mirror of
https://github.com/infiniflow/ragflow.git
synced 2025-12-24 23:46:52 +08:00
5dfdbcce3a
### What problem does this PR solve? Pipeline supports parsing PPTX naively (text only). ### Type of change - [x] New Feature (non-breaking change which adds functionality)
124 lines
3.2 KiB
JSON
124 lines
3.2 KiB
JSON
{
|
|
"components": {
|
|
"File": {
|
|
"obj":{
|
|
"component_name": "File",
|
|
"params": {
|
|
}
|
|
},
|
|
"downstream": ["Parser:0"],
|
|
"upstream": []
|
|
},
|
|
"Parser:0": {
|
|
"obj": {
|
|
"component_name": "Parser",
|
|
"params": {
|
|
"setups": {
|
|
"pdf": {
|
|
"parse_method": "deepdoc",
|
|
"vlm_name": "",
|
|
"lang": "Chinese",
|
|
"suffix": [
|
|
"pdf"
|
|
],
|
|
"output_format": "json"
|
|
},
|
|
"spreadsheet": {
|
|
"suffix": [
|
|
"xls",
|
|
"xlsx",
|
|
"csv"
|
|
],
|
|
"output_format": "html"
|
|
},
|
|
"word": {
|
|
"suffix": [
|
|
"doc",
|
|
"docx"
|
|
],
|
|
"output_format": "json"
|
|
},
|
|
"slides": {
|
|
"parse_method": "presentation",
|
|
"suffix": [
|
|
"pptx"
|
|
],
|
|
"output_format": "json"
|
|
},
|
|
"markdown": {
|
|
"suffix": [
|
|
"md",
|
|
"markdown"
|
|
],
|
|
"output_format": "json"
|
|
},
|
|
"text": {
|
|
"suffix": ["txt"],
|
|
"output_format": "json"
|
|
},
|
|
"image": {
|
|
"parse_method": "vlm",
|
|
"llm_id":"glm-4.5v",
|
|
"lang": "Chinese",
|
|
"suffix": [
|
|
"jpg",
|
|
"jpeg",
|
|
"png",
|
|
"gif"
|
|
],
|
|
"output_format": "text"
|
|
},
|
|
"audio": {
|
|
"suffix": [
|
|
"da",
|
|
"wave",
|
|
"wav",
|
|
"mp3",
|
|
"aac",
|
|
"flac",
|
|
"ogg",
|
|
"aiff",
|
|
"au",
|
|
"midi",
|
|
"wma",
|
|
"realaudio",
|
|
"vqf",
|
|
"oggvorbis",
|
|
"ape"
|
|
],
|
|
"lang": "Chinese",
|
|
"llm_id": "SenseVoiceSmall",
|
|
"output_format": "json"
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"downstream": ["Splitter:0"],
|
|
"upstream": ["Begin"]
|
|
},
|
|
"Splitter:0": {
|
|
"obj": {
|
|
"component_name": "Splitter",
|
|
"params": {
|
|
"chunk_token_size": 512,
|
|
"delimiters": ["\n"],
|
|
"overlapped_percent": 0
|
|
}
|
|
},
|
|
"downstream": ["Tokenizer:0"],
|
|
"upstream": ["Parser:0"]
|
|
},
|
|
"Tokenizer:0": {
|
|
"obj": {
|
|
"component_name": "Tokenizer",
|
|
"params": {
|
|
}
|
|
},
|
|
"downstream": [],
|
|
"upstream": ["Chunker:0"]
|
|
}
|
|
},
|
|
"path": []
|
|
}
|
|
|