dify Tool HTTP request, returned application/octet-stream, file type not recognized

Self Checks

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

Dify version

1.0.0

Cloud or Self Hosted

Self Hosted (Docker)

Steps to reproduce

curl --location --request POST 'http://10.27.24.233:8081/api/v1/convert/pdf/word'
--form 'fileInput=@"/Users/soya/Desktop/1219926043.pdf"'
--form 'outputFormat="docx"' The above is the request curl,postman the returned information is as follows： Content-Type | application/octet-stream Content-Disposition | form-data; name="attachment"; filename="1219926043.docx" Content-Length | 116257

The workflow is as follows：

The debug trace information is as follows: { "status_code": 200, "body": "", "headers": { "date": "Sat, 08 Mar 2025 04:42:33 GMT", "content-type": "application/octet-stream", "content-disposition": "form-data; name="attachment"; filename="1219926043.docx"", "content-length": "116257", "cache-status": "ac5c6a2e8fde;detail=mismatch", "via": "1.1 ac5c6a2e8fde (squid/6.10)", "connection": "keep-alive" }, "files": [ { "dify_model_identity": "dify__file", "id": null, "tenant_id": "5185fd0a-047c-4abf-b27f-f91ed93f6c0f", "type": "custom", "transfer_method": "tool_file", "remote_url": null, "related_id": "71aef328-2b8a-4808-b49d-5dd1079f900f", "filename": "2ff8edd9bb0d4557a2c68d2a02e98271.bin", "extension": ".bin", "mime_type": "application/octet-stream", "size": 116257, "url": "http://10.27.24.230//files/tools/71aef328-2b8a-4808-b49d-5dd1079f900f.bin?timestamp=1741408957&nonce=d0a6a819f20afe7291b99d35a98c1082&sign=T9Qu-AmUwlsvOJmomm6CrrxJaZCmQEkAl1Vc2Fwp6ng=" } ] }

✔️ Expected Behavior

{ "status_code": 200, "body": "", "headers": { "date": "Sat, 08 Mar 2025 04:42:33 GMT", "content-type": "application/octet-stream", "content-disposition": "form-data; name="attachment"; filename="1219926043.docx"", "content-length": "116257", "cache-status": "ac5c6a2e8fde;detail=mismatch", "via": "1.1 ac5c6a2e8fde (squid/6.10)", "connection": "keep-alive" }, "files": [ { "dify_model_identity": "dify__file", "id": null, "tenant_id": "5185fd0a-047c-4abf-b27f-f91ed93f6c0f", "type": "custom", "transfer_method": "tool_file", "remote_url": null, "related_id": "71aef328-2b8a-4808-b49d-5dd1079f900f", "filename": "2ff8edd9bb0d4557a2c68d2a02e98271.docx", "extension": ".docx", "mime_type": "application/octet-stream", "size": 116257, "url": "http://10.27.24.230//files/tools/71aef328-2b8a-4808-b49d-5dd1079f900f.bin?timestamp=1741408957&nonce=d0a6a819f20afe7291b99d35a98c1082&sign=T9Qu-AmUwlsvOJmomm6CrrxJaZCmQEkAl1Vc2Fwp6ng=" } ] }

❌ Actual Behavior

{ "status_code": 200, "body": "", "headers": { "date": "Sat, 08 Mar 2025 04:42:33 GMT", "content-type": "application/octet-stream", "content-disposition": "form-data; name="attachment"; filename="1219926043.docx"", "content-length": "116257", "cache-status": "ac5c6a2e8fde;detail=mismatch", "via": "1.1 ac5c6a2e8fde (squid/6.10)", "connection": "keep-alive" }, "files": [ { "dify_model_identity": "dify__file", "id": null, "tenant_id": "5185fd0a-047c-4abf-b27f-f91ed93f6c0f", "type": "custom", "transfer_method": "tool_file", "remote_url": null, "related_id": "71aef328-2b8a-4808-b49d-5dd1079f900f", "filename": "2ff8edd9bb0d4557a2c68d2a02e98271.bin", "extension": ".bin", "mime_type": "application/octet-stream", "size": 116257, "url": "http://10.27.24.230//files/tools/71aef328-2b8a-4808-b49d-5dd1079f900f.bin?timestamp=1741408957&nonce=d0a6a819f20afe7291b99d35a98c1082&sign=T9Qu-AmUwlsvOJmomm6CrrxJaZCmQEkAl1Vc2Fwp6ng=" } ] }

Mar 08 '25 04:03 fq393

How do I get the tool http request recognition extension to docx?

Mar 08 '25 04:03 fq393

To configure the tool HTTP request to recognize the file extension as .docx instead of .bin when the content type is application/octet-stream, you can implement a detection process that checks the Content-Disposition header for the filename and adjusts the file extension accordingly. Here's a Python snippet that demonstrates this approach:

import os
import re
import mimetypes
from uuid import uuid4
import magic

def guess_file_info_from_response(response):
    url = str(response.url)
    # Try to extract filename from URL
    parsed_url = urllib.parse.urlparse(url)
    url_path = parsed_url.path
    filename = os.path.basename(url_path)

    # If filename couldn't be extracted, use Content-Disposition header
    if not filename:
        content_disposition = response.headers.get("Content-Disposition")
        if content_disposition:
            filename_match = re.search(r'filename="?(.+)"?', content_disposition)
            if filename_match:
                filename = filename_match.group(1)

    # If still no filename, generate a unique one
    if not filename:
        unique_name = str(uuid4())
        filename = f"{unique_name}"

    # Guess MIME type from filename first, then URL
    mimetype, _ = mimetypes.guess_type(filename)
    if mimetype is None:
        mimetype, _ = mimetypes.guess_type(url)
    if mimetype is None:
        # If guessing fails, use Content-Type from response headers
        mimetype = response.headers.get("Content-Type", "application/octet-stream")

    # Use python-magic to guess MIME type if still unknown or generic
    if mimetype == "application/octet-stream" and magic is not None:
        try:
            mimetype = magic.from_buffer(response.content[:1024], mime=True)
        except magic.MagicException:
            pass

    extension = os.path.splitext(filename)[1]

    # Ensure filename has an extension
    if not extension:
        extension = mimetypes.guess_extension(mimetype) or ".bin"
        filename = f"{filename}{extension}"

    # Handle special cases for extensions
    if extension == ".bin" and mimetype == "application/vnd.openxmlformats-officedocument.wordprocessingml.document":
        extension = ".docx"
        filename = f"{os.path.splitext(filename)[0]}.docx"

    return {
        "filename": filename,
        "extension": extension,
        "mimetype": mimetype,
        "size": int(response.headers.get("Content-Length", -1)),
    }

This code uses the python-magic library to perform content-based MIME type detection and adjusts the file extension to .docx if necessary ^[1].

To continue talking to Dosu, mention @dosu.

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Bug Report | Other}

Mar 08 '25 04:03 dosubot[bot]

To configure the tool HTTP request to recognize the file extension as .docx instead of .bin when the content type is application/octet-stream, you can implement a detection process that checks the Content-Disposition header for the filename and adjusts the file extension accordingly. Here's a Python snippet that demonstrates this approach:要配置工具 HTTP 请求，使其在内容类型为application/octet-stream时将文件扩展名识别为.docx而不是.bin ，您可以实施一个检测过程，检查文件名的Content-Disposition标头并相应地调整文件扩展名。以下是演示此方法的 Python 代码片段：

import os import re import mimetypes from uuid import uuid4 import magic

def guess_file_info_from_response(response): url = str(response.url) # Try to extract filename from URL parsed_url = urllib.parse.urlparse(url) url_path = parsed_url.path filename = os.path.basename(url_path)
# If filename couldn't be extracted, use Content-Disposition header
if not filename:
    content_disposition = response.headers.get("Content-Disposition")
    if content_disposition:
        filename_match = re.search(r'filename="?(.+)"?', content_disposition)
        if filename_match:
            filename = filename_match.group(1)

# If still no filename, generate a unique one
if not filename:
    unique_name = str(uuid4())
    filename = f"{unique_name}"

