Update app.py

app.py

@@ -216,7 +216,7 @@ st.markdown(f"""
 </style>
 """, unsafe_allow_html=True)
 
-# --- 3. FUNCTIONS (تبقى كما هي) ---
+# --- 3. FUNCTIONS ---
 
 def clean_isbn(isbn_str):
     """تنظيف وتصحيح ISBN"""
@@ -265,9 +265,10 @@ def search_web_context(isbn, title=None):
 
 def enhanced_ai_librarian_analysis(isbn, meta_data, web_context):
     """تحليل محسن مع دقة عالية"""
+    lang = st.session_state.language  # استخدام اللغة الحالية
     if not client or not USE_AI:
         st.info(languages.get_text("local_classification_mode", lang))
-        return fallback_local_classification(isbn, meta_data, web_context)
+        return fallback_local_classification(isbn, meta_data, web_context, lang)
     
     metadata_fetcher = EnhancedMetadataFetcher()
     enhanced_meta = metadata_fetcher.fetch_metadata(isbn)
@@ -400,13 +401,13 @@ CRITICAL: The NLM classification must be accurate and specific to the main subje
             st.warning(languages.get_text("ai_payment_warning", lang))
         else:
             st.error(languages.get_text("ai_unexpected_error", lang).format(e))
-        return fallback_local_classification(isbn, enhanced_meta, web_context)
+        return fallback_local_classification(isbn, enhanced_meta, web_context, lang)
     except Exception as e:
         st.error(languages.get_text("ai_unexpected_error", lang).format(str(e)))
-        return fallback_local_classification(isbn, enhanced_meta, web_context)
+        return fallback_local_classification(isbn, enhanced_meta, web_context, lang)
 
-def fallback_local_classification(isbn, enhanced_meta, web_context):
-    """تصنيف محلي متقدم مع بيانات افتراضية ذكية"""
+def fallback_local_classification(isbn, enhanced_meta, web_context, lang='en'):
+    """تصنيف محلي متقدم مع بيانات افتراضية ذكية وقابلية للترجمة"""
     nlm_classifier = AdvancedNLMClassifier()
     nlm_result = nlm_classifier.classify_with_confidence(
         enhanced_meta.get('title', ''), 
@@ -420,7 +421,7 @@ def fallback_local_classification(isbn, enhanced_meta, web_context):
     publisher = enhanced_meta.get('publisher', 'Unknown')
     pub_year = enhanced_meta.get('published_date', '')
     
-    # توليد ملخص ذكي
+    # توليد ملخص ذكي (يبقى بالإنجليزية)
     if description and len(description) > 20:
         summary = description[:500] + ('...' if len(description) > 500 else '')
     else:
@@ -429,7 +430,7 @@ def fallback_local_classification(isbn, enhanced_meta, web_context):
                   f"Published by {publisher} in {pub_year if pub_year else 'unknown year'}. " \
                   f"The work covers key concepts in {nlm_result['nlm_description'].lower() if nlm_result['nlm_description'] else 'medicine'}."
     
-    # توليد محتويات افتراضية بناءً على التصنيف
+    # توليد محتويات افتراضية بناءً على التصنيف (تبقى بالإنجليزية)
     main_topic = nlm_result.get('nlm_name', 'Medicine')
     contents_templates = {
         'Textbook': [
@@ -497,21 +498,72 @@ def fallback_local_classification(isbn, enhanced_meta, web_context):
             "8. Review and self-assessment"
         ]
     
-    # توليد موضوعات MeSH افتراضية
+    # توليد موضوعات MeSH مع ترجمة إذا وجدت
     mesh_mapping = {
-        'W 18': ['Education, Medical', 'Textbooks as Topic', 'Curriculum'],
-        'WB 100': ['Clinical Medicine', 'Diagnosis', 'Therapeutics'],
-        'WB 105': ['Emergency Medicine', 'Traumatology', 'Critical Care'],
-        'WO 100': ['General Surgery', 'Surgical Procedures, Operative'],
-        'WS 1': ['Pediatrics', 'Child Development', 'Adolescent Medicine'],
-        'WG': ['Cardiology', 'Cardiovascular Diseases', 'Heart Diseases'],
-        'WL': ['Neurology', 'Nervous System Diseases', 'Brain'],
-        'QS 1': ['Anatomy', 'Dissection', 'Embryology'],
-        'QV 1': ['Pharmacology', 'Pharmaceutical Preparations', 'Drug Therapy'],
-        'QZ 4': ['Pathology', 'Disease', 'Clinical Pathology'],
-        'WY 100': ['Nursing Care', 'Nursing Process', 'Clinical Nursing Research'],
-        'WA 1': ['Public Health', 'Preventive Medicine', 'Epidemiology'],
-        'WM 1': ['Psychiatry', 'Mental Disorders', 'Psychotherapy']
+        'W 18': [
+            languages.get_text("mesh_education_medical", lang),
+            languages.get_text("mesh_textbooks", lang),
+            languages.get_text("mesh_curriculum", lang)
+        ],
+        'WB 100': [
+            languages.get_text("mesh_clinical_medicine", lang),
+            languages.get_text("mesh_diagnosis", lang),
+            languages.get_text("mesh_therapeutics", lang)
+        ],
+        'WB 105': [
+            languages.get_text("mesh_emergency_medicine", lang),
+            languages.get_text("mesh_traumatology", lang),
+            languages.get_text("mesh_critical_care", lang)
+        ],
+        'WO 100': [
+            languages.get_text("mesh_general_surgery", lang),
+            languages.get_text("mesh_surgical_procedures", lang)
+        ],
+        'WS 1': [
+            languages.get_text("mesh_pediatrics", lang),
+            languages.get_text("mesh_child_development", lang),
+            languages.get_text("mesh_adolescent_medicine", lang)
+        ],
+        'WG': [
+            languages.get_text("mesh_cardiology", lang),
+            languages.get_text("mesh_cardiovascular_diseases", lang),
+            languages.get_text("mesh_heart_diseases", lang)
+        ],
+        'WL': [
+            languages.get_text("mesh_neurology", lang),
+            languages.get_text("mesh_nervous_system_diseases", lang),
+            languages.get_text("mesh_brain", lang)
+        ],
+        'QS 1': [
+            languages.get_text("mesh_anatomy", lang),
+            languages.get_text("mesh_dissection", lang),
+            languages.get_text("mesh_embryology", lang)
+        ],
+        'QV 1': [
+            languages.get_text("mesh_pharmacology", lang),
+            languages.get_text("mesh_pharmaceutical_preparations", lang),
+            languages.get_text("mesh_drug_therapy", lang)
+        ],
+        'QZ 4': [
+            languages.get_text("mesh_pathology", lang),
+            languages.get_text("mesh_disease", lang),
+            languages.get_text("mesh_clinical_pathology", lang)
+        ],
+        'WY 100': [
+            languages.get_text("mesh_nursing_care", lang),
+            languages.get_text("mesh_nursing_process", lang),
+            languages.get_text("mesh_clinical_nursing_research", lang)
+        ],
+        'WA 1': [
+            languages.get_text("mesh_public_health", lang),
+            languages.get_text("mesh_preventive_medicine", lang),
+            languages.get_text("mesh_epidemiology", lang)
+        ],
+        'WM 1': [
+            languages.get_text("mesh_psychiatry", lang),
+            languages.get_text("mesh_mental_disorders", lang),
+            languages.get_text("mesh_psychotherapy", lang)
+        ]
     }
     
