### What problem does this PR solve?
Feat: Load the dialog page, prohibit calling the dialog/get interface
#6798
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
fix redis pvc in helm deployment
### What problem does this PR solve?
_Briefly describe what this PR aims to solve. Include background context
that will help reviewers understand the purpose of the PR._
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
add openai agent
_Briefly describe what this PR aims to solve. Include background context
that will help reviewers understand the purpose of the PR._
### 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: Kevin Hu <kevinhu.sh@gmail.com>
### What problem does this PR solve?
Feat: Clarify the use of OpenAI-API-compatible #6782
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
…gic to return the correct deletion message. Add handling for empty
arrays to ensure no errors occur during the deletion operation. Update
the test cases to verify the new logic.
### What problem does this PR solve?
fix this bug：https://github.com/infiniflow/ragflow/issues/6607
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
---------
Co-authored-by: wenju.li <wenju.li@deepctr.cn>
### What problem does this PR solve?
align default values in create chat assistant HTTP API dos with
implementation.
llm.presence_penalty 0.2 -> 0.4
prompt.top_n 8->6
### Type of change
- [x] Documentation Update
### What problem does this PR solve?
Fix: Issue with Markdown Code Blocks Breaking Frontend Layout #5789
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
…cument_ids" to maintain consistency.
### What problem does this PR solve?
Close#6752
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
Co-authored-by: wenju.li <wenju.li@deepctr.cn>
### What problem does this PR solve?
Fix: The file upload prompt indicates "No authorization." #6516
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Fix: Using the Enter key does not send a complete message #6754
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Feat: Support deleting knowledge graph #6747
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
Feat: Add a notification logic to the team member invite feature #6610
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
Feat: Interrupt streaming #6515
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
fix#6085
RagTokenizer's dfs_() function falls into infinite recursion when
processing text with repetitive Chinese characters (e.g.,
"一一一一一十一十一十一..." or "一一一一一一十十十十十十十二十二十二..."), causing memory leaks.
### Type of change
Implemented three optimizations to the dfs_() function:
1.Added memoization with _memo dictionary to cache computed results
2.Added recursion depth limiting with _depth parameter (max 10 levels)
3.Implemented special handling for repetitive character sequences
- [x] Bug Fix (non-breaking change which fixes an issue)
Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
I'm really sorry, I found that in graphrag/general/extractor.py under
def __call__, the line change.removed_nodes.extend(nodes[1:]) causes an
AttributeError: 'set' object has no attribute 'extend'. Could you please
merge the branch e666528 again? I made some modifications.
### What problem does this PR solve?
Feat: Allows users to search for models in the model selection drop-down
box #3221
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### Related Issue:
https://github.com/infiniflow/ragflow/issues/6653
### Environment:
Using nightly version [ece5903]
Elasticsearch database
Thanks for the review! My fault! I realize my initial testing wasn't
passed.
In graphrag/entity_resolution.py
`sub_connect_graph` is a set like` {'HELLO', 'Hi', 'How are you'}`,
Neither accessing `.nodes` nor `.nodes()` will work, **it still causes
`AttributeError: 'set' object has no attribute 'nodes'`**
In graphrag/general/extractor.py
The `list.extend() `method performs an in-place operation, directly
modifying the original list and returning ‘None’ rather than the
modified list.
Neither accessing
`sorted(set(node0_attrs[attr].extend(node1_attrs.get(attr, []))))` nor
`sorted(set(node0_attrs[attr].extend(node1_attrs[attr])))` will work,
**it still causes `TypeError: 'NoneType' object is not iterable`**
### Type of change
- [ ] Bug Fix AttributeError: graphrag/entity_resolution.py
- [ ] Bug Fix TypeError: graphrag/general/extractor.py
### Related Issue: #6653
### Environment:
Using nightly version
Elasticsearch database
### Bug Description:
When clicking the "Entity Resolution" button in KnowledgeGraph,
encountered the following errors:
graphrag/entity_resolution.py
```
list(sub_connect_graph.nodes) AttributeError
```
graphrag/general/extractor.py
```
node0_attrs[attr] = sorted(set(node0_attrs[attr].extend(node1_attrs[attr])))
TypeError: 'NoneType' object is not iterable
```
```
for attr in ["keywords", "source_id"]:
KeyError I think attribute "keywords" is in edges not nodes
```
graphrag/utils.py
```
settings.docStoreConn.delete() # Sync function called as async
```
### Changes Made:
Fixed AttributeError in entity_resolution.py by properly handling graph
nodes
Fixed TypeError and KeyError in extractor.py by separate operations
Corrected async/sync mismatch in document deletion call
### What problem does this PR solve?
- Added support for S3-compatible protocols.
- Enabled the use of knowledge base ID as a file prefix when storing
files in S3.
- Updated docker/README.md to include detailed S3 and OSS configuration
instructions.
