Update base_models.py

Signed-off-by: ShiroYasha18 <85089952+ShiroYasha18@users.noreply.github.com>
This commit is contained in:
ShiroYasha18 2025-05-27 01:50:53 +05:30 committed by GitHub
parent 7c7baf814d
commit 96377cb81e
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194

View File

@ -37,12 +37,11 @@ class ConversionStatus(str, Enum):
class InputFormat(str, Enum): class InputFormat(str, Enum):
"""A document format supported by document backend parsers.""" """A document format supported by document backend parsers."""
PDF = "pdf"
DOCX = "docx" DOCX = "docx"
XLSM = "xlsm"
PPTX = "pptx" PPTX = "pptx"
HTML = "html" HTML = "html"
IMAGE = "image" IMAGE = "image"
PDF = "pdf"
ASCIIDOC = "asciidoc" ASCIIDOC = "asciidoc"
MD = "md" MD = "md"
CSV = "csv" CSV = "csv"
@ -71,7 +70,7 @@ FormatToExtensions: Dict[InputFormat, List[str]] = {
InputFormat.IMAGE: ["jpg", "jpeg", "png", "tif", "tiff", "bmp"], InputFormat.IMAGE: ["jpg", "jpeg", "png", "tif", "tiff", "bmp"],
InputFormat.ASCIIDOC: ["adoc", "asciidoc", "asc"], InputFormat.ASCIIDOC: ["adoc", "asciidoc", "asc"],
InputFormat.CSV: ["csv"], InputFormat.CSV: ["csv"],
InputFormat.XLSX: ["xlsx"], InputFormat.XLSX: ["xlsx","xlsm"],
InputFormat.XML_USPTO: ["xml", "txt"], InputFormat.XML_USPTO: ["xml", "txt"],
InputFormat.JSON_DOCLING: ["json"], InputFormat.JSON_DOCLING: ["json"],
} }