Update app/core/summarizer.py

app/core/summarizer.py

@@ -1,8 +1,19 @@
+import logging
 from transformers import pipeline
 
+# Setup logging
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger("AutoTLDR")
+
+# Load summarizer
 summarizer = pipeline("summarization", model="facebook/bart-large-cnn")
 
-def chunk_text(text, max_len=700):
+# Configuration
+MAX_TEXT_LENGTH = 6000  # max total input length
+MAX_CHUNKS = 3          # max number of chunks
+CHUNK_SIZE = 700        # approx number of characters per chunk
+
+def chunk_text(text, max_len=CHUNK_SIZE):
     words = text.split()
     chunks = []
     current = []
@@ -16,18 +27,29 @@ def chunk_text(text, max_len=700):
     if current:
         chunks.append(" ".join(current))
 
-    return chunks[:3]  # Limit to first 3 chunks
+    return chunks[:MAX_CHUNKS]  # Only take the first 3 chunks
+
+def get_summary(text: str) -> str:
+    logger.info(f"Input text length: {len(text)} characters")
 
-def get_summary(text):
-    print("Input length:", len(text))
+    if not text or len(text.strip()) < 100:
+        logger.warning("Input too short to summarize.")
+        return "The text is too short to summarize."
+
+    if len(text) > MAX_TEXT_LENGTH:
+        logger.error(f"Text too long ({len(text)} chars). Aborting.")
+        return f"Input too long to summarize (max {MAX_TEXT_LENGTH} characters allowed)."
 
     summaries = []
-    for chunk in chunk_text(text):
+    chunks = chunk_text(text)
+
+    for idx, chunk in enumerate(chunks):
         try:
+            logger.info(f"Summarizing chunk {idx + 1}/{len(chunks)} (length: {len(chunk)})")
             result = summarizer(chunk, max_length=150, min_length=30, do_sample=False)
             summaries.append(result[0]['summary_text'])
         except Exception as e:
-            print("Summarization failed:", e)
-            summaries.append("...")
+            logger.exception(f"Error summarizing chunk {idx + 1}")
+            summaries.append("[Summarization failed for this section.]")
 
-    return "\n".join(summaries)
+    return "\n\n".join(summaries)