File size: 3,257 Bytes
bfe3429
 
 
 
 
 
0e67ca9
bfe3429
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0e67ca9
bfe3429
0e67ca9
bfe3429
 
0e67ca9
bfe3429
 
 
 
0e67ca9
bfe3429
0e67ca9
bfe3429
 
 
 
 
 
 
 
 
0e67ca9
 
 
 
 
bfe3429
0e67ca9
 
 
 
 
 
 
 
 
 
bfe3429
 
 
 
 
0e67ca9
 
 
 
 
 
 
bfe3429
 
 
 
0e67ca9
 
 
 
 
 
 
f245c77
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
import json
from typing import Dict

from core.doc_processor import extract_text_from_bytes, extract_text_from_docx
from core.text_processor import chunk_text
from core.vector_store import create_or_load_vectorstore, similarity_search
from services.llm_service import get_llm, get_unified_resume_prompt


def analyze_resume_document(
    file_bytes: bytes, job_role: str, file_extension: str
) -> Dict:
    if file_extension.lower() == ".pdf":
        resume_text = extract_text_from_bytes(file_bytes)
    elif file_extension.lower() == ".docx":
        resume_text = extract_text_from_docx(file_bytes)
    else:
        raise ValueError(f"Unsupported file format: {file_extension}")

    if resume_text.startswith("Error"):
        raise ValueError(resume_text)

    chunks = chunk_text(resume_text)
    vector_store = create_or_load_vectorstore(chunks)
    all_data = analyze_resume_with_llm(vector_store, resume_text, job_role)
    
    return all_data


def analyze_resume_with_llm(vectorstore, resume_text: str, job_role: str) -> Dict:
    try:
        retrieved_docs = similarity_search(vectorstore, job_role)
        relevant_content = "\n".join([doc.page_content for doc in retrieved_docs])

        unified_prompt = get_unified_resume_prompt()
        llm = get_llm()
        chain = unified_prompt | llm

        response = chain.invoke(
            {
                "resume_text": resume_text,
                "relevant_content": relevant_content,
                "job_role": job_role,
            }
        )

        response_text = response.content.strip()
        if response_text.startswith("`"):
            response_text = response_text.strip("`")
        if response_text.startswith("json"):
            response_text = response_text[4:]

        parsed_response = json.loads(response_text.strip())
        
        return {
            "candidate_name": parsed_response.get("candidate_name", ""),
            "university_name": parsed_response.get("university_name"),
            "location": parsed_response.get("location"),
            "physical_address": parsed_response.get("physical_address"),
            "role": parsed_response.get("role"),
            "relevant_experiences": parsed_response.get("relevant_experiences", "No relevant experience found"),
            "experience_summary": parsed_response.get("experience_summary", "No experience analysis available")
        }

    except json.JSONDecodeError as e:
        print(f"JSON Parse Error: {str(e)}")
        return {
            "candidate_name": "",
            "university_name": None,
            "location": None,
            "physical_address": None,
            "role": None,
            "relevant_experiences": "Error: Could not parse resume details from response",
            "experience_summary": "Could not parse resume details from response"
        }
    except Exception as e:
        print(f"Unexpected Error: {str(e)}")
        return {
            "candidate_name": "",
            "university_name": None,
            "location": None,
            "physical_address": None,
            "role": None,
            "relevant_experiences": f"Error analyzing resume: {str(e)}",
            "experience_summary": f"Error analyzing resume: {str(e)}"
        }