app.py

from fastapi import FastAPI
from pydantic import BaseModel
from typing import List
from pathlib import Path
import shutil

from langchain_docling import DoclingLoader
from langchain_docling.loader import ExportType


app = FastAPI()

resumes = []
jobs = []

UPLOAD_DIR = Path("uploads")
UPLOAD_DIR.mkdir(exist_ok=True)

@app.post("/upload")
async def upload_file(file: UploadFile = File(...)):
    file_path = UPLOAD_DIR / file.filename
    with file_path.open("wb") as buffer:
        shutil.copyfileobj(file.file, buffer)

    # result = process_with_langchain(file_path)
    
    loader = DoclingLoader(file_path=FILE_PATH, export_type=ExportType.MARKDOWN)
    docs = loader.load()
    # docs = docs.model_dump()
    return {
        "code":201,
        "message":"Request was successful.",
        "data": docs[0].model_dump()
    }

    # return {"filename": file.filename, "path": str(file_path), "status": "uploaded"}


# class InputResume(BaseModel):
#     content: str
    


# @app.post("/suggest/")
# async def suggestion(data: InputResume):
#     return {
#         "code":201,
#         "message":"Request was successful.",
#         "data": InputResume.model_dump_json()
#     }





from ranker import rank_resume 
from embeddings import rank_jobs

# Function to wrap the existing rank_resume
def process_input(job_description, resumes):
    print("[JOB DESC]", job_description)
    print("[RESUMES]", resumes)
    resumes = [r for r in resumes if r and r.strip() != ""]  # Remove empty
    if not job_description.strip() or not resumes:
        return "Please provide both job description and at least one resume."
    
    return rank_resume(job_description, resumes)[1]


def process_input_suggestion(resume, job_descriptions):
    # print("[JOB DESC]", job_description)
    # print("[RESUMES]", resumes)
    # resumes = [r for r in resumes if r and r.strip() != ""]  # Remove empty
    # if not job_description.strip() or not resumes:
    #     return "Please provide both resume and at least one job description."

    return rank_jobs(job_descriptions, resume)[1]
    
    # results = zip(*rank_jobs(resumes, job_description))
    # formatted_output = ""
    # for i, (resume, score) in enumerate(results, 1):
    #     formatted_output += f"Job #{i}:\nScore: {score:.2f}\nJob Description Snippet: {resume[:200]}...\n\n-------\n\n"
    # return formatted_output


    
app.get("/")
def read_root():
    return {"message": "Hello, World!"}


class InputData(BaseModel):
    resumes: List[str]
    job_description: str

class InputData2(BaseModel):
    job_descriptions: List[str]
    resume: str

@app.post("/rank/")
async def process_data(data: InputData):
    return dict(scores=process_input(data.job_description, data.resumes))


@app.post("/suggest/")
async def suggestion(data: InputData2):
    return {
        "scores":process_input_suggestion(data.resume, data.job_descriptions)
    }