Update app.py

app.py

@@ -5,96 +5,74 @@ from sentence_transformers import SentenceTransformer
 from sklearn.decomposition import PCA
 import plotly.graph_objects as go
 import re
-
-# Load model
+import pandas as pd
+import zipfile
+import os
+
+# ------------------------
+# تنظیمات مسیر فایل‌ها
+# ------------------------
+# مسیر فایل CSV یا ZIP (در Hugging Face داخل repo قرار بده)
+csv_path = "data/stoic_quotes_full.csv"
+zip_path = "data/archive.zip"
+
+# ------------------------
+# بارگذاری مدل
+# ------------------------
 model = SentenceTransformer('paraphrase-multilingual-MiniLM-L12-v2')
 
-# تابع برای استخراج نقل‌قول‌ها از فایل
-def load_quotes_from_file(filename, philosopher):
-    try:
-        with open(filename, 'r', encoding='utf-8') as file:
-            content = file.read()
-        
-        # حذف بخش‌های متا
-        content = re.sub(r'\[file name\].*?\[file content begin\]', '', content, flags=re.DOTALL)
-        content = re.sub(r'\[file content end\].*', '', content, flags=re.DOTALL)
-        
-        # تقسیم به جملات معنادار
-        sentences = re.split(r'(?<!\w\.\w.)(?<![A-Z][a-z]\.)(?<=\.|\?|\!)\s', content)
-        
-        # فیلتر جملات کوتاه و بی‌معنی
-        meaningful_quotes = [
-            s.strip() for s in sentences 
-            if len(s.split()) > 4  # جملات حداقل 4 کلمه‌ای
-            and not s.startswith((' ', '\n')) 
-            and len(s) < 500  # جملات خیلی طولانی نباشند
-        ]
-        
-        return meaningful_quotes
-    except Exception as e:
-        print(f"Error loading {filename}: {str(e)}")
-        return []
-
-# ساختار داده‌های فلسفی جدید
-school_data = {
-    "Hegelianism": {
-        "philosophers": ["Hegel"],
-        "quotes": []
-    },
-    "Aristotelianism": {
-        "philosophers": ["Aristotle"],
-        "quotes": []
-    },
-    "Schopenhauerian": {
-        "philosophers": ["Schopenhauer"],
-        "quotes": []
-    },
-    "Nietzschean": {
-        "philosophers": ["Nietzsche"],
-        "quotes": []
-    }
-}
-
-# تابع بارگیری نقل‌قول‌ها
-def load_quotes():
-    try:
-        school_data["Hegelianism"]["quotes"] = load_quotes_from_file("Hegel.txt", "Hegel")
-        school_data["Aristotelianism"]["quotes"] = load_quotes_from_file("Aristotle.txt", "Aristotle")
-        school_data["Schopenhauerian"]["quotes"] = load_quotes_from_file("Arthur-Schopenhauer-Quotes.txt", "Schopenhauer")
-        school_data["Nietzschean"]["quotes"] = load_quotes_from_file("Friedrich-Nietzsche.txt", "Nietzsche")
-        
-        # حذف مکاتب خالی
-        for school in list(school_data.keys()):
-            if not school_data[school]["quotes"]:
-                del school_data[school]
-        
-        print(f"Loaded quotes from {len(school_data)} philosophical schools")
-    except Exception as e:
-        print(f"Error loading quotes: {str(e)}")
-
-# فراخوانی تابع بارگیری
-load_quotes()
-
-# پروفایل مکاتب
-school_profiles = {
-    "Hegelianism": {
-        "timeline": "19th century",
-        "profile": "Dialectical, Historical, Idealist"
-    },
-    "Aristotelianism": {
-        "timeline": "4th century BCE",
-        "profile": "Logical, Empirical, Teleological"
-    },
-    "Schopenhauerian": {
-        "timeline": "19th century",
-        "profile": "Pessimistic, Compassionate, Will-centered"
-    },
-    "Nietzschean": {
-        "timeline": "19th century",
-        "profile": "Existential, Will-to-Power, Übermensch"
-    }
-}
-
+# ------------------------
+# بارگذاری داده‌ها از CSV یا ZIP
+# ------------------------
+def build_school_data():
+    school_data_dynamic = {}
+
+    # حالت CSV
+    if os.path.exists(csv_path):
+        df = pd.read_csv(csv_path)
+        if "philosopher" in df.columns and "quote" in df.columns:
+            grouped = df.groupby("philosopher")["quote"].apply(list)
+            for philosopher, quotes in grouped.items():
+                school_data_dynamic[philosopher] = {
+                    "philosophers": [philosopher],
+                    "quotes": quotes
+                }
+
+    # حالت ZIP (فایل‌های txt)
+    elif os.path.exists(zip_path):
+        with zipfile.ZipFile(zip_path, 'r') as z:
+            for filename in z.namelist():
+                if filename.lower().endswith(".txt"):
+                    with z.open(filename) as f:
+                        content = f.read().decode("utf-8", errors="ignore")
+                        sentences = re.split(r'(?<=[.!?])\s+', content)
+                        quotes = [
+                            s.strip() for s in sentences
+                            if len(s.split()) > 4 and len(s) < 500
+                        ]
+                        philosopher_name = os.path.splitext(os.path.basename(filename))[0]
+                        school_data_dynamic[philosopher_name] = {
+                            "philosophers": [philosopher_name],
+                            "quotes": quotes
+                        }
+    else:
+        print("⚠ No valid dataset found.")
+    
+    return school_data_dynamic
+
+school_data = build_school_data()
+
+# ------------------------
+# پروفایل مکاتب (دلخواه: می‌تونی این رو هم داینامیک بسازی)
+# ------------------------
+school_profiles = {name: {
+        "timeline": "Unknown",
+        "profile": "No profile available"
+    } for name in school_data.keys()}
+
+# ------------------------
+# دسته‌بندی روانشناختی
+# ------------------------
 psychological_categories = [
     {
         "name": "Moral guilt",
@@ -129,6 +107,9 @@ psychological_categories = [
     }
 ]
 
