Update app.py

app.py

@@ -13,11 +13,11 @@ from collections import OrderedDict
 
 import gradio as gr
 
-# Detect if running on Hugging Face Spaces (so we don't set share=True there)
+# Detect if running on Hugging Face Spaces (don't use share=True there)
 IN_SPACES = bool(os.getenv("SPACE_ID") or os.getenv("HF_SPACE_ID"))
 
 # ---- Optional NLTK pieces (no downloads at startup) ----
-# We try to use real stopwords & lemmatizer if available; otherwise fall back.
+# Use real stopwords/lemmatizer if available; otherwise fall back.
 try:
     import nltk  # noqa: F401
     from nltk.corpus import stopwords as nltk_stopwords
@@ -39,15 +39,13 @@ except Exception:
         _stemmer = PorterStemmer()
         _use_porter = True
     except Exception:
-        # Last-resort fallback: identity lemmatizer
         _lemmatizer = None
         _use_porter = None
 
 
 # ---- Pipeline helpers ----
 def tokenize(text: str):
-    # Simple, dependency-free tokenizer:
-    # split into "word" blocks and single non-space symbols to preserve punctuation step
+    # Simple, dependency-free tokenizer: words or single non-space symbols
     return re.findall(r"\w+|[^\w\s]", text or "", flags=re.UNICODE)
 
 def remove_non_ascii(tokens):
@@ -75,7 +73,6 @@ def lemmatize_list(tokens):
     elif _use_porter is False:
         return [_lemmatizer.lemmatize(w) for w in tokens]
     else:
-        # identity if no lemmatizer/stemmer
         return tokens
 
 
@@ -105,7 +102,7 @@ def normalize(text: str) -> OrderedDict:
     return steps
 
 
-# ---- Gradio wiring (keeps your original Interface style) ----
+# ---- Gradio wiring ----
 examples = [
     "The quick brown fox jumps over the lazy dog!",
     "NLTK is a leading platform for building Python programs to work with human language data.",
@@ -115,14 +112,14 @@ examples = [
 
 def show_steps(text):
     steps = normalize(text)
-    html = []
+    parts = []
     for step, value in steps.items():
         if isinstance(value, list):
             pretty = " ".join(value)
-            html.append(f"<b>{step}</b>: {pretty} <small>({len(value)} tokens)</small>")
+            parts.append(f"<b>{step}</b>: {pretty} <small>({len(value)} tokens)</small>")
         else:
-            html.append(f"<b>{step}</b>: {value}")
-    return "<br>".join(html)
+            parts.append(f"<b>{step}</b>: {value}")
+    return "<br>".join(parts)
 
 iface = gr.Interface(
     fn=show_steps,
@@ -130,11 +127,9 @@ iface = gr.Interface(
     outputs=gr.HTML(label="Step-by-step normalization"),
     examples=[[ex] for ex in examples],
     title="Text Normalization Pipeline",
-    description="Enter text or select an example to see each step of the normalization process.",
+    description="Enter text or select an example to see each step of the normalization process."
 )
 
 if __name__ == "__main__":
-    iface.launch(
-        server_name="0.0.0.0",
-        server_port=7860,
-        share=not
+    # share=True only when running locally (avoids Spaces warning)
+    iface.launch(server_name="0.0.0.0", server_port=7860, share=(not IN_SPACES))