interview_logic.py

# PrepGenie/interview_logic.py
"""Core logic for the mock interview process."""

import os
import tempfile
import PyPDF2
import google.generativeai as genai
from transformers import BertTokenizer, TFBertModel
import numpy as np
import speech_recognition as sr
import soundfile as sf
import json
import matplotlib.pyplot as plt
import io
import re

# --- Configuration ---
# These could potentially be moved to a config file or environment variables
# For now, they are initialized here or passed in.
# genai.configure(api_key=os.getenv("GOOGLE_API_KEY") or "YOUR_DEFAULT_API_KEY_HERE")
# text_model = genai.GenerativeModel("gemini-1.5-flash") # This should be initialized in app.py or a central config

# --- BERT Model Loading ---
# It's generally better to load large models once. This can be handled in app.py and passed if needed,
# or loaded here if this module is imported once at startup.
# For simplicity, we'll handle loading here, assuming it's imported once.
try:
    model = TFBertModel.from_pretrained("bert-base-uncased")
    tokenizer = BertTokenizer.from_pretrained("bert-base-uncased")
    BERT_AVAILABLE = True
    print("BERT model loaded successfully in interview_logic.")
except Exception as e:
    print(f"Warning: Could not load BERT model/tokenizer in interview_logic: {e}")
    BERT_AVAILABLE = False
    model = None
    tokenizer = None

# --- Core Logic Functions ---

def getallinfo(data, text_model):
    """Processes raw resume text into a structured overview."""
    if not data or not data.strip():
        return "No data provided or data is empty."
    text = f"""{data} is given by the user. Make sure you are getting the details like name, experience,
            education, skills of the user like in a resume. If the details are not provided return: not a resume.
            If details are provided then please try again and format the whole in a single paragraph covering all the information. """
    try:
        response = text_model.generate_content(text)
        response.resolve()
        return response.text
    except Exception as e:
        print(f"Error in getallinfo: {e}")
        return "Error processing resume data."

def file_processing(pdf_file_path):
    """Processes the uploaded PDF file given its path."""
    if not pdf_file_path or not os.path.exists(pdf_file_path):
        print(f"File path is invalid or file does not exist: {pdf_file_path}")
        return ""
    try:
        print(f"Attempting to process file at path: {pdf_file_path}")
        with open(pdf_file_path, "rb") as f:
            reader = PyPDF2.PdfReader(f)
            text = ""
            for page in reader.pages:
                text += page.extract_text() or "" # Handle None from extract_text
        return text
    except FileNotFoundError:
        error_msg = f"File not found at path: {pdf_file_path}"
        print(error_msg)
        return ""
    except PyPDF2.errors.PdfReadError as e:
        error_msg = f"Error reading PDF file {pdf_file_path}: {e}"
        print(error_msg)
        return ""
    except Exception as e:
        error_msg = f"Unexpected error processing PDF from path {pdf_file_path}: {e}"
        print(error_msg)
        return ""

def get_embedding(text):
    """Generates BERT embedding for a given text."""
    if not text or not text.strip():
         return np.zeros((1, 768))
    if not BERT_AVAILABLE or not model or not tokenizer:
        print("BERT model not available for embedding in interview_logic.")
        return np.zeros((1, 768))
    try:
        encoded_text = tokenizer(text, return_tensors="tf", truncation=True, padding=True, max_length=512)
        output = model(encoded_text)
        embedding = output.last_hidden_state[:, 0, :]
        return embedding.numpy()
    except Exception as e:
        print(f"Error getting embedding in interview_logic: {e}")
        return np.zeros((1, 768))

def generate_feedback(question, answer):
    """Calculates similarity score between question and answer."""
    if not question or not question.strip() or not answer or not answer.strip():
        return "0.00"
    try:
        question_embedding = get_embedding(question)
        answer_embedding = get_embedding(answer)
        q_emb = np.squeeze(question_embedding)
        a_emb = np.squeeze(answer_embedding)
        dot_product = np.dot(q_emb, a_emb)
        norms = np.linalg.norm(q_emb) * np.linalg.norm(a_emb)
        if norms == 0:
            similarity_score = 0.0
        else:
            similarity_score = dot_product / norms
        return f"{similarity_score:.2f}"
    except Exception as e:
        print(f"Error generating feedback in interview_logic: {e}")
        return "0.00"

