status: api is working but not chromadb

chroma/db.py

@@ -1,7 +1,6 @@
 import chromadb
 import random
-import os
-import sqlite3
+
 
 # Recommendations
 def recommend(user_details, repos):
@@ -28,7 +27,7 @@ def recommend(user_details, repos):
         new_doc = f"{user_proj['project_name']} : {user_proj['description']}"
         results = collection.query(
             query_texts = [new_doc],
-            n_results = 2,
+            n_results = 3,
         )
         try:
             # recommending the repos in random

src/api.py

@@ -13,23 +13,26 @@ class User(BaseModel):
 
 
 @app.post('/recommendations/')
-async def get_recommendations(request : User) -> dict:
-    username = request.username
+async def get_recommendations(user: User) -> dict:
+    username = user.username
 
     try:
+        print(f'Fetching recommendations for {username}')
         user_details, language_topics = await get_repos(username)
+        print(f'--------\n{user_details}')
+        print(f'--------\n{language_topics}')
         unique_repos = await main(language_topics)
-
+        print(f'--------\n{unique_repos}')
         urls = recommend(user_details,unique_repos)
         return {'recommendations': urls}
     except Exception as e:
         print(e)
         raise HTTPException(status_code=500, detail = 'Error generating recommendatoins')
 
 
-async def main():
+async def run_server():
     uvicorn.run(app, host='0.0.0.0', port=8000)
 
 
 if __name__ == '__main__':
-    asyncio.run(main())
+    asyncio.run(run_server())

src/search.py

@@ -64,11 +64,12 @@ async def main(language_topics):
         tasks = []
 
         for language in languages:
+            print(f"Searching for {language} repositories")
             base_params = {
                 'q': f'stars:>=2000 forks:>=500 language:{language} pushed:>=2024-01-01',
                 'sort': 'stars',
                 'order': 'desc',
-                'per_page': 100,
+                'per_page': 30,
                 'page': 1,
             }
 
@@ -81,11 +82,12 @@ async def main(language_topics):
             tasks.append(asyncio.create_task(search_repositories(octokit, good_first_issues_params)))
 
         for topic in topics:
+            print(f"Searching for {topic} repositories")
             base_params = {
                 'q': f'stars:>=2000 forks:>=500 topic:{topic} pushed:>=2024-01-01',
                 'sort': 'stars',
                 'order': 'desc',
-                'per_page': 100,
+                'per_page': 30,
                 'page': 1,
             }
 
@@ -101,19 +103,22 @@ async def main(language_topics):
         for result in results:
             unique_repos.update(result)
 
-
+    print(f"Found {len(unique_repos)} unique repositories")
     return unique_repos
 
 
 if __name__ == '__main__':
-    language_topics = {
-        'languages': ['python'],
-        'topics': ['ai', 'web-development']
-    }
+    language_topics = {'languages': ['Jupyter Notebook', 'Python', 'C++', 'go'], 
+                       'topics': ['openai', 'llm-agent', 'agentic-agi', 'agentic']}
+
+    import time
 
+    start = time.time()
     loop = asyncio.get_event_loop()
     result = loop.run_until_complete(main(language_topics))
     loop.close()
+    end = time.time()
+    print(f"Time taken: {end - start:.2f} seconds")
 
 
     for repo_id, repo_info in result.items():

src/user_data.py

@@ -10,16 +10,19 @@
 GPAT = os.getenv('GPAT')
 
 async def get_repos(username: str):
+    """
+    Fetches the repositories of the user
+    """
     user_details = []
     language_topics = {}
 
     try:
         async with ClientSession() as session:
             octokit = Octokit(GPAT, session)
 
-        # To store the unique data of the user 
-            languages_set = set()
-            topics_set = set()
+            # To store the unique data of the user 
+            languages_map = {} # store the freq of the languages
+            topics_map = {}
 
             url = f'/users/{username}/repos'
             repos_data = await octokit.request('GET', url)
@@ -36,10 +39,16 @@ async def get_repos(username: str):
                     }
                     user_details.append(user_repo)
 
-                    languages_set.update(languages_data.keys())
-                    topics_set.update(repo['topics'])
+                    for lang in languages_data.keys():
+                        languages_map[lang] = languages_map.get(lang, 0) + 1
 
-            language_topics = {"languages" : list(languages_set), "topics" : list(topics_set)}
+                    for topic in repo['topics']:
+                        topics_map[topic] = topics_map.get(topic, 0) + 1
+
+            # return the top5 languages
+            top5_languages = sorted(languages_map, key=languages_map.get, reverse=True)[:5]
+            top_topics = sorted(topics_map, key=topics_map.get, reverse=True)[:7]
+            language_topics = {"languages" : list(top5_languages), "topics" : list(top_topics)}
 
     except aiohttp.ClientError as e:
         print(f"Error fetching data: {e}")