# Guess MIME type from filename first, then URL
mimetype, _ = mimetypes.guess_type(filename)
if mimetype is None:
    mimetype, _ = mimetypes.guess_type(url)
if mimetype is None:
    # If guessing fails, use Content-Type from response headers
    mimetype = response.headers.get("Content-Type", "application/octet-stream")

# Use python-magic to guess MIME type if still unknown or generic
if mimetype == "application/octet-stream" and magic is not None:
    try:
        mimetype = magic.from_buffer(response.content[:1024], mime=True)
    except magic.MagicException:
        pass

extension = os.path.splitext(filename)[1]

# Ensure filename has an extension
if not extension:
    extension = mimetypes.guess_extension(mimetype) or ".bin"
    filename = f"{filename}{extension}"

# Handle special cases for extensions
if extension == ".bin" and mimetype == "application/vnd.openxmlformats-officedocument.wordprocessingml.document":
    extension = ".docx"
    filename = f"{os.path.splitext(filename)[0]}.docx"

return {
    "filename": filename,
    "extension": extension,
    "mimetype": mimetype,
    "size": int(response.headers.get("Content-Length", -1)),
}
This code uses the python-magic library to perform content-based MIME type detection and adjusts the file extension to .docx if necessary [1].此代码使用python-magic库执行基于内容的 MIME 类型检测，并在必要时将文件扩展名调整为.docx [1] 。

To continue talking to Dosu, mention @dosu.要继续与Dosu交谈，请提及。

Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Bug Report | Other帮助 Dosu 学习！提供反馈：很好的回应|不相关的答案|不正确的来源|太冗长|幻觉|错误报告|其他

How to configure the detection process?

Mar 08 '25 13:03 fq393

To configure the detection process to ensure the tool HTTP request recognizes the file extension as .docx instead of .bin when the content type is application/octet-stream, you can follow these steps:

Extract the Filename: First, try to extract the filename from the URL or the Content-Disposition header.
Determine MIME Type: Use mimetypes.guess_type() to determine the MIME type based on the filename and URL.
Content-Based Detection: If the MIME type is still unknown or generic (application/octet-stream), use the python-magic library to perform content-based MIME type detection.
Adjust File Extension: Ensure the filename has the correct extension. Specifically, if the MIME type indicates a .docx file, adjust the extension accordingly.

Here's a Python snippet demonstrating this approach:

import os
import re
import mimetypes
from uuid import uuid4
import magic

def guess_file_info_from_response(response):
    url = str(response.url)
    parsed_url = urllib.parse.urlparse(url)
    url_path = parsed_url.path
    filename = os.path.basename(url_path)

    if not filename:
        content_disposition = response.headers.get("Content-Disposition")
        if content_disposition:
            filename_match = re.search(r'filename="?(.+)"?', content_disposition)
            if filename_match:
                filename = filename_match.group(1)

    if not filename:
        unique_name = str(uuid4())
        filename = f"{unique_name}"

    mimetype, _ = mimetypes.guess_type(filename)
    if mimetype is None:
        mimetype, _ = mimetypes.guess_type(url)
    if mimetype is None:
        mimetype = response.headers.get("Content-Type", "application/octet-stream")

    if mimetype == "application/octet-stream" and magic is not None:
        try:
            mimetype = magic.from_buffer(response.content[:1024], mime=True)
        except magic.MagicException:
            pass

    extension = os.path.splitext(filename)[1]

    if not extension:
        extension = mimetypes.guess_extension(mimetype) or ".bin"
        filename = f"{filename}{extension}"

    if extension == ".bin" and mimetype == "application/vnd.openxmlformats-officedocument.wordprocessingml.document":
        extension = ".docx"
        filename = f"{os.path.splitext(filename)[0]}.docx"

    return {
        "filename": filename,
        "extension": extension,
        "mimetype": mimetype,
        "size": int(response.headers.get("Content-Length", -1)),
    }

This code ensures that if the MIME type is detected as application/vnd.openxmlformats-officedocument.wordprocessingml.document, the file extension is set to .docx ^[1].

To continue talking to Dosu, mention @dosu.

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Bug Report | Other}

Mar 08 '25 13:03 dosubot[bot]