def generate_questions(roles, data, text_model):
    """Generates interview questions based on resume and roles."""
    if not roles or (isinstance(roles, list) and not any(roles)) or not data or not data.strip():
        return ["Could you please introduce yourself based on your resume?"]
    questions = []
    if isinstance(roles, list):
        roles_str = ", ".join(roles)
    else:
        roles_str = str(roles)
    text = f"""If this is not a resume then return text uploaded pdf is not a resume. this is a resume overview of the candidate.
            The candidate details are in {data}. The candidate has applied for the role of {roles_str}.
            Generate questions for the candidate based on the role applied and on the Resume of the candidate.
            Not always necessary to ask only technical questions related to the role but the logic of question
            should include the job applied for because there might be some deep tech questions which the user might not know.
            Ask some personal questions too. Ask no additional questions. Don't categorize the questions.
            ask 2 questions only. directly ask the questions not anything else.
            Also ask the questions in a polite way. Ask the questions in a way that the candidate can understand the question.
            and make sure the questions are related to these metrics: Communication skills, Teamwork and collaboration,
            Problem-solving and critical thinking, Time management and organization, Adaptability and resilience."""
    try:
        response = text_model.generate_content(text)
        response.resolve()
        questions_text = response.text.strip()
        questions = [q.strip() for q in questions_text.split('\n') if q.strip()]
        if not questions:
             questions = [q.strip() for q in questions_text.split('?') if q.strip()]
        if not questions:
             questions = [q.strip() for q in questions_text.split('.') if q.strip()]
        questions = questions[:2] if questions else ["Could you please introduce yourself based on your resume?"]
    except Exception as e:
        print(f"Error generating questions in interview_logic: {e}")
        questions = ["Could you please introduce yourself based on your resume?"]
    return questions

def generate_overall_feedback(data, percent, answer, question, text_model):
    """Generates overall feedback for an answer."""
    if not data or not data.strip() or not answer or not answer.strip() or not question or not question.strip():
        return "Unable to generate feedback due to missing information."
    if isinstance(percent, float):
        percent_str = f"{percent:.2f}"
    else:
        percent_str = str(percent)
    prompt = f"""As an interviewer, provide concise feedback (max 150 words) for candidate {data}.
    Questions asked: {question} # Pass single question
    Candidate's answers: {answer}
    Score: {percent_str}
    Feedback should include:
    1. Overall performance assessment (2-3 sentences)
    2. Key strengths (2-3 points)
    3. Areas for improvement (2-3 points)
    Be honest and constructive. Do not mention the exact score, but rate the candidate out of 10 based on their answers."""
    try:
        response = text_model.generate_content(prompt)
        response.resolve()
        return response.text
    except Exception as e:
        print(f"Error generating overall feedback in interview_logic: {e}")
        return "Feedback could not be generated."

def generate_metrics(data, answer, question, text_model):
    """Generates skill metrics for an answer."""
    if not data or not data.strip() or not answer or not answer.strip() or not question or not question.strip():
        return {
            "Communication skills": 0.0, "Teamwork and collaboration": 0.0,
            "Problem-solving and critical thinking": 0.0, "Time management and organization": 0.0,
            "Adaptability and resilience": 0.0
        }
    metrics = {}
    text = f"""Here is the overview of the candidate {data}. In the interview the question asked was {question}.
    The candidate has answered the question as follows: {answer}. Based on the answers provided, give me the metrics related to:
    Communication skills, Teamwork and collaboration, Problem-solving and critical thinking, Time management and organization,
    Adaptability and resilience.
    Rules for rating:
    - Rate each skill from 0 to 10
    - If the answer is empty, 'Sorry could not recognize your voice', meaningless, or irrelevant: rate all skills as 0
    - Only provide numeric ratings without any additional text or '/10'
    - Ratings must reflect actual content quality - do not give courtesy points
    - Consider answer relevance to the specific skill being rated
    Format:
    Communication skills: [rating]
    Teamwork and collaboration: [rating]
    Problem-solving and critical thinking: [rating]
    Time management and organization: [rating]
    Adaptability and resilience: [rating]"""
    try:
        response = text_model.generate_content(text)
        response.resolve()
        metrics_text = response.text.strip()
        for line in metrics_text.split('\n'):
            if ':' in line:
                key, value_str = line.split(':', 1)
                key = key.strip()
                try:
                    value_clean = value_str.strip().split()[0]
                    value = float(value_clean)
                    metrics[key] = value
                except (ValueError, IndexError):
                    metrics[key] = 0.0
        expected_metrics = [
            "Communication skills", "Teamwork and collaboration",
            "Problem-solving and critical thinking", "Time management and organization",
            "Adaptability and resilience"
        ]
        for m in expected_metrics:
            if m not in metrics:
                metrics[m] = 0.0
    except Exception as e:
        print(f"Error generating metrics in interview_logic: {e}")
        metrics = {
            "Communication skills": 0.0, "Teamwork and collaboration": 0.0,
            "Problem-solving and critical thinking": 0.0, "Time management and organization": 0.0,
            "Adaptability and resilience": 0.0
        }
    return metrics

def getmetrics(interaction, resume, text_model):
    """Gets overall metrics from AI based on interaction."""
    interaction_text = "\n".join([f"{q}: {a}" for q, a in interaction.items()])
    text = f"""This is the user's resume: {resume}.
    And here is the interaction of the interview: {interaction_text}.
    Please evaluate the interview based on the interaction and the resume.
    Rate me the following metrics on a scale of 1 to 10. 1 being the lowest and 10 being the highest.
    Communication skills, Teamwork and collaboration, Problem-solving and critical thinking,
    Time management and organization, Adaptability and resilience. Just give the ratings for the metrics.
    I do not need the feedback. Just the ratings in the format:
    Communication skills: X
    Teamwork and collaboration: Y
    Problem-solving and critical thinking: Z
    Time management and organization: A
    Adaptability and resilience: B
    """
    try:
        response = text_model.generate_content(text)
        response.resolve()
        return response.text
    except Exception as e:
        print(f"Error fetching metrics from AI in interview_logic: {e}")
        return ""

def parse_metrics(metric_text):
    """Parses raw metric text into a dictionary."""
    metrics = {
        "Communication skills": 0,
        "Teamwork and collaboration": 0,
        "Problem-solving and critical thinking": 0,
        "Time management and organization": 0,
        "Adaptability and resilience": 0
    }
    if not metric_text:
        return metrics
    for line in metric_text.split("\n"):
        if ":" in line:
            key, value = line.split(":", 1)
            key = key.strip()
            value = value.strip()
            if value and value not in ['N/A', 'nan'] and not value.isspace():
                try:
                    numbers = re.findall(r'\d+\.?\d*', value)
                    if numbers:
                        metrics[key] = int(float(numbers[0]))
                    else:
                        metrics[key] = 0
                except (ValueError, IndexError, TypeError):
                    print(f"Warning: Could not parse metric value '{value}' for '{key}' in interview_logic. Setting to 0.")
                    metrics[key] = 0
            else:
                metrics[key] = 0
    return metrics

def create_metrics_chart(metrics_dict):
    """Creates a pie chart image from metrics."""
    try:
        labels = list(metrics_dict.keys())
        sizes = list(metrics_dict.values())
        if not any(sizes):
             fig, ax = plt.subplots(figsize=(4, 4))
             ax.text(0.5, 0.5, 'No Data Available', ha='center', va='center', transform=ax.transAxes)
             ax.axis('off')
        else:
            fig, ax = plt.subplots(figsize=(6, 6))
            wedges, texts, autotexts = ax.pie(sizes, labels=labels, autopct='%1.1f%%', startangle=90)
            ax.axis('equal')
            for autotext in autotexts:
                autotext.set_color('white')
                autotext.set_fontsize(8)
        buf = io.BytesIO()
        plt.savefig(buf, format='png', bbox_inches='tight')
        buf.seek(0)
        plt.close(fig)
        return buf
    except Exception as e:
        print(f"Error creating chart in interview_logic: {e}")
        fig, ax = plt.subplots(figsize=(4, 4))
        ax.text(0.5, 0.5, 'Chart Error', ha='center', va='center', transform=ax.transAxes)
        ax.axis('off')
        buf = io.BytesIO()
        plt.savefig(buf, format='png')
        buf.seek(0)
        plt.close(fig)
        return buf

def generate_evaluation_report(metrics_data, average_rating, feedback_list, interaction_dict):
    """Generates a formatted evaluation report."""
    try:
        report_lines = [f"## Hey Candidate, here is your interview evaluation:\n"]
        report_lines.append("### Skill Ratings:\n")
        for metric, rating in metrics_data.items():
            report_lines.append(f"* **{metric}:** {rating}/10\n")
        report_lines.append(f"\n### Overall Average Rating: {average_rating:.2f}/10\n")
        report_lines.append("### Feedback Summary:\n")
        if feedback_list:
            last_feedback = feedback_list[-1] if feedback_list else "No feedback available."
            report_lines.append(last_feedback)
        else:
             report_lines.append("No detailed feedback was generated.")
        report_lines.append("\n### Interview Interaction:\n")
        if interaction_dict:
            for q, a in interaction_dict.items():
                report_lines.append(f"* **{q}**\n  {a}\n")
        else:
             report_lines.append("Interaction data not available.")
        improvement_content = """
### Areas for Improvement:
*   **Communication:** Focus on clarity, conciseness, and tailoring your responses to the audience. Use examples and evidence to support your points.
*   **Teamwork and collaboration:** Highlight your teamwork skills through specific examples and demonstrate your ability to work effectively with others.
*   **Problem-solving and critical thinking:** Clearly explain your problem-solving approach and thought process. Show your ability to analyze information and arrive at logical solutions.
*   **Time management and organization:** Emphasize your ability to manage time effectively and stay organized during challenging situations.
*   **Adaptability and resilience:** Demonstrate your ability to adapt to new situations and overcome challenges. Share examples of how you have handled unexpected situations or setbacks in the past.
**Remember:** This is just a starting point. Customize the feedback based on the specific strengths and weaknesses identified in your interview.
"""
        report_lines.append(improvement_content)
        report_text = "".join(report_lines)
        return report_text
    except Exception as e:
        error_msg = f"Error generating evaluation report in interview_logic: {e}"
        print(error_msg)
        return error_msg

# --- Interview State Management Functions ---
# These functions operate on the interview_state dictionary

def process_resume_logic(file_obj):
    """Handles resume upload and processing logic."""
    print(f"process_resume_logic called with: {file_obj}")
    if not file_obj:
        return {
            "status": "Please upload a PDF resume.",
            "processed_data": "",
            "ui_updates": {
                "role_selection": "gr_hide", "start_interview_btn": "gr_hide",
                "question_display": "gr_hide", "answer_instructions": "gr_hide",
                "audio_input": "gr_hide", "submit_answer_btn": "gr_hide",
                "next_question_btn": "gr_hide", "submit_interview_btn": "gr_hide",
                "answer_display": "gr_hide", "feedback_display": "gr_hide",
                "metrics_display": "gr_hide"
            }
        }
    try:
        if hasattr(file_obj, 'name'):
            file_path = file_obj.name
        else:
            file_path = str(file_obj)
        print(f"File path to process: {file_path}")
        raw_text = file_processing(file_path)
        print(f"Raw text extracted (length: {len(raw_text) if raw_text else 0})")
        if not raw_text or not raw_text.strip():
             print("Failed to extract text or text is empty.")
             return {
                "status": "Could not extract text from the PDF.",
                "processed_data": "",
                "ui_updates": {
                    "role_selection": "gr_hide", "start_interview_btn": "gr_hide",
                    "question_display": "gr_hide", "answer_instructions": "gr_hide",
                    "audio_input": "gr_hide", "submit_answer_btn": "gr_hide",
                    "next_question_btn": "gr_hide", "submit_interview_btn": "gr_hide",
                    "answer_display": "gr_hide", "feedback_display": "gr_hide",
                    "metrics_display": "gr_hide"
                }
             }
        # processed_data = getallinfo(raw_text, text_model) # text_model needs to be passed
        # Placeholder, actual call in app.py
        return {
            "status": f"File processed successfully!",
            "processed_data": raw_text, # Return raw text, let app.py call getallinfo
            "ui_updates": {
                "role_selection": "gr_show", "start_interview_btn": "gr_show",
                "question_display": "gr_hide", "answer_instructions": "gr_hide",
                "audio_input": "gr_hide", "submit_answer_btn": "gr_hide",
                "next_question_btn": "gr_hide", "submit_interview_btn": "gr_hide",
                "answer_display": "gr_hide", "feedback_display": "gr_hide",
                "metrics_display": "gr_hide"
            }
        }
    except Exception as e:
        error_msg = f"Error processing file in interview_logic: {str(e)}"
        print(error_msg)
        import traceback
        traceback.print_exc()
        return {
            "status": error_msg,
            "processed_data": "",
            "ui_updates": {
                "role_selection": "gr_hide", "start_interview_btn": "gr_hide",
                "question_display": "gr_hide", "answer_instructions": "gr_hide",
                "audio_input": "gr_hide", "submit_answer_btn": "gr_hide",
                "next_question_btn": "gr_hide", "submit_interview_btn": "gr_hide",
                "answer_display": "gr_hide", "feedback_display": "gr_hide",
                "metrics_display": "gr_hide"
            }
        }

def start_interview_logic(roles, processed_resume_data, text_model):
    """Starts the interview process logic."""
    if not roles or (isinstance(roles, list) and not any(roles)) or not processed_resume_data or not processed_resume_data.strip():
        return {
            "status": "Please select a role and ensure resume is processed.",
            "initial_question": "",
            "interview_state": {},
            "ui_updates": {
                "audio_input": "gr_hide", "submit_answer_btn": "gr_hide", "next_question_btn": "gr_hide",
                "submit_interview_btn": "gr_hide", "feedback_display": "gr_hide", "metrics_display": "gr_hide",
                "question_display": "gr_hide", "answer_instructions": "gr_hide"
            }
        }
    try:
        questions = generate_questions(roles, processed_resume_data, text_model)
        initial_question = questions[0] if questions else "Could you please introduce yourself?"
        interview_state = {
            "questions": questions,
            "current_q_index": 0,
            "answers": [],
            "feedback": [],
            "interactions": {},
            "metrics_list": [],
            "resume_data": processed_resume_data,
            "selected_roles": roles # Store roles for history
        }
        return {
            "status": "Interview started. Please answer the first question.",
            "initial_question": initial_question,
            "interview_state": interview_state,
            "ui_updates": {
                "audio_input": "gr_show", "submit_answer_btn": "gr_show", "next_question_btn": "gr_show",
                "submit_interview_btn": "gr_hide", "feedback_display": "gr_hide", "metrics_display": "gr_hide",
                "question_display": "gr_show", "answer_instructions": "gr_show"
            }
        }
    except Exception as e:
        error_msg = f"Error starting interview in interview_logic: {str(e)}"
        print(error_msg)
        return {
            "status": error_msg,
            "initial_question": "",
            "interview_state": {},
            "ui_updates": {
                "audio_input": "gr_hide", "submit_answer_btn": "gr_hide", "next_question_btn": "gr_hide",
                "submit_interview_btn": "gr_hide", "feedback_display": "gr_hide", "metrics_display": "gr_hide",
                "question_display": "gr_hide", "answer_instructions": "gr_hide"
            }
        }