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
https://github.com/infiniflow/ragflow/issues/6138
This PR is going to support vision llm for gpustack, modify url path
from `/v1-openai` to `/v1`
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
### Type of change
- [x] Documentation Update
---------
Co-authored-by: balibabu <cike8899@users.noreply.github.com>
### What problem does this PR solve?
Fix#6334
Hello, I encountered the same problem in #6334. In the
`api/db/db_models.py`, it calls `obj.create_table()` unconditionally in
`init_database_tables`, before the `migrate_db()`. Specially for the
`permission` field of `user_canvas` table, it has `index=True`, which
causes `peewee` to issue a SQL trying to create the index when the field
does not exist (the `user_canvas` table already exists), so
`psycopg2.errors.UndefinedColumn: column "permission" does not exist`
occurred.
I've added a judgement in the code, to only call `create_table()` when
the table does not exist, delegate the migration process to
`migrate_db()`.
Then another problem occurs: the `migrate_db()` actually does nothing
because it failed on the first migration! The `playhouse` blindly issue
DDLs without things like `IF NOT EXISTS`, so it fails... even if the
exception is `pass`, the transaction is still rolled back. So I removed
the transaction in `migrate_db()` to make it work.
### 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?
Actually fix#6241
Hello, I ran into the same problem as #6241. When I'm testing my agent
flow in the web ui using `Run` button with a file input, the retrieval
component always gave an empty output.
In the code I found that:
`web/src/pages/flow/debug-content/index.tsx`:
```tsx
const onOk = useCallback(async () => {
const values = await form.validateFields();
const nextValues = Object.entries(values).map(([key, value]) => {
const item = parameters[Number(key)];
let nextValue = value;
if (Array.isArray(value)) {
nextValue = ``;
value.forEach((x) => {
nextValue +=
x?.originFileObj instanceof File
? `${x.name}\n${x.response?.data}\n----\n` // Here, the file content always ends in '\n'
: `${x.url}\n${x.result}\n----\n`;
});
}
return { ...item, value: nextValue };
});
ok(nextValues);
}, [form, ok, parameters]);
```
while in the `agent/component/retrieval.py`:
```python
def _run(self, history, **kwargs):
query = self.get_input()
query = str(query["content"][0]) if "content" in query else ""
lines = query.split('\n') # inputs are split to ['xxx','yyy','----','']
query = lines[-1] if lines else "" # Here we always get '', thus no result
kbs = KnowledgebaseService.get_by_ids(self._param.kb_ids)
if not kbs:
return Retrieval.be_output("")
```
so the code will never got correct result.
I'm not sure why the input needs such a split here, so I just removed
the splitting, and it works well on my side.
### 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?
Fix#5418
Actually, the fix#4329 also works for agent flows with parameters, so
this PR just relaxes the `else` branch of that. With this PR, it works
fine on my side, may need more testing to make sure this does not break
something.
I guess the real problem may be deeply hidden in the code which relates
to conversation and canvas execution. After a few hours of debugging, I
see the only difference between with and without parameters in `begin`
component, is the `history` field of canvas data. When the `begin`
component contains some parameters, the debug log shows:
```
025-03-29 19:50:38,521 DEBUG 356590 {
"component_name": "Begin",
"params": {"output_var_name": "output", "message_history_window_size": 22, "query": [{"type": "fileUrls", "key": "fileUrls", "name": "files", "optional": true, "value": "问题.txt\n今天天气怎么样"}], "inputs": [], "debug_inputs": [], "prologue": "你好！ 我是你的助理，有什么可以帮到你的吗？", "output": null},
"output": null,
"inputs": []
}, history: [["user", "请回答我上传文件中的问题。"]], kwargs: {"stream": false}
2025-03-29 19:50:38,523 DEBUG 356590 {
"component_name": "Answer",
"params": {"output_var_name": "output", "message_history_window_size": 22, "query": [], "inputs": [], "debug_inputs": [], "post_answers": [], "output": null},
"output": null,
"inputs": []
}, history: [["user", "请回答我上传文件中的问题。"]], kwargs: {"stream": false}
```
Then it does not go further along the flow.
When the `begin` component does not contain any parameter, the debug log
shows:
```
2025-03-29 19:41:13,518 DEBUG 353596 {
"component_name": "Begin",
"params": {"output_var_name": "output", "message_history_window_size": 22, "query": [], "inputs": [], "debug_inputs": [], "prologue": "你好！ 我是你的助理，有什么可以帮到你的吗？", "output": null},
"output": null,
"inputs": []
}, history: [], kwargs: {"stream": false}
2025-03-29 19:41:13,520 DEBUG 353596 {
"component_name": "Answer",
"params": {"output_var_name": "output", "message_history_window_size": 22, "query": [], "inputs": [], "debug_inputs": [], "post_answers": [], "output": null},
"output": null,
"inputs": []
}, history: [], kwargs: {"stream": false}
2025-03-29 19:41:13,556 INFO 353596 127.0.0.1 - - [29/Mar/2025 19:41:13] "POST /api/v1/agents/fee6886a0c6f11f09b48eb8798e9aa9b/sessions?user_id=123 HTTP/1.1" 200 -
2025-03-29 19:41:21,115 DEBUG 353596 Canvas.prepare2run: Retrieval:LateGuestsNotice
2025-03-29 19:41:21,116 DEBUG 353596 {
"component_name": "Retrieval",
"params": {"output_var_name": "output", "message_history_window_size": 22, "query": [], "inputs": [], "debug_inputs": [], "similarity_threshold": 0.2, "keywords_similarity_weight": 0.3, "top_n": 8, "top_k": 1024, "kb_ids": ["9aca3c700c5911f0811caf35658b9385"], "rerank_id": "", "empty_response": "", "tavily_api_key": "", "use_kg": false, "output": null},
"output": null,
"inputs": []
}, history: [["user", "请回答我上传文件中的问题。"]], kwargs: {"stream": false}
```
It correctly goes along the flow and generates correct answer.