+# ------------------------
+# تحلیل روانشناختی
+# ------------------------
 def psychological_analysis(text):
     text_lower = text.lower()
     results = []
@@ -146,8 +127,11 @@ def psychological_analysis(text):
         output += f"🧠 {item['name']}\n✅ If Followed: {item['followed']}\n❌ If Ignored: {item['ignored']}\n\n"
     return output
 
+# ------------------------
+# رسم نقشه مفهومی
+# ------------------------
 def create_semantic_plot(user_vec, best_school):
-    ref_quotes = school_data[best_school]["quotes"]  # تغییر اینجا برای تطابق با ساختار جدید
+    ref_quotes = school_data[best_school]["quotes"]
     quote_vecs = model.encode(ref_quotes)
     labels = [f"Ref {i+1}" for i in range(len(ref_quotes))]
 
@@ -179,6 +163,9 @@ def create_semantic_plot(user_vec, best_school):
     fig.update_layout(title="🧭 Conceptual Map", showlegend=False)
     return fig
 
+# ------------------------
+# تحلیل متن
+# ------------------------
 def analyze_text(text):
     if not text.strip():
         return "Please enter a philosophical text.", "", "", "", "", None, ""
@@ -186,7 +173,7 @@ def analyze_text(text):
     user_vec = model.encode([text])[0]
     best_school, best_score, best_match = None, -1, ""
 
-    for school, data in school_data.items():  # تغییر اینجا برای تطابق با ساختار جدید
+    for school, data in school_data.items():
         for quote in data["quotes"]:
             quote_vec = model.encode([quote])[0]
             score = cosine_similarity([user_vec], [quote_vec])[0][0]
@@ -203,6 +190,9 @@ def analyze_text(text):
 def clear_fields():
     return "", "", "", "", "", "", None, ""
 
+# ------------------------
+# رابط Gradio
+# ------------------------
 with gr.Blocks(title="Philosophical Analyzer") as demo:
     gr.Markdown("## 📝 Enter Philosophical Text")
     input_text = gr.Textbox(lines=4, placeholder="Type or paste a philosophical text...")
@@ -227,4 +217,4 @@ with gr.Blocks(title="Philosophical Analyzer") as demo:
                      outputs=[school, score, profile_box, timeline, best_quote, conceptual_map, psych_box])
     clear_btn.click(clear_fields, outputs=[input_text, school, score, profile_box, timeline, best_quote, conceptual_map, psych_box])
 
-demo.launch()
+demo.launch()