Fix: enhance type check robustness with mixed String/Enum types

This commit is contained in:
少卿 2025-12-09 19:30:27 +08:00
parent 0d298cd5c9
commit 1bc77ddb22

View file

@ -849,7 +849,11 @@ class MinerUParser(RAGFlowPdfParser):
return return
img_root = subdir / "generated_images" img_root = subdir / "generated_images"
img_root.mkdir(parents=True, exist_ok=True) img_root.mkdir(parents=True, exist_ok=True)
text_types = {"text", "list", "code", "header"} text_types = {
"text", "list", "code", "header", "equation",
MinerUContentType.TEXT, MinerUContentType.LIST, MinerUContentType.CODE,
MinerUContentType.HEADER, MinerUContentType.EQUATION
}
generated = 0 generated = 0
for idx, item in enumerate(outputs): for idx, item in enumerate(outputs):
if item.get("type") not in text_types: if item.get("type") not in text_types: