h

SYS/html_table.py

@@ -19,6 +19,7 @@ from typing import Any, Dict, List, Optional, Tuple
 from lxml import html as lxml_html
 from urllib.parse import urljoin
 import re
+from SYS.logger import debug
 
 # Default xpaths for candidate result containers
 _DEFAULT_XPATHS = [
@@ -50,8 +51,8 @@ def _text_or_img_title(el) -> str:
         imgs = el.xpath('.//img/@title')
         if imgs:
             return str(imgs[0]).strip()
-    except Exception:
-        pass
+    except Exception as exc:
+        debug("Failed to retrieve img title from element: %s", exc, exc_info=True)
     return (el.text_content() or "").strip()
 
 
@@ -66,7 +67,8 @@ def find_candidate_nodes(doc_or_html: Any, xpaths: Optional[List[str]] = None) -
             found = doc.xpath(xp)
             if found:
                 return list(found), xp
-        except Exception:
+        except Exception as exc:
+            debug("Failed to execute xpath %s: %s", xp, exc, exc_info=True)
             continue
     return [], None
 
@@ -214,7 +216,8 @@ def extract_records(doc_or_html: Any, base_url: Optional[str] = None, xpaths: Op
                 df = max(dfs, key=lambda d: getattr(d, "shape", (len(getattr(d, 'index', [])), 0))[0])
                 try:
                     rows = df.to_dict("records")
-                except Exception:
+                except Exception as exc:
+                    debug("pandas HTML table parse: df.to_dict('records') failed: %s", exc, exc_info=True)
                     # Some DataFrame-like objects may have slightly different APIs
                     rows = [dict(r) for r in df]
 
@@ -240,13 +243,13 @@ def extract_records(doc_or_html: Any, base_url: Optional[str] = None, xpaths: Op
                             href = anchors.get(rec["title"])
                             if href:
                                 rec["path"] = urljoin(base_url, href) if base_url else href
-                except Exception:
-                    pass
+                except Exception as exc:
+                    debug("pandas: failed to recover anchor hrefs for table rows: %s", exc, exc_info=True)
 
                 return records, "pandas"
-        except Exception:
+        except Exception as exc:
             # Pandas not present or parsing failed; fall back to node parsing
-            pass
+            debug("pandas: not available or parsing failed: %s", exc, exc_info=True)
 
     # Fallback to node-based parsing
     nodes, chosen = find_candidate_nodes(doc_or_html, xpaths=xpaths)