### Summary
Adopt sync.WaitGroup.Go (Go 1.25) to simplify tracked goroutine
spawning. This replaces the error-prone trio of wg.Add(1), go func(),
and defer wg.Done() with a single, self-contained call.
More info: https://github.com/golang/go/issues/63796
Signed-off-by: grandpig <grandpig@outlook.com>
### Summary
This PR fixes a Go backend bug where updating agent settings, such as
description, could clear the agent DSL.
Root cause:
PUT /api/v1/agents/:canvas_id only bound the dsl field in Go. When the
frontend submitted settings without dsl, the service still updated the
canvas with an empty DSL value.
Changes:
- Treat agent updates