Update app.py

app.py

@@ -8,6 +8,11 @@ matplotlib.use('Agg')
 import matplotlib.pyplot as plt
 import io
 from PIL import Image
+import nltk
+
+# Download punkt tokenizer if not already
+nltk.download('punkt', quiet=True)
+from nltk.tokenize import sent_tokenize
 
 logging.basicConfig(level=logging.ERROR)
 device = -1  # CPU-only
@@ -37,24 +42,33 @@ def visualize_chunk_status(chunk_data):
     return Image.open(buf)
 
 def create_summary_flowchart(summaries):
-    filtered = [s for s in summaries if s.startswith("**Chunk") and "Skipped" not in s and "Error" not in s]
+    # Filter only successful summaries
+    filtered = []
+    for s in summaries:
+        if s.startswith("**Chunk") and "Skipped" not in s and "Error" not in s:
+            parts = s.split("**:", 1)
+            if len(parts) > 1:
+                filtered.append(parts[1].strip())
+
     if not filtered:
         return None
 
-    fig, ax = plt.subplots(figsize=(6, len(filtered) * 0.8 + 1))
+    fig_height = max(2, len(filtered) * 1.5)
+    fig, ax = plt.subplots(figsize=(6, fig_height))
     ax.axis('off')
 
     ypos = list(range(len(filtered) * 2, 0, -2))
     boxprops = dict(boxstyle="round,pad=0.5", facecolor="lightblue", edgecolor="black")
 
     for i, (y, summary) in enumerate(zip(ypos, filtered)):
-        summary_text = summary.split("**Chunk")[1]
-        summary_text = summary_text.replace("**:", ":").split("\n", 1)[-1].strip()[:120]
-        ax.text(0.5, y, summary_text + ("..." if len(summary_text) > 100 else ""),
-                ha='center', va='center', bbox=boxprops, fontsize=9)
+        summary_text = summary.replace("\n", " ").strip()[:120]
+        if len(summary_text) == 120:
+            summary_text += "..."
+
+        ax.text(0.5, y, summary_text, ha='center', va='center', bbox=boxprops, fontsize=9)
 
         if i < len(filtered) - 1:
-            ax.annotate('', xy=(0.5, y - 1), xytext=(0.5, y - 0.2),
+            ax.annotate('', xy=(0.5, y - 1.5), xytext=(0.5, y - 0.5),
                         arrowprops=dict(arrowstyle="->", lw=1.5))
 
     buf = io.BytesIO()
@@ -64,6 +78,23 @@ def create_summary_flowchart(summaries):
     plt.close(fig)
     return Image.open(buf)
 
+def split_text_into_chunks(text, max_tokens=1500):
+    sentences = sent_tokenize(text)
+    chunks = []
+    current_chunk = ""
+
+    for sentence in sentences:
+        if len(current_chunk) + len(sentence) <= max_tokens:
+            current_chunk += " " + sentence
+        else:
+            chunks.append(current_chunk.strip())
+            current_chunk = sentence
+
+    if current_chunk:
+        chunks.append(current_chunk.strip())
+
+    return chunks[:20]  # Limit to 20 chunks max
+
 def summarize_file(file_bytes):
     start = time.time()
     chunk_info = []
@@ -82,7 +113,8 @@ def summarize_file(file_bytes):
     if not text.strip():
         return "❌ No text found", None, None
 
-    chunks = [text[i:i+1500] for i in range(0, min(len(text), 30000), 1500)]  # max 20 chunks
+    chunks = split_text_into_chunks(text)
+
     for i, chunk in enumerate(chunks):
         chunk_start = time.time()
         chunk_result = {'chunk': i + 1, 'status': '', 'time': 0}
@@ -127,3 +159,4 @@ if __name__ == "__main__":
 
 
 
+