Fix poppler PDF processing: add comprehensive packages and multiple path attempts

apt.txt

@@ -1,4 +1,6 @@
 poppler-utils
+poppler-data
+libpoppler-dev
+libpoppler-cpp-dev
 tesseract-ocr
 libzbar0
-libpoppler-cpp-dev

pdf_comparator.py

@@ -49,21 +49,26 @@ def _is_pdf(path: str) -> bool:
 
 def load_first_page(path: str, dpi: int = 300) -> Image.Image:
     if _is_pdf(path):
-        try:
-            # Try with poppler_path explicitly set
-            imgs = convert_from_path(path, dpi=dpi, first_page=1, last_page=1, poppler_path="/usr/bin")
-            if not imgs:
-                raise ValueError(f"No pages in PDF: {path}")
-            return imgs[0].convert("RGB")
-        except Exception as e1:
+        # Try multiple poppler paths
+        poppler_paths = ["/usr/bin", "/usr/local/bin", "/bin", None]
+        
+        for poppler_path in poppler_paths:
             try:
-                # Fallback: try without explicit poppler_path
-                imgs = convert_from_path(path, dpi=dpi, first_page=1, last_page=1)
+                if poppler_path:
+                    imgs = convert_from_path(path, dpi=dpi, first_page=1, last_page=1, poppler_path=poppler_path)
+                else:
+                    imgs = convert_from_path(path, dpi=dpi, first_page=1, last_page=1)
+                
                 if not imgs:
-                    raise ValueError(f"No pages in PDF: {path}")
+                    continue
+                    
                 return imgs[0].convert("RGB")
-            except Exception as e2:
-                raise ValueError(f"Failed to convert PDF to image. Error 1: {str(e1)}. Error 2: {str(e2)}. Make sure poppler-utils is installed.")
+            except Exception as e:
+                if poppler_path is None:  # Last attempt failed
+                    raise ValueError(f"Failed to convert PDF to image with all poppler paths. Last error: {str(e)}. Make sure poppler-utils is installed.")
+                continue  # Try next path
+                
+        raise ValueError(f"No pages in PDF: {path}")
     return Image.open(path).convert("RGB")
 
 def match_sizes(a: Image.Image, b: Image.Image) -> Tuple[Image.Image, Image.Image]: