Shrujan/integrate api py

.env

@@ -1,2 +1,4 @@
 # For VS Code Python extension
 PYTHONPATH=backend/src
+GEMINI_API_KEY=AIzaSyA-id6fL6AZ84X_NE3OzoLiFSBys4aWDD8
+GEMINI_MODEL=gemini-2.0-flash

.gitignore

@@ -0,0 +1 @@
+.DS_Store

backend/.gitignore

@@ -1,3 +1,4 @@
 .venv
 **/.env
 **/__pycache__/
+.env

backend/README.md

@@ -29,7 +29,7 @@ don't conflict with the ones on your main system by running:
 ### macOS/Unix
 
 ```bash
-$ python3 -m venv .venv
+$ python3 -m venv .venbash
 $ source .venv/bin/activate
 ```
 

backend/src/api.py

@@ -1,31 +1,109 @@
-"""
-This file defines the FastAPI app for the API and all of its routes.
-To run this API, use the FastAPI CLI
-$ fastapi dev src/api.py
-"""
+from __future__ import annotations
+import io
+import os
+import re
+import json
+import textwrap
+import uuid
+from typing import List, Dict, Optional
+import fitz
 
-import random
+from dotenv import load_dotenv
+load_dotenv()  # load .env into os.environ
 
-from fastapi import FastAPI
+#import google.generativeai as genai
+from anyio import to_thread
+from google import genai
 
-# The app which manages all of the API routes
-app = FastAPI()
+from fastapi import FastAPI, UploadFile, File, Form, HTTPException
+from fastapi.middleware.cors import CORSMiddleware
+from fastapi.responses import JSONResponse, StreamingResponse
+from pydantic import BaseModel
 
+from pypdf import PdfReader
+from reportlab.lib.pagesizes import letter
+from reportlab.pdfgen import canvas
 
-# The decorator declares the function as a FastAPI route on the given path.
-# This route in particular is a GET route at "/hello" which returns the example
-# dictionary as a JSON response with the status code 200 by default.
-@app.get("/hello")
-async def hello() -> dict[str, str]:
-    """Get hello message."""
-    return {"message": "Hello from FastAPI"}
+import uvicorn
 
 
-# The route can also handle query parameters encoded in the URL after the path,
-# e.g. `/random?maximum=1000`
-# If the value isn't an integer, FastAPI will return an error response
-# with a validation error describing the invalid input.
-@app.get("/random")
-async def get_random_item(maximum: int) -> dict[str, int]:
-    """Get an item with a random ID."""
-    return {"itemId": random.randint(0, maximum)}
+# App + CORS
+app = FastAPI(title="Resume Coach API (Gemini)")
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["http://localhost:3000", "*"],  # Change as needed for prod!
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+
+
+
+# Gemini & Resume Endpoints
+
+'''GEMINI_API_KEY = os.getenv("GEMINI_API_KEY", "")
+GEMINI_MODEL = os.getenv("GEMINI_MODEL", "gemini-2.0-flash")
+
+if GEMINI_API_KEY:
+    genai.configure(api_key=GEMINI_API_KEY)'''
+
+
+
+
+
+
+@app.post("/upload")
+async def upload_pdf(file: UploadFile = File(...)):
+    # This function checks and validates the file type
+    if file.content_type != "application/pdf":
+        raise HTTPException(status_code=400, detail="Only PDF files are allowed")
+    try:
+        contents = await file.read()
+        # PDF processing function could go here
+        doc = fitz.open(stream=contents, filetype= "pdf")
+        text = ""
+        for page in doc:
+            text += page.get_text()
+
+        #print(text)
+        job_description = "Software Engineer"
+        prompt = f"""You are an expert resume reviewer with extensive experience in recruitment and career coaching. You will receive the text content extracted from a resume. Analyze this text and provide detailed, actionable critiques focusing on content quality, language, and structure.
+                    Evaluate the following aspects:
+                    Action Verbs and Language Strength: Assess the use of strong action verbs vs. passive or weak language
+                    Quantifiable Achievements: Identify where metrics and numbers should be added to demonstrate impact
+                    Keyword Optimization: Evaluate the presence of industry-relevant keywords and ATS-friendly terminology
+                    Clarity and Conciseness: Check for verbose descriptions, unclear statements, or redundant information
+                    Professional Tone: Assess grammar, punctuation, spelling, and overall professionalism
+                    Content Structure: Evaluate the logical flow and organization of information
+                    Experience Descriptions: Analyze how effectively accomplishments are communicated
+                    Skills Relevance: Review whether skills listed are specific and marketable
+                    Gaps and Missing Information: Identify where important details or context may be missing
+                    Provide each critique as a separate, specific, actionable suggestion. Format your response as follows:
+                    Each piece of feedback must be on its own line
+                    Separate each feedback item with the pipe symbol "|"
+                    Be specific and actionable in your critiques
+                    Focus only on content, language, and text-based elements
+                    Limit each critique to one clear point
+                    Do not include the pipe symbol at the beginning or end of your response
+                    Example format:
+                    Replace "responsible for" with strong action verbs like "Led," "Developed," or "Implemented"|Add quantifiable metrics to "improved sales" - specify percentage increase or dollar amount|Remove generic phrase "team player" and replace with specific examples of collaboration|Spell out acronyms on first use to ensure ATS compatibility
+                    Now analyze the resume text and provide your critiques. Here is the resume in text form: {text}. This is the job title: {job_description}"""
+
+        client = genai.Client()
+        response = client.models.generate_content(
+            model = "gemini-2.5-flash", contents = prompt
+        )
+
+        processed_result = {
+            "message": response.text,
+        }
+        return JSONResponse(content=processed_result, status_code=200)
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Error processing file: {str(e)}")
+    finally:
+        await file.close()
+
+
+if __name__ == "__main__":
+    uvicorn.run("api:app", host="0.0.0.0", port=8000, reload=True)