     nlm_code = nlm_result['nlm_code']
@@ -521,32 +573,37 @@ def fallback_local_classification(isbn, enhanced_meta, web_context):
             mesh_subjects = subjects
             break
     if not mesh_subjects:
-        mesh_subjects = ['Medicine', 'Medical Sciences', 'Health Occupations']
+        mesh_subjects = [
+            languages.get_text("mesh_medicine", lang),
+            languages.get_text("mesh_medical_sciences", lang),
+            languages.get_text("mesh_health_occupations", lang)
+        ]
     
-    # تحديد الجمهور المستهدف
-    audience_category = "Medical Students and Healthcare Professionals"
-    audience_reason = "Based on medical content and typical audience for this subject area."
+    # تحديد الجمهور المستهدف مع الترجمة
     if 'Textbook' in nlm_result.get('nlm_name', ''):
-        audience_category = "Medical Students (Undergraduate)"
-        audience_reason = "Introductory textbook format indicates undergraduate medical education."
+        audience_category = languages.get_text("audience_undergraduate", lang)
+        audience_reason = languages.get_text("audience_reason_textbook", lang)
     elif 'Education' in nlm_result.get('nlm_name', ''):
-        audience_category = "Medical Educators and Students"
-        audience_reason = "Focus on educational methods and curriculum."
+        audience_category = languages.get_text("audience_educators", lang)
+        audience_reason = languages.get_text("audience_reason_education", lang)
     elif 'Surgery' in nlm_result.get('nlm_name', ''):
-        audience_category = "Surgical Residents and Practicing Surgeons"
-        audience_reason = "Specialized surgical content for advanced trainees and clinicians."
+        audience_category = languages.get_text("audience_surgeons", lang)
+        audience_reason = languages.get_text("audience_reason_surgery", lang)
+    else:
+        audience_category = languages.get_text("audience_medical_students", lang)
+        audience_reason = languages.get_text("audience_reason_default", lang)
     
-    # قرار الشراء
+    # قرار الشراء مع الترجمة
     score = nlm_result['confidence_score']
     if score >= 8:
-        acquisition_decision = "Highly Recommended"
-        acquisition_reason = "Strong alignment with collection development policy and high relevance."
+        acquisition_decision = languages.get_text("decision_highly_recommended", lang)
+        acquisition_reason = languages.get_text("acquisition_reason_high", lang)
     elif score >= 4:
-        acquisition_decision = "Recommended"
-        acquisition_reason = "Good fit for the collection; moderate relevance."
+        acquisition_decision = languages.get_text("decision_recommended", lang)
+        acquisition_reason = languages.get_text("acquisition_reason_medium", lang)
     else:
-        acquisition_decision = "Review Required"
-        acquisition_reason = "Limited information; further evaluation recommended."
+        acquisition_decision = languages.get_text("decision_review_required", lang)
+        acquisition_reason = languages.get_text("acquisition_reason_low", lang)
     
     return {
         'title': title,
@@ -738,10 +795,12 @@ if st.session_state.analysis_data:
             if isbns:
                 st.markdown(f"**📋 Identifiers:** {' | '.join(isbns)}")
             st.markdown("---")
-            decision = ai.get('acquisition_decision', 'Optional')
-            if "Highly" in decision:
+            decision = ai.get('acquisition_decision', languages.get_text("decision_recommended", lang))
+            if decision == languages.get_text("decision_highly_recommended", lang):
                 pill_class = "status-high"
-            elif "Not" in decision:
+            elif decision == languages.get_text("decision_recommended", lang):
+                pill_class = "status-medium"
+            elif decision == languages.get_text("decision_review_required", lang) or decision == languages.get_text("decision_optional", lang):
                 pill_class = "status-low"
             else:
                 pill_class = "status-medium"