Update annotations.py

annotations.py

@@ -11,8 +11,11 @@ def extract_pdf_text(file) -> str:
         doc = fitz.open(stream=file.read(), filetype="pdf") if not isinstance(file, str) else fitz.open(file)
         full_text = ""
         for page in doc:
-            full_text += page.get_text("text") + "\n"
+            text = page.get_text("text")
+            full_text += text + "\n"
+            print(f"Extracted text from page {page.number + 1}: {len(text)} characters")
         doc.close()
+        print(f"Total extracted text length: {len(full_text)} characters")
         return full_text
     except Exception as e:
         print(f"Error extracting text from PDF: {e}")
@@ -20,23 +23,28 @@ def extract_pdf_text(file) -> str:
 
 def check_language_issues(full_text: str) -> Dict[str, Any]:
     """Check for language issues using LanguageTool."""
-    language_tool = language_tool_python.LanguageTool('en-US')
-    matches = language_tool.check(full_text)
-    issues = []
-    for match in matches:
-        issues.append({
-            "message": match.message,
-            "context": match.context,
-            "suggestions": match.replacements[:3] if match.replacements else [],
-            "category": match.category,
-            "rule_id": match.ruleId,
-            "offset": match.offset,
-            "length": match.errorLength
-        })
-    return {
-        "total_issues": len(issues),
-        "issues": issues
-    }
+    try:
+        language_tool = language_tool_python.LanguageTool('en-US')
+        matches = language_tool.check(full_text)
+        issues = []
+        for match in matches:
+            issues.append({
+                "message": match.message,
+                "context": match.context,
+                "suggestions": match.replacements[:3] if match.replacements else [],
+                "category": match.category,
+                "rule_id": match.ruleId,
+                "offset": match.offset,
+                "length": match.errorLength
+            })
+        print(f"Total language issues found: {len(issues)}")
+        return {
+            "total_issues": len(issues),
+            "issues": issues
+        }
+    except Exception as e:
+        print(f"Error checking language issues: {e}")
+        return {"error": str(e)}
 
 def highlight_issues_in_pdf(file, language_matches: List[Dict[str, Any]]) -> bytes:
     """
@@ -47,26 +55,29 @@ def highlight_issues_in_pdf(file, language_matches: List[Dict[str, Any]]) -> byt
     try:
         # Open the PDF
         doc = fitz.open(stream=file.read(), filetype="pdf") if not isinstance(file, str) else fitz.open(file)
+        print(f"Opened PDF with {len(doc)} pages.")
 
         # Extract words with positions from each page
         word_list = []  # List of tuples: (page_number, word, x0, y0, x1, y1)
-        page_text = ""
         for page_number in range(len(doc)):
             page = doc[page_number]
             words = page.get_text("words")  # List of tuples: (x0, y0, x1, y1, "word", block_no, line_no, word_no)
             for w in words:
                 word_text = w[4]
                 word_list.append((page_number, word_text, w[0], w[1], w[2], w[3]))
-        
+            print(f"Page {page_number + 1}: Extracted {len(words)} words.")
+
         # Concatenate all words to form the full text
         concatenated_text = " ".join([w[1] for w in word_list])
-        
+        print(f"Concatenated text length: {len(concatenated_text)} characters.")
+
         # Iterate over each language issue
-        for issue in language_matches:
+        for idx, issue in enumerate(language_matches, 1):
             offset = issue["offset"]
             length = issue["length"]
             error_text = concatenated_text[offset:offset+length]
-
+            print(f"\nIssue {idx}: '{error_text}' at offset {offset} with length {length}")
+            
             # Find the words that fall within the error span
             current_pos = 0
             target_words = []
@@ -78,22 +89,28 @@ def highlight_issues_in_pdf(file, language_matches: List[Dict[str, Any]]) -> byt
                     target_words.append(word)
                 current_pos += word_length
 
+            if not target_words:
+                print("No matching words found for this issue.")
+                continue
+
             # Add highlight annotations to the target words
             for target in target_words:
                 page_num, word_text, x0, y0, x1, y1 = target
                 page = doc[page_num]
-                # Define a rectangle around the word
-                rect = fitz.Rect(x0, y0, x1, y1)
+                # Define a rectangle around the word with some padding
+                rect = fitz.Rect(x0 - 1, y0 - 1, x1 + 1, y1 + 1)
                 # Add a highlight annotation
                 highlight = page.add_highlight_annot(rect)
                 highlight.set_colors(stroke=(1, 1, 0))  # Yellow color
                 highlight.update()
+                print(f"Highlighted '{word_text}' on page {page_num + 1} at position ({x0}, {y0}, {x1}, {y1})")
 
         # Save annotated PDF to bytes
         byte_stream = io.BytesIO()
         doc.save(byte_stream)
         annotated_pdf_bytes = byte_stream.getvalue()
         doc.close()
+        print("Annotated PDF successfully created.")
         return annotated_pdf_bytes
     except Exception as e:
         print(f"Error in highlighting PDF: {e}")
@@ -107,8 +124,12 @@ def analyze_pdf(file) -> Tuple[Dict[str, Any], bytes]:
             return {"error": "Failed to extract text from PDF."}, None
 
         language_issues = check_language_issues(full_text)
+        if "error" in language_issues:
+            return language_issues, None
+
         issues = language_issues.get("issues", [])
         annotated_pdf = highlight_issues_in_pdf(file, issues) if issues else None
         return language_issues, annotated_pdf
     except Exception as e:
+        print(f"Error in analyze_pdf: {e}")
         return {"error": str(e)}, None