Add HTML support to DocumentIntelligenceConverter (#1352)

This commit is contained in:
safen0s
2025-08-26 22:34:43 +01:00
committed by GitHub
parent b6e5da8874
commit ea1a3dfb60
2 changed files with 31 additions and 0 deletions

View File

@@ -84,6 +84,9 @@ def _get_mime_type_prefixes(types: List[DocumentIntelligenceFileType]) -> List[s
prefixes.append(
"application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
)
elif type_ == DocumentIntelligenceFileType.HTML:
prefixes.append("text/html")
prefixes.append("application/xhtml+xml")
elif type_ == DocumentIntelligenceFileType.PDF:
prefixes.append("application/pdf")
prefixes.append("application/x-pdf")
@@ -119,6 +122,8 @@ def _get_file_extensions(types: List[DocumentIntelligenceFileType]) -> List[str]
extensions.append(".bmp")
elif type_ == DocumentIntelligenceFileType.TIFF:
extensions.append(".tiff")
elif type_ == DocumentIntelligenceFileType.HTML:
extensions.append(".html")
return extensions

View File

@@ -0,0 +1,26 @@
import io
from markitdown.converters._doc_intel_converter import (
DocumentIntelligenceConverter,
DocumentIntelligenceFileType,
)
from markitdown._stream_info import StreamInfo
def _make_converter(file_types):
conv = DocumentIntelligenceConverter.__new__(DocumentIntelligenceConverter)
conv._file_types = file_types
return conv
def test_docintel_accepts_html_extension():
conv = _make_converter([DocumentIntelligenceFileType.HTML])
stream_info = StreamInfo(mimetype=None, extension=".html")
assert conv.accepts(io.BytesIO(b""), stream_info)
def test_docintel_accepts_html_mimetype():
conv = _make_converter([DocumentIntelligenceFileType.HTML])
stream_info = StreamInfo(mimetype="text/html", extension=None)
assert conv.accepts(io.BytesIO(b""), stream_info)
stream_info = StreamInfo(mimetype="application/xhtml+xml", extension=None)
assert conv.accepts(io.BytesIO(b""), stream_info)