fix: replace bare excepts, use 'is None', fix file handle leak

src/evaluation/PresentQuiz/docling/backend/asciidoc_backend.py

@@ -273,14 +273,14 @@ def _parse(self, doc: DoclingDocument):
 
     def _get_current_level(self, parents):
         for k, v in parents.items():
-            if v == None and k > 0:
+            if v is None and k > 0:
                 return k - 1
 
         return 0
 
     def _get_current_parent(self, parents):
         for k, v in parents.items():
-            if v == None and k > 0:
+            if v is None and k > 0:
                 return parents[k - 1]
 
         return None
@@ -328,15 +328,15 @@ def _parse_list_item(self, line):
                     "marker": marker,
                     "text": text.strip(),
                     "numbered": False,
-                    "indent": 0 if indent == None else len(indent),
+                    "indent": 0 if indent is None else len(indent),
                 }
             else:
                 return {
                     "type": "list_item",
                     "marker": marker,
                     "text": text.strip(),
                     "numbered": True,
-                    "indent": 0 if indent == None else len(indent),
+                    "indent": 0 if indent is None else len(indent),
                 }
         else:
             # Fallback if no match

src/evaluation/PresentQuiz/docling/backend/html_backend.py

@@ -158,9 +158,9 @@ def extract_text_recursively(self, item: Tag):
                     try:
                         # Recursively get the child's text content
                         result.extend(self.extract_text_recursively(child))
-                    except:
+                    except Exception:
                         pass
-            except:
+            except Exception:
                 _log.warn("item has no children")
                 pass
 

src/evaluation/PresentQuiz/docling/backend/msexcel_backend.py

@@ -337,7 +337,7 @@ def _find_images_in_sheet(
                     image=ImageRef.from_pil(image=pil_image, dpi=72),
                     caption=None,
                 )
-            except:
+            except Exception:
                 _log.error("could not extract the image from excel sheets")
 
         """

src/evaluation/PresentQuiz/docling/backend/msword_backend.py

@@ -151,7 +151,7 @@ def prev_indent(self) -> Optional[int]:
     def get_level(self) -> int:
         """Return the first None index."""
         for k, v in self.parents.items():
-            if k >= 0 and v == None:
+            if k >= 0 and v is None:
                 return k
         return 0
 

src/evaluation/PresentQuiz/docling/models/tesseract_ocr_model.py

@@ -46,7 +46,7 @@ def __init__(self, enabled: bool, options: TesseractOcrOptions):
                 raise ImportError(install_errmsg)
             try:
                 tesseract_version = tesserocr.tesseract_version()
-            except:
+            except Exception:
                 raise ImportError(install_errmsg)
 
             _, self._tesserocr_languages = tesserocr.get_languages()

src/evaluation/PresentQuiz/docling/utils/glm_utils.py

@@ -29,7 +29,7 @@ def resolve_item(paths, obj):
 
     try:
         key = int(paths[0])
-    except:
+    except Exception:
         key = paths[0]
 
     if len(paths) == 1:

src/evaluation/PresentQuiz/utils/src/experiment/baseline_docpres.py

@@ -175,7 +175,8 @@ def generate(model: Literal["Qwen2.5", "gpt"]):
     progress = tqdm(total=len(folders))
 
     def process_folder(pdf_folder, model, processor):
-        source_text = open(f"{pdf_folder}/source.md").read()
+        with open(f"{pdf_folder}/source.md") as f:
+            source_text = f.read()
         bird_eye = json.load(open(f"{pdf_folder}/refined_doc.json"))
         images = json.load(open(f"{pdf_folder}/image_caption.json")).keys()
         output_dir = f"{pdf_folder}/docpres/{llm_name}"