def submit_answer_logic(audio, interview_state, text_model):
    """Handles submitting an answer via audio logic."""
    if not audio or not interview_state:
        return {
            "status": "No audio recorded or interview not started.",
            "answer_text": "",
            "interview_state": interview_state,
            "feedback_text": "",
            "metrics": {},
            "ui_updates": {
                "feedback_display": "gr_hide", "metrics_display": "gr_hide",
                "audio_input": "gr_show", "submit_answer_btn": "gr_show", "next_question_btn": "gr_show",
                "submit_interview_btn": "gr_hide", "question_display": "gr_show", "answer_instructions": "gr_show"
            }
        }
    try:
        temp_dir = tempfile.mkdtemp()
        audio_file_path = os.path.join(temp_dir, "recorded_audio.wav")
        sample_rate, audio_data = audio
        sf.write(audio_file_path, audio_data, sample_rate)
        r = sr.Recognizer()
        with sr.AudioFile(audio_file_path) as source:
            audio_data_sr = r.record(source)
        answer_text = r.recognize_google(audio_data_sr)
        print(f"Recognized Answer: {answer_text}")
        os.remove(audio_file_path)
        os.rmdir(temp_dir)
        interview_state["answers"].append(answer_text)
        current_q_index = interview_state["current_q_index"]
        current_question = interview_state["questions"][current_q_index]
        interview_state["interactions"][f"Q{current_q_index + 1}: {current_question}"] = f"A{current_q_index + 1}: {answer_text}"
        percent_str = generate_feedback(current_question, answer_text)
        try:
            percent = float(percent_str)
        except ValueError:
            percent = 0.0
        feedback_text = generate_overall_feedback(interview_state["resume_data"], percent_str, answer_text, current_question, text_model)
        interview_state["feedback"].append(feedback_text)
        metrics = generate_metrics(interview_state["resume_data"], answer_text, current_question, text_model)
        interview_state["metrics_list"].append(metrics)
        interview_state["current_q_index"] += 1
        return {
            "status": f"Answer submitted: {answer_text}",
            "answer_text": answer_text,
            "interview_state": interview_state,
            "feedback_text": feedback_text,
            "metrics": metrics,
            "ui_updates": {
                "feedback_display": "gr_show_and_update", "metrics_display": "gr_show_and_update",
                "audio_input": "gr_show", "submit_answer_btn": "gr_show", "next_question_btn": "gr_show",
                "submit_interview_btn": "gr_hide", "question_display": "gr_show", "answer_instructions": "gr_show"
            }
        }
    except Exception as e:
        print(f"Error processing audio answer in interview_logic: {e}")
        return {
            "status": "Error processing audio. Please try again.",
            "answer_text": "",
            "interview_state": interview_state,
            "feedback_text": "",
            "metrics": {},
            "ui_updates": {
                "feedback_display": "gr_hide", "metrics_display": "gr_hide",
                "audio_input": "gr_show", "submit_answer_btn": "gr_show", "next_question_btn": "gr_show",
                "submit_interview_btn": "gr_hide", "question_display": "gr_show", "answer_instructions": "gr_show"
            }
        }

