Update file_utils.py

file_utils.py

@@ -1,69 +1,105 @@
-import gradio as gr
+import fitz  # PyMuPDF
+from pptx import Presentation
+from openai import OpenAI
 import os
 
+# === GPT-4 Client Setup ===
+client = OpenAI(api_key=os.getenv("OPENAI_API_KEY"))
 
-def file_dashboard():
-    with gr.Column(elem_classes="file-panel") as file_panel:
-        gr.Markdown("""
-        <div style='text-align: center; margin-bottom: 1rem;'>
-            <img src='file/images/chatter_owl.png' width='100'>
-            <h2 style='margin-top: 0.5rem;'>📄 File-Based Learning with <strong>Chatter the Owl</strong></h2>
-            <p style='margin-bottom: 1rem;'>Upload your notes, textbooks, or slides, and let me turn them into summaries, quizzes, and flashcards!</p>
-        </div>
-        """)
-
-        file_upload = gr.File(label="📁 Upload Your Study Material", file_types=[".pdf", ".txt", ".md", ".pptx"])
-        question_box = gr.Textbox(label="💬 Ask a question about the uploaded file", lines=2)
-
-        with gr.Tabs():
-            with gr.Tab("📚 Summary"):
-                summary_output = gr.Textbox(label="Generated Summary", lines=6, interactive=False)
-
-            with gr.Tab("🧠 Flashcards"):
-                with gr.Accordion("Click to view generated flashcards", open=False):
-                    flashcard_output = gr.Textbox(label=None, lines=6, interactive=False, show_label=False)
-
-            with gr.Tab("❓ Quiz"):
-                with gr.Accordion("Click to view generated quiz questions", open=False):
-                    quiz_output = gr.Textbox(label=None, lines=6, interactive=False, show_label=False)
-
-            with gr.Tab("💬 Answer to Your Question"):
-                answer_output = gr.Textbox(label="Answer", lines=4, interactive=False)
-
-        text_state = gr.State("")
-
-        def handle_file_upload(file):
-            if file is None:
-                return "", "No file uploaded.", "", "", ""
-
-            text = extract_text_from_file(file)
-            if not text or "Unsupported file type" in text:
-                return "", text, "", "", ""
-
-            summary = generate_summary(text)
-            flashcards = generate_flashcards(text)
-            quiz = generate_quiz(text)
-            return text, summary, flashcards, quiz, ""
-
-        def handle_question(text, question):
-            if not text.strip():
-                return "Please upload a file first."
-            if not question.strip():
-                return "Please enter a question."
-            return answer_question(text, question)
-
-        file_upload.change(
-            fn=handle_file_upload,
-            inputs=[file_upload],
-            outputs=[text_state, summary_output, flashcard_output, quiz_output, answer_output],
-            show_progress="minimal"
-        )
-
-        question_box.change(
-            fn=handle_question,
-            inputs=[text_state, question_box],
-            outputs=[answer_output],
-            show_progress="minimal"
-        )
-
-        return file_panel
+# === Centralized LLM Caller ===
+def call_llm(prompt, system_message="You are a helpful assistant.", temperature=0.7):
+    response = client.chat.completions.create(
+        model="gpt-4",
+        messages=[
+            {"role": "system", "content": system_message},
+            {"role": "user", "content": prompt}
+        ],
+        temperature=temperature
+    )
+    return response.choices[0].message.content
+
+# === File Parsers ===
+def extract_text_from_pdf(pdf_path):
+    text = ""
+    with fitz.open(pdf_path) as doc:
+        for page in doc:
+            text += page.get_text()
+    return text
+
+
+def extract_text_from_txt(file_path):
+    with open(file_path, 'r', encoding='utf-8') as f:
+        return f.read()
+
+
+def extract_text_from_md(file_path):
+    with open(file_path, 'r', encoding='utf-8') as f:
+        return f.read()
+
+
+def extract_text_from_pptx(file_path):
+    text = ""
+    prs = Presentation(file_path)
+    for slide in prs.slides:
+        for shape in slide.shapes:
+            if hasattr(shape, "text"):
+                text += shape.text + "\n"
+    return text
+
+
+def extract_text_from_file(file):
+    if file is None:
+        return ""
+
+    name = file.name.lower()
+    if name.endswith(".pdf"):
+        return extract_text_from_pdf(file.name)
+    elif name.endswith(".txt"):
+        return extract_text_from_txt(file.name)
+    elif name.endswith(".md"):
+        return extract_text_from_md(file.name)
+    elif name.endswith(".pptx"):
+        return extract_text_from_pptx(file.name)
+    else:
+        return "Unsupported file type."
+
+
+# === GPT-4 Powered Content Generators ===
+def generate_summary(text):
+    prompt = f"""Please summarize the following study material in a concise and organized manner:
+
+{text}
+"""
+    return call_llm(prompt, system_message="You are a helpful study assistant.")
+
+
+def generate_flashcards(text):
+    prompt = f"""Based on the content below, create a set of helpful flashcards.
+Each flashcard should be formatted as:
+Q: Question?
+A: Answer.
+
+{text}
+"""
+    return call_llm(prompt, system_message="You are a flashcard generator assistant.")
+
+
+def generate_quiz(text):
+    prompt = f"""Create a short quiz (3-5 questions) based on the content below.
+Include a mix of multiple choice and short answer questions.
+
+{text}
+"""
+    return call_llm(prompt, system_message="You are a quiz generator assistant.")
+
+
+def answer_question(text, question):
+    prompt = f"""You are an AI tutor. Answer the following question based only on the content below.
+
+Content:
+{text}
+
+Question:
+{question}
+"""
+    return call_llm(prompt, system_message="You are a helpful and accurate tutor that stays grounded in provided content.")