Sfoglia il codice sorgente

fix: support mdx files close #11557 (#11565)

Signed-off-by: yihong0618 <zouzou0208@gmail.com>
tags/0.14.0
yihong 10 mesi fa
parent
commit
36cb25b341
Nessun account collegato all'indirizzo email del committer

+ 2
- 2
api/constants/__init__.py Vedi File





if dify_config.ETL_TYPE == "Unstructured": if dify_config.ETL_TYPE == "Unstructured":
DOCUMENT_EXTENSIONS = ["txt", "markdown", "md", "pdf", "html", "htm", "xlsx", "xls"]
DOCUMENT_EXTENSIONS = ["txt", "markdown", "md", "mdx", "pdf", "html", "htm", "xlsx", "xls"]
DOCUMENT_EXTENSIONS.extend(("docx", "csv", "eml", "msg", "pptx", "xml", "epub")) DOCUMENT_EXTENSIONS.extend(("docx", "csv", "eml", "msg", "pptx", "xml", "epub"))
if dify_config.UNSTRUCTURED_API_URL: if dify_config.UNSTRUCTURED_API_URL:
DOCUMENT_EXTENSIONS.append("ppt") DOCUMENT_EXTENSIONS.append("ppt")
DOCUMENT_EXTENSIONS.extend([ext.upper() for ext in DOCUMENT_EXTENSIONS]) DOCUMENT_EXTENSIONS.extend([ext.upper() for ext in DOCUMENT_EXTENSIONS])
else: else:
DOCUMENT_EXTENSIONS = ["txt", "markdown", "md", "pdf", "html", "htm", "xlsx", "xls", "docx", "csv"]
DOCUMENT_EXTENSIONS = ["txt", "markdown", "md", "mdx", "pdf", "html", "htm", "xlsx", "xls", "docx", "csv"]
DOCUMENT_EXTENSIONS.extend([ext.upper() for ext in DOCUMENT_EXTENSIONS]) DOCUMENT_EXTENSIONS.extend([ext.upper() for ext in DOCUMENT_EXTENSIONS])

+ 2
- 2
api/core/rag/extractor/extract_processor.py Vedi File

extractor = ExcelExtractor(file_path) extractor = ExcelExtractor(file_path)
elif file_extension == ".pdf": elif file_extension == ".pdf":
extractor = PdfExtractor(file_path) extractor = PdfExtractor(file_path)
elif file_extension in {".md", ".markdown"}:
elif file_extension in {".md", ".markdown", ".mdx"}:
extractor = ( extractor = (
UnstructuredMarkdownExtractor(file_path, unstructured_api_url, unstructured_api_key) UnstructuredMarkdownExtractor(file_path, unstructured_api_url, unstructured_api_key)
if is_automatic if is_automatic
extractor = ExcelExtractor(file_path) extractor = ExcelExtractor(file_path)
elif file_extension == ".pdf": elif file_extension == ".pdf":
extractor = PdfExtractor(file_path) extractor = PdfExtractor(file_path)
elif file_extension in {".md", ".markdown"}:
elif file_extension in {".md", ".markdown", ".mdx"}:
extractor = MarkdownExtractor(file_path, autodetect_encoding=True) extractor = MarkdownExtractor(file_path, autodetect_encoding=True)
elif file_extension in {".htm", ".html"}: elif file_extension in {".htm", ".html"}:
extractor = HtmlExtractor(file_path) extractor = HtmlExtractor(file_path)

+ 1
- 0
web/app/components/base/file-icon/index.tsx Vedi File

return <Json className={className} /> return <Json className={className} />
case 'md': case 'md':
case 'markdown': case 'markdown':
case 'mdx':
return <Md className={className} /> return <Md className={className} />
case 'pdf': case 'pdf':
return <Pdf className={className} /> return <Pdf className={className} />

+ 1
- 1
web/app/components/base/file-uploader/utils.ts Vedi File

if (extension === 'pdf') if (extension === 'pdf')
return FileAppearanceTypeEnum.pdf return FileAppearanceTypeEnum.pdf


if (extension === 'md' || extension === 'markdown')
if (extension === 'md' || extension === 'markdown' || extension === 'mdx')
return FileAppearanceTypeEnum.markdown return FileAppearanceTypeEnum.markdown


if (extension === 'xlsx' || extension === 'xls') if (extension === 'xlsx' || extension === 'xls')

+ 1
- 1
web/app/components/base/prompt-editor/constants.tsx Vedi File



export const FILE_EXTS: Record<string, string[]> = { export const FILE_EXTS: Record<string, string[]> = {
[SupportUploadFileTypes.image]: ['JPG', 'JPEG', 'PNG', 'GIF', 'WEBP', 'SVG'], [SupportUploadFileTypes.image]: ['JPG', 'JPEG', 'PNG', 'GIF', 'WEBP', 'SVG'],
[SupportUploadFileTypes.document]: ['TXT', 'MD', 'MARKDOWN', 'PDF', 'HTML', 'XLSX', 'XLS', 'DOCX', 'CSV', 'EML', 'MSG', 'PPTX', 'PPT', 'XML', 'EPUB'],
[SupportUploadFileTypes.document]: ['TXT', 'MD', 'MDX', 'MARKDOWN', 'PDF', 'HTML', 'XLSX', 'XLS', 'DOCX', 'CSV', 'EML', 'MSG', 'PPTX', 'PPT', 'XML', 'EPUB'],
[SupportUploadFileTypes.audio]: ['MP3', 'M4A', 'WAV', 'WEBM', 'AMR', 'MPGA'], [SupportUploadFileTypes.audio]: ['MP3', 'M4A', 'WAV', 'WEBM', 'AMR', 'MPGA'],
[SupportUploadFileTypes.video]: ['MP4', 'MOV', 'MPEG', 'MPGA'], [SupportUploadFileTypes.video]: ['MP4', 'MOV', 'MPEG', 'MPGA'],
} }

Loading…
Annulla
Salva