youngkingdom/dify
1
0
Fork 0
mirror of https://github.com/langgenius/dify.git synced 2026-04-27 14:08:18 +08:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'main' into feat/rag-2

Browse Source
This commit is contained in:
twwu
2025-08-11 11:15:58 +08:00
parent 58aca75ee0 69c3439c3a
commit fc779d00df
214 changed files with 8987 additions and 838 deletions
Download Patch File Download Diff File Expand all files Collapse all files

54
api/core/workflow/nodes/document_extractor/node.py
View File

@ -168,7 +168,57 @@ def _extract_text_by_mime_type(*, file_content: bytes, mime_type: str) -> str:
def _extract_text_by_file_extension(*, file_content: bytes, file_extension: str) -> str:
"""Extract text from a file based on its file extension."""
match file_extension:
case ".txt" | ".markdown" | ".md" | ".html" | ".htm" | ".xml":
case (
".txt"
| ".markdown"
| ".md"
| ".html"
| ".htm"
| ".xml"
| ".c"
| ".h"
| ".cpp"
| ".hpp"
| ".cc"
| ".cxx"
| ".c++"
| ".py"
| ".js"
| ".ts"
| ".jsx"
| ".tsx"
| ".java"
| ".php"
| ".rb"
| ".go"
| ".rs"
| ".swift"
| ".kt"
| ".scala"
| ".sh"
| ".bash"
| ".bat"
| ".ps1"
| ".sql"
| ".r"
| ".m"
| ".pl"
| ".lua"
| ".vim"
| ".asm"
| ".s"
| ".css"
| ".scss"
| ".less"
| ".sass"
| ".ini"
| ".cfg"
| ".conf"
| ".toml"
| ".env"
| ".log"
| ".vtt"
):
return _extract_text_from_plain_text(file_content)
case ".json":
return _extract_text_from_json(file_content)
@ -194,8 +244,6 @@ def _extract_text_by_file_extension(*, file_content: bytes, file_extension: str)
return _extract_text_from_eml(file_content)
case ".msg":
return _extract_text_from_msg(file_content)
case ".vtt":
return _extract_text_from_vtt(file_content)
case ".properties":
return _extract_text_from_properties(file_content)
case _: