]:
"""Parse table.md and extract availability data"""
with open('table.md', 'r') as f:
content = f.read()
lines = content.strip().split('\n')
header_idx = None
for i, line in enumerate(lines):
if line.startswith('| Discord ID'):
header_idx = i
break
if header_idx is None:
raise ValueError("Could not find table header")
headers = [h.strip() for h in lines[header_idx].split('|')[1:-1]]
applicants = []
for line in lines[header_idx + 2:]:
if not line.startswith('|'):
continue
cells = [c.strip() for c in line.split('|')[1:-1]]
if len(cells) < len(headers):
continue
row = dict(zip(headers, cells))
applicants.append(row)
return applicants
def main():
with open('discord_verification.json', 'r') as f:
verification = json.load(f)
verified_ids = set(v[0] for v in verification['verified'])
print(f"Verified applicants: {len(verified_ids)}")
applicants = parse_table_md()
print(f"Total applicants in table: {len(applicants)}")
availability_results = []
for applicant in applicants:
discord_id = applicant.get('Discord ID', '')
if discord_id not in verified_ids:
continue
timezone = applicant.get('Timezone', '')
day_slots = {}
for day in DAYS:
time_str = applicant.get(day, '')
day_slots[day] = parse_time_slots(time_str)
analysis = analyze_applicant_availability(timezone, day_slots)
availability_results.append({
'discord_id': discord_id,
'timezone': timezone,
'utc_offset': analysis['utc_offset'],
'available_blocks': analysis['available_blocks'],
'block_counts': analysis['block_counts'],
'total_unique_utc_hours': analysis['total_unique_utc_hours']
})
with open('availability_analysis.json', 'w') as f:
json.dump(availability_results, f, indent=2)
block_distribution = defaultdict(int)
for result in availability_results:
for block in result['available_blocks']:
block_distribution[block] += 1
print(f"\n=== AVAILABILITY ANALYSIS COMPLETE ===")
print(f"Analyzed: {len(availability_results)} applicants")
print(f"\nBlock Distribution (applicants available in each block):")
for block in ['mornings', 'afternoons', 'evenings', 'nights']:
print(f" {block.capitalize()}: {block_distribution[block]}")
no_blocks = sum(1 for r in availability_results if not r['available_blocks'])
print(f"\nApplicants with no clear block availability: {no_blocks}")
print(f"\nResults saved to availability_analysis.json")
if __name__ == "__main__":
main()
@@ -0,0 +1,238 @@
import json
import re
import time
import urllib.request
import urllib.error
from typing import Optional
# GitHub API (no auth needed for public repos, but rate limited)
GITHUB_API = "https://api.github.com"
def extract_github_info(url: str) -> tuple[Optional[str], Optional[str]]:
"""Extract owner and repo from GitHub URL."""
# Handle various GitHub URL formats
patterns = [
r'github\.com/([^/]+)/([^/\s?#]+)', # github.com/owner/repo
r'github\.com/([^/\s?#]+)/?$', # github.com/owner (profile)
]
for pattern in patterns:
match = re.search(pattern, url)
if match:
groups = match.groups()
if len(groups) == 2:
return groups[0], groups[1].rstrip('.git')
elif len(groups) == 1:
return groups[0], None
return None, None
def fetch_github_user(username: str) -> Optional[dict]:
"""Fetch GitHub user profile."""
url = f"{GITHUB_API}/users/{username}"
req = urllib.request.Request(url)
req.add_header("Accept", "application/vnd.github.v3+json")
req.add_header("User-Agent", "Cohort-Evaluator")
try:
response = urllib.request.urlopen(req, timeout=10)
return json.loads(response.read().decode())
except Exception as e:
return None
def fetch_github_repos(username: str) -> list:
"""Fetch user's public repos."""
url = f"{GITHUB_API}/users/{username}/repos?per_page=100&sort=updated"
req = urllib.request.Request(url)
req.add_header("Accept", "application/vnd.github.v3+json")
req.add_header("User-Agent", "Cohort-Evaluator")
try:
response = urllib.request.urlopen(req, timeout=10)
return json.loads(response.read().decode())
except Exception as e:
return []
def fetch_repo_languages(owner: str, repo: str) -> dict:
"""Fetch languages used in a repo."""
url = f"{GITHUB_API}/repos/{owner}/{repo}/languages"
req = urllib.request.Request(url)
req.add_header("Accept", "application/vnd.github.v3+json")
req.add_header("User-Agent", "Cohort-Evaluator")
try:
response = urllib.request.urlopen(req, timeout=10)
return json.loads(response.read().decode())
except Exception as e:
return {}
def analyze_proficiency_text(text: str) -> tuple[str, list[str]]:
"""Analyze self-described proficiency text."""
