After following the Open WebUI documentation for uploading files, I successfully uploaded several files using the suggested code snippet. However, these files are not visible in the interface when typing # in the prompt field. Based on previous functionality, I expected to see all uploaded documents prefixed with # for easy selection and access. Also, if you ask the chat to find something in the uploaded documents, it thinks that there aren't any.
Use the code above to upload a file to Open WebUI.
In the prompt field, type # to bring up the list of documents.
Observe that the uploaded files are not listed.
Expected Behavior:
All uploaded files should be visible in the interface, prefixed by #, to enable quick access and reference.
Current Behavior:
Uploaded files do not appear in the document list, despite the upload operation returning success.
Environment:
Open WebUI Version: v0.3.35
OS: macOS 14.1
Additional Context:
This affects workflows that rely on the ability to quickly view and reference uploaded documents through the prompt interface. Any guidance on resolving this or confirmation of a potential bug would be greatly appreciated.
Thank you for taking a look!
Originally created by @tdiprima on GitHub (Oct 29, 2024).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/6565
**Description:**
After following the [Open WebUI documentation for uploading files](https://docs.openwebui.com/api/#uploading-files), I successfully uploaded several files using the suggested code snippet. However, these files are not visible in the interface when typing `#` in the prompt field. Based on previous functionality, I expected to see all uploaded documents prefixed with `#` for easy selection and access. Also, if you ask the chat to find something in the uploaded documents, it thinks that there aren't any.
**Code Used for Uploading Files:**
```python
import requests
def upload_file(token, file_path):
url = 'http://localhost:3000/api/v1/files/'
headers = {
'Authorization': f'Bearer {token}',
'Accept': 'application/json'
}
files = {'file': open(file_path, 'rb')}
response = requests.post(url, headers=headers, files=files)
return response.json()
```
**Steps to Reproduce:**
1. Use the code above to upload a file to Open WebUI.
2. In the prompt field, type `#` to bring up the list of documents.
3. Observe that the uploaded files are not listed.
**Expected Behavior:**
All uploaded files should be visible in the interface, prefixed by `#`, to enable quick access and reference.
**Current Behavior:**
Uploaded files do not appear in the document list, despite the upload operation returning success.
**Environment:**
- **Open WebUI Version:** v0.3.35
- **OS:** macOS 14.1
**Additional Context:**
This affects workflows that rely on the ability to quickly view and reference uploaded documents through the prompt interface. Any guidance on resolving this or confirmation of a potential bug would be greatly appreciated.
---
Thank you for taking a look!
Can you please tell me exactly which pdfs you are uploading in knowledge those pdfs are scanned or searchable/digitize?
<!-- gh-comment-id:2445151922 -->
@medrpa commented on GitHub (Oct 29, 2024):
Can you please tell me exactly which pdfs you are uploading in knowledge those pdfs are scanned or searchable/digitize?
In the knowledge base, we upload document files or directories into a collection. We then use the "#" symbol in the prompt field to select the appropriate collection.
Can you please share the screenshot?
<!-- gh-comment-id:2445178007 -->
@medrpa commented on GitHub (Oct 29, 2024):
In the knowledge base, we upload document files or directories into a collection. We then use the "#" symbol in the prompt field to select the appropriate collection.
Can you please share the screenshot?
In the knowledge base, we upload document files or directories into a collection.
Right. But I said I uploaded them via the script that was recommended in the documentation(Python Example). I haven't gotten to the Adding Files to Knowledge Collections part. But I have questions about that code also: #6553 .
<!-- gh-comment-id:2445190052 -->
@tdiprima commented on GitHub (Oct 29, 2024):
> In the knowledge base, we upload document files or directories into a collection.
Right. But I said I uploaded them via the script that was recommended in the [documentation](https://docs.openwebui.com/api/#uploading-files) **(Python Example)**. I haven't gotten to the [Adding Files to Knowledge Collections](https://docs.openwebui.com/api/#adding-files-to-knowledge-collections) part. But I have questions about that code also: #6553 .
@medrpa could you please maybe tag somebody who is familiar with Open WebUI's REST API, and bring them into the conversation?
<!-- gh-comment-id:2445260522 -->
@tdiprima commented on GitHub (Oct 29, 2024):
@medrpa could you please maybe tag somebody who is familiar with Open WebUI's REST API, and bring them into the conversation?
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Originally created by @tdiprima on GitHub (Oct 29, 2024).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/6565
Description:
After following the Open WebUI documentation for uploading files, I successfully uploaded several files using the suggested code snippet. However, these files are not visible in the interface when typing
#in the prompt field. Based on previous functionality, I expected to see all uploaded documents prefixed with#for easy selection and access. Also, if you ask the chat to find something in the uploaded documents, it thinks that there aren't any.Code Used for Uploading Files:
Steps to Reproduce:
#to bring up the list of documents.Expected Behavior:
All uploaded files should be visible in the interface, prefixed by
#, to enable quick access and reference.Current Behavior:
Uploaded files do not appear in the document list, despite the upload operation returning success.
Environment:
Additional Context:
This affects workflows that rely on the ability to quickly view and reference uploaded documents through the prompt interface. Any guidance on resolving this or confirmation of a potential bug would be greatly appreciated.
Thank you for taking a look!
@medrpa commented on GitHub (Oct 29, 2024):
Can you please tell me exactly which pdfs you are uploading in knowledge those pdfs are scanned or searchable/digitize?
@tdiprima commented on GitHub (Oct 29, 2024):
They are text documents. So yes, they are searchable.
@medrpa commented on GitHub (Oct 29, 2024):
In the knowledge base, we upload document files or directories into a collection. We then use the "#" symbol in the prompt field to select the appropriate collection.
Can you please share the screenshot?
@tdiprima commented on GitHub (Oct 29, 2024):
Right. But I said I uploaded them via the script that was recommended in the documentation (Python Example). I haven't gotten to the Adding Files to Knowledge Collections part. But I have questions about that code also: #6553 .
@tdiprima commented on GitHub (Oct 29, 2024):
@medrpa could you please maybe tag somebody who is familiar with Open WebUI's REST API, and bring them into the conversation?