Update app.py

app.py

@@ -14,10 +14,8 @@ def preprocess_capitalization(text: str) -> str:
     processed_words = []
     
     for word in words:
-        # Check if the word is an acronym (all uppercase letters)
         if re.match(r"^[A-Z]+$", word):
             processed_words.append(word)  # Leave acronyms unchanged
-        # Check if the word has mixed capitalization (e.g., "HEllo")
         elif re.search(r"[A-Z]", word) and re.search(r"[a-z]", word):
             processed_words.append(word[0].upper() + word[1:].lower())  # Correct capitalization
         else:
@@ -25,11 +23,11 @@ def preprocess_capitalization(text: str) -> str:
     
     return " ".join(processed_words)
 
-def preprocess_text(text: str):
+def preprocess_text(text: str, is_spell_corrected: bool = False):
     """Process text and return corrections with position information."""
     result = {
         "spell_suggestions": [],
-        "other_suggestions": [],
+        "other_suggestions": [],  # Added to distinguish style suggestions
         "entities": [],
         "tags": []
     }
@@ -39,42 +37,27 @@ def preprocess_text(text: str):
     if capitalized_text != text:
         result["spell_suggestions"].append({
             "original": text,
-            "corrected": capitalized_text,
-            "type": "spell"
+            "corrected": capitalized_text
         })
         text = capitalized_text  # Update text for further processing
 
-    # Transformer spell check - only for words that look misspelled
-    spell_checked = spell_checker(text, max_length=512)[0]['generated_text']
-    if spell_checked != text:
-        spell_words = spell_checked.split()
-        text_words = text.split()
-        
-        # Only suggest spell corrections for words that are actually different
-        for orig, corrected in zip(text_words, spell_words):
-            if orig.lower() != corrected.lower():  # Compare case-insensitive
-                result["spell_suggestions"].append({
-                    "original": orig,
-                    "corrected": corrected,
-                    "type": "spell"
-                })
+    # Transformer spell check, skipped if already spell-corrected
+    if not is_spell_corrected:
+        spell_checked = spell_checker(text, max_length=512)[0]['generated_text']
+        if spell_checked != text:
+            result["spell_suggestions"].append({
+                "original": text,
+                "corrected": spell_checked
+            })
+            text = spell_checked  # Update text after spell correction
 
-    # Add fluency/style suggestions (other suggestions)
-    # Only add if the word isn't already in spell suggestions
-    spell_originals = {s["original"] for s in result["spell_suggestions"]}
+    # Add style suggestions (other_suggestions)
     words = text.split()
-    
     for word in words:
-        # Skip words that have spell suggestions or already have punctuation
-        if word in spell_originals or word.endswith(('.', '?', '!')):
-            continue
-            
-        # Add style suggestions (like adding punctuation)
-        if not word.endswith('!'):  # Only suggest adding exclamation if it doesn't already have one
+        if not word.endswith(('.', '?', '!')):  # Skip if already punctuated
             result["other_suggestions"].append({
                 "original": word,
-                "corrected": word + "!",
-                "type": "other"
+                "corrected": word + "!"
             })
 
     # Add entities and tags
@@ -84,51 +67,25 @@ def preprocess_text(text: str):
 
     return text, result
 
-def translate_text(text: str):
-    """Just translate the text without preprocessing."""
-    client = Client("Frenchizer/space_21")
-    try:
-        translation = client.predict(text)
-        # Return an empty suggestions object since we're just translating
-        result = {
-            "spell_suggestions": [],
-            "other_suggestions": [],
-            "entities": [],
-            "tags": []
-        }
-        return translation, result
-    except Exception as e:
-        return f"Error: {str(e)}", {}
-
-def preprocess_and_forward(text: str):
+def preprocess_and_forward(text: str, is_spell_corrected: bool = False):
     """Process text and forward to translation service."""
-    original_text, preprocessing_result = preprocess_text(text)
+    original_text, preprocessing_result = preprocess_text(text, is_spell_corrected)
     
     # Forward original text to translation service
-    client = Client("Frenchizer/space_21")
+    client = Client("Frenchizer/space_17")
     try:
         translation = client.predict(original_text)
         return translation, preprocessing_result
     except Exception as e:
         return f"Error: {str(e)}", preprocessing_result
 
-def translate_only(text: str):
-    """Endpoint that only does translation without preprocessing or suggestions."""
-    translation, empty_result = translate_text(text)
-    return translation, empty_result
-
 # Gradio interface
 with gr.Blocks() as demo:
-    with gr.Tab("Main"):
-        input_text = gr.Textbox(label="Input Text")
-        output_text = gr.Textbox(label="Output Text")
-        preprocess_button = gr.Button("Process and Translate")
-        preprocess_button.click(fn=preprocess_and_forward, inputs=input_text, outputs=output_text)
-    with gr.Tab("Translation Only"):
-        translate_input = gr.Textbox(label="Input Text")
-        translate_output = gr.Textbox(label="Output Text")
-        translate_button = gr.Button("Translate")
-        translate_button.click(fn=translate_only, inputs=translate_input, outputs=translate_output)
-                                
+    input_text = gr.Textbox(label="Input Text")
+    is_spell_corrected = gr.Checkbox(label="Spell Corrected", value=False, visible=False)  # Hidden flag
+    output_text = gr.Textbox(label="Output Text")
+    preprocess_button = gr.Button("Process")
+    preprocess_button.click(fn=preprocess_and_forward, inputs=[input_text, is_spell_corrected], outputs=[output_text])
+
 if __name__ == "__main__":
     demo.launch()