text_lower = text.lower()
# Extract languages/technologies mentioned
tech_patterns = [
r'\b(python|java|javascript|typescript|c\+\+|c#|ruby|go|rust|swift|kotlin|php|perl|scala|r)\b',
r'\b(react|angular|vue|node|express|django|flask|spring|rails|laravel)\b',
r'\b(html|css|sass|scss|tailwind|bootstrap)\b',
r'\b(sql|mysql|postgresql|mongodb|redis|firebase)\b',
r'\b(docker|kubernetes|aws|azure|gcp|git)\b',
r'\b(machine learning|ml|ai|data science|tensorflow|pytorch)\b',
]
technologies = set()
for pattern in tech_patterns:
matches = re.findall(pattern, text_lower)
technologies.update(matches)
# Determine level from keywords
beginner_keywords = ['beginner', 'learning', 'new to', 'just started', 'basic', 'novice', 'early']
intermediate_keywords = ['intermediate', 'comfortable', 'familiar', 'some experience', 'worked with']
advanced_keywords = ['advanced', 'expert', 'senior', 'professional', 'years of experience', 'proficient', 'strong']
level = 'intermediate' # default
if any(kw in text_lower for kw in advanced_keywords):
level = 'advanced'
elif any(kw in text_lower for kw in beginner_keywords):
level = 'beginner'
elif any(kw in text_lower for kw in intermediate_keywords):
level = 'intermediate'
return level, list(technologies)
def evaluate_applicant(applicant: dict, index: int, total: int) -> dict:
"""Evaluate a single applicant's technical proficiency."""
discord_id = applicant['discord_id']
project_url = applicant['project_url']
proficiency_self = applicant['proficiency_self']
project_reason = applicant['project_reason']
print(f"[{index+1}/{total}] Evaluating {discord_id}...")
result = {
'discord_id': discord_id,
'github_username': None,
'github_repos_count': 0,
'github_followers': 0,
'languages_from_github': [],
'languages_from_text': [],
'self_described_level': None,
'final_proficiency': 'intermediate', # default
'tech_stack': [],
'notes': []
}
# Analyze self-description
text_level, text_techs = analyze_proficiency_text(proficiency_self + " " + project_reason)
result['self_described_level'] = text_level
result['languages_from_text'] = text_techs
# Fetch GitHub data if URL provided
if project_url and 'github.com' in project_url:
owner, repo = extract_github_info(project_url)
if owner:
result['github_username'] = owner
# Fetch user profile
user_data = fetch_github_user(owner)
if user_data:
result['github_repos_count'] = user_data.get('public_repos', 0)
result['github_followers'] = user_data.get('followers', 0)
# Fetch repos to get languages
repos = fetch_github_repos(owner)
all_languages = set()
for r in repos[:10]: # Check top 10 repos
if r.get('language'):
all_languages.add(r['language'].lower())
result['languages_from_github'] = list(all_languages)
# If specific repo provided, get its languages
if repo:
repo_langs = fetch_repo_languages(owner, repo)
for lang in repo_langs.keys():
all_languages.add(lang.lower())
result['languages_from_github'] = list(all_languages)
time.sleep(0.5) # Rate limiting
# Combine tech stack
all_tech = set(result['languages_from_github']) | set(result['languages_from_text'])
result['tech_stack'] = sorted(list(all_tech))
# Determine final proficiency
# Factors: self-description, GitHub activity, tech diversity
github_score = 0
if result['github_repos_count'] >= 20:
github_score += 2
elif result['github_repos_count'] >= 10:
github_score += 1
if result['github_followers'] >= 50:
github_score += 2
elif result['github_followers'] >= 10:
github_score += 1
tech_count = len(result['tech_stack'])
if tech_count >= 6:
github_score += 2
elif tech_count >= 3:
github_score += 1
# Map self-described level to score
level_scores = {'beginner': 0, 'intermediate': 2, 'advanced': 4}
self_score = level_scores.get(text_level, 2)
# Combined score
total_score = github_score + self_score
if total_score >= 7:
result['final_proficiency'] = 'advanced'
elif total_score >= 3:
result['final_proficiency'] = 'intermediate'
else:
result['final_proficiency'] = 'beginner'
# Add notes
if not project_url or 'github.com' not in project_url:
result['notes'].append('No GitHub URL provided')
if result['github_repos_count'] == 0 and result['github_username']:
result['notes'].append('GitHub profile has no public repos')
return result
def main():
# Load applicants
with open('applicants_to_evaluate.json', 'r') as f:
applicants = json.load(f)
print(f"Evaluating {len(applicants)} applicants...\n")
evaluations = []
for i, applicant in enumerate(applicants):
result = evaluate_applicant(applicant, i, len(applicants))
evaluations.append(result)
# Progress update every 10
if (i + 1) % 10 == 0:
print(f" Progress: {i+1}/{len(applicants)} complete")
# Save results
with open('proficiency_evaluations.json', 'w') as f:
json.dump(evaluations, f, indent=2)
# Summary
beginner = sum(1 for e in evaluations if e['final_proficiency'] == 'beginner')
intermediate = sum(1 for e in evaluations if e['final_proficiency'] == 'intermediate')
advanced = sum(1 for e in evaluations if e['final_proficiency'] == 'advanced')
print(f"\n=== EVALUATION COMPLETE ===")
print(f"Beginner: {beginner}")
print(f"Intermediate: {intermediate}")
print(f"Advanced: {advanced}")
print(f"Total: {len(evaluations)}")
print(f"\nResults saved to proficiency_evaluations.json")
if __name__ == "__main__":
main()