def next_question_logic(interview_state):
    """Moves to the next question or ends the interview logic."""
    if not interview_state:
        return {
            "status": "Interview not started.",
            "next_q": "",
            "interview_state": interview_state,
            "ui_updates": {
                "audio_input": "gr_show", "submit_answer_btn": "gr_show", "next_question_btn": "gr_show",
                "feedback_display": "gr_hide", "metrics_display": "gr_hide", "submit_interview_btn": "gr_hide",
                "question_display": "gr_hide", "answer_instructions": "gr_hide",
                "answer_display": "gr_clear", "metrics_display_clear": "gr_clear"
            }
        }
    current_q_index = interview_state["current_q_index"]
    total_questions = len(interview_state["questions"])
    if current_q_index < total_questions:
        next_q = interview_state["questions"][current_q_index]
        return {
            "status": f"Question {current_q_index + 1}/{total_questions}",
            "next_q": next_q,
            "interview_state": interview_state,
            "ui_updates": {
                "audio_input": "gr_show", "submit_answer_btn": "gr_show", "next_question_btn": "gr_show",
                "feedback_display": "gr_hide", "metrics_display": "gr_hide", "submit_interview_btn": "gr_hide",
                "question_display": "gr_show", "answer_instructions": "gr_show",
                "answer_display": "gr_clear", "metrics_display_clear": "gr_clear"
            }
        }
    else:
        return {
            "status": "Interview completed! Click 'Submit Interview' to see your evaluation.",
            "next_q": "Interview Finished",
            "interview_state": interview_state,
            "ui_updates": {
                "audio_input": "gr_hide", "submit_answer_btn": "gr_hide", "next_question_btn": "gr_hide",
                "feedback_display": "gr_hide", "metrics_display": "gr_hide", "submit_interview_btn": "gr_show", # Show submit button
                "question_display": "gr_show", "answer_instructions": "gr_hide",
                "answer_display": "gr_clear", "metrics_display_clear": "gr_clear"
            }
        }

def submit_interview_logic(interview_state, text_model):
    """Handles final submission, triggers evaluation, prepares results logic."""
    if not interview_state or not isinstance(interview_state, dict):
        return {
            "status": "Interview state is missing or invalid.",
            "interview_state": interview_state,
            "report_text": "",
            "chart_buffer": None,
            "ui_updates": {
                "evaluation_report_display": "gr_hide", "evaluation_chart_display": "gr_hide"
            }
        }
    try:
        print("Interview submitted for evaluation in interview_logic.")
        interactions = interview_state.get("interactions", {})
        resume_data = interview_state.get("resume_data", "")
        feedback_list = interview_state.get("feedback", [])
        metrics_history = interview_state.get("metrics_list", [])
        # selected_roles = interview_state.get("selected_roles", []) # Not used here directly

        if not interactions:
            error_msg = "No interview interactions found to evaluate."
            print(error_msg)
            return {
                "status": error_msg,
                "interview_state": interview_state,
                "report_text": "",
                "chart_buffer": None,
                "ui_updates": {
                    "evaluation_report_display": "gr_hide", "evaluation_chart_display": "gr_hide"
                }
            }
        raw_metrics_text = getmetrics(interactions, resume_data, text_model)
        print(f"Raw Metrics Text:\n{raw_metrics_text}")
        final_metrics = parse_metrics(raw_metrics_text)
        print(f"Parsed Metrics: {final_metrics}")
        if final_metrics:
            average_rating = sum(final_metrics.values()) / len(final_metrics)
        else:
            average_rating = 0.0
        report_text = generate_evaluation_report(final_metrics, average_rating, feedback_list, interactions)
        print("Evaluation report generated in interview_logic.")
        chart_buffer = create_metrics_chart(final_metrics)
        print("Evaluation chart generated in interview_logic.")

        return {
            "status": "Evaluation Complete! See your results below.",
            "interview_state": interview_state, # Pass through
            "report_text": report_text,
            "chart_buffer": chart_buffer,
            "ui_updates": {
                "evaluation_report_display": "gr_show_and_update", "evaluation_chart_display": "gr_show_and_update"
            }
        }
    except Exception as e:
        error_msg = f"Error during evaluation submission in interview_logic: {str(e)}"
        print(error_msg)
        import traceback
        traceback.print_exc()
        return {
            "status": error_msg,
            "interview_state": interview_state,
            "report_text": error_msg,
            "chart_buffer": None,
            "ui_updates": {
                "evaluation_report_display": "gr_show_and_update_error", "evaluation_chart_display": "gr_hide"
            }
        }

# Add similar logic functions for chat if needed, or keep chat in its own module.