You can see the difference: when the `begin` component has any
parameter, the `history` field is filled from the beginning, while it is
just `[]` if the `begin` component has no parameter.
### 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?
Fix knowledge_graph_kwd on infinity. Close#6476 and #6624
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Fix entity_types. Close#6287 and #6608
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
This PR gives better control over how we distribute which service will
be loaded. With this approach, we can create containers to run only the
web server and others to run the task executor. It also introduces the
unique ID per task executor host, this will be important when scaling
task executors horizontally, considering unique task executor ids will
be required.
This new `entrypoint.sh` maintains the default behavior of starting the
web server and task executor in the same host.
### 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):
# Dynamic Context Window Size for Ollama Chat
## Problem Statement
Previously, the Ollama chat implementation used a fixed context window
size of 32768 tokens. This caused two main issues:
1. Performance degradation due to unnecessarily large context windows
for small conversations
2. Potential business logic failures when using smaller fixed sizes
(e.g., 2048 tokens)
## Solution
Implemented a dynamic context window size calculation that:
1. Uses a base context size of 8192 tokens
2. Applies a 1.2x buffer ratio to the total token count
3. Adds multiples of 8192 tokens based on the buffered token count
4. Implements a smart context size update strategy
## Implementation Details
### Token Counting Logic
```python
def count_tokens(text):
"""Calculate token count for text"""
# Simple calculation: 1 token per ASCII character
# 2 tokens for non-ASCII characters (Chinese, Japanese, Korean, etc.)
total = 0
for char in text:
if ord(char) < 128: # ASCII characters
total += 1
else: # Non-ASCII characters
total += 2
return total
```
### Dynamic Context Calculation
```python
def _calculate_dynamic_ctx(self, history):
"""Calculate dynamic context window size"""
# Calculate total tokens for all messages
total_tokens = 0
for message in history:
content = message.get("content", "")
content_tokens = count_tokens(content)
role_tokens = 4 # Role marker token overhead
total_tokens += content_tokens + role_tokens
# Apply 1.2x buffer ratio
total_tokens_with_buffer = int(total_tokens * 1.2)
# Calculate context size in multiples of 8192
if total_tokens_with_buffer <= 8192:
ctx_size = 8192
else:
ctx_multiplier = (total_tokens_with_buffer // 8192) + 1
ctx_size = ctx_multiplier * 8192
return ctx_size
```
### Integration in Chat Method
```python
def chat(self, system, history, gen_conf):
if system:
history.insert(0, {"role": "system", "content": system})
if "max_tokens" in gen_conf:
del gen_conf["max_tokens"]
try:
# Calculate new context size
new_ctx_size = self._calculate_dynamic_ctx(history)
# Prepare options with context size
options = {
"num_ctx": new_ctx_size
}
# Add other generation options
if "temperature" in gen_conf:
options["temperature"] = gen_conf["temperature"]
if "max_tokens" in gen_conf:
options["num_predict"] = gen_conf["max_tokens"]
if "top_p" in gen_conf:
options["top_p"] = gen_conf["top_p"]
if "presence_penalty" in gen_conf:
options["presence_penalty"] = gen_conf["presence_penalty"]
if "frequency_penalty" in gen_conf:
options["frequency_penalty"] = gen_conf["frequency_penalty"]
# Make API call with dynamic context size
response = self.client.chat(
model=self.model_name,
messages=history,
options=options,
keep_alive=60
)
return response["message"]["content"].strip(), response.get("eval_count", 0) + response.get("prompt_eval_count", 0)
except Exception as e:
return "**ERROR**: " + str(e), 0
```
## Benefits
1. **Improved Performance**: Uses appropriate context windows based on
conversation length
2. **Better Resource Utilization**: Context window size scales with
content
3. **Maintained Compatibility**: Works with existing business logic
4. **Predictable Scaling**: Context growth in 8192-token increments
5. **Smart Updates**: Context size updates are optimized to reduce
unnecessary model reloads
## Future Considerations
1. Fine-tune buffer ratio based on usage patterns
2. Add monitoring for context window utilization
3. Consider language-specific token counting optimizations
4. Implement adaptive threshold based on conversation patterns
5. Add metrics for context size update frequency
---------
Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
