Help! Anyone know if I use this api correctly?

[jason-Genesis-101]

Self Checks

• I have searched for existing issues search for existing issues, including closed ones.
• I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
• [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:)
• Please do not modify this template :) and fill in all the required fields.

1. Is this request related to a challenge you're experiencing? Tell me about your story.

I'm using the creat-by-file api to update document in a dataset. At first, I simply upload the file and choose parent-child sturcture to build the dataset. Then I want to achieve parent-child structure with full-doc in parent when I use api to update file. However, everytime I tried to update file thru this api, document structure changed to one chunk. Can anyone helps? Much appreciate!

Here is my input:

url = f"{env}/datasets/{dataset_id}/document/create-by-file"
headers = {"Authorization": f"Bearer {api_secret}"}

data = '''{
    "original_document_id": "1234",
    "indexing_technique": "high_quality",
    "doc_form": "hierarchical_model",
    "doc_language": "Chinese",
    "process_rule": {
        "mode": "custom",
        "rules": {
            "pre_processing_rules": [
                {"id": "remove_extra_spaces", "enabled": true},
                {"id": "remove_urls_emails", "enabled": false}
            ],
            "segmentation": {
                "separator": "\\n",
                "max_tokens": 10000
            },
            "parent_mode": "full-doc",
            "subchunk_segmentation":{
                "separator":"\\n",
                "max_tokens":4000
                }
            }
        },
    "retrieval_model": {
	    "search_method": "hybrid_search",
	    "reranking_enable": true,
	    "reranking_model": {
		    "reranking_provider_name": "TONGYI",
		    "reranking_model_name": "gte-rerank-v2"
	    },
	    "top_k": 3,
	    "score_threshold_enabled": false
    },
    "embedding_model": "text-embedding-v4",
    "embedding_model_provider": "TONGYI"
}'''

with open(r"file.xlsx", "rb") as file:
    files = {
        "data": (None, data, "text/plain"),
        "file": file
    }
    # Start request
    response = requests.post(url, headers=headers, files=files)

2. Additional context or comments

No response

[dosubot[bot]]

You're very close! To keep the parent-child structure (with the parent as the full document) when updating via the API, you need to set "process_rule": { "mode": "hierarchical", ... } instead of "custom". The "parent_mode" value "full-doc" isn't documented as a standard option—most examples use "paragraph" for parent_mode. If you use "custom", the API will treat the whole document as a single chunk, which is why your structure collapses to one chunk.

Try changing your payload to use "mode": "hierarchical" and "parent_mode": "paragraph" (or another supported value), and keep your segmentation/subchunk_segmentation settings. This should restore the hierarchical structure you want when updating the file. For reference, see this example and this thread.

Let me know if you need a sample payload!

To reply, just mention @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}    

[jason-Genesis-101]

@dosu How to input this code block into HTTP request node? I've tried using form-data with json block, yet it still return with 400. I've assigned a input variable to "original_document_id" in HTTP request node. Since I can't see what exactly is the reason of 400, please point it out:

data = '''{
"original_document_id": "123",
"indexing_technique": "high_quality",
"doc_form": "hierarchical_model",
"doc_language": "Chinese",
"process_rule": {
"mode": "hierarchical",
"rules": {
"pre_processing_rules": [
{"id": "remove_extra_spaces", "enabled": true},
{"id": "remove_urls_emails", "enabled": false}
],
"segmentation": {
"separator": "\n",
"max_tokens": 10000
},
"parent_mode": "full-doc",
"subchunk_segmentation":{
"separator":"\n",
"max_tokens":4000
}
}
},
"retrieval_model": {
"search_method": "hybrid_search",
"reranking_enable": true,
"reranking_model": {
"reranking_provider_name": "TONGYI",
"reranking_model_name": "gte-rerank-v2"
},
"top_k": 3,
"score_threshold_enabled": false
},
"embedding_model": "text-embedding-v4",
"embedding_model_provider": "TONGYI"
}'''

with open(r"file.xlsx", "rb") as file:
    files = {
        "data": (None, data, "text/plain"),
        "file": file
    }

Here is a reference picture:

[zxh0305]

May I ask how your create-by-file is implemented? When I use Apifox, it keeps showing 'Invalid user account,' and it also doesn't work in the HTTP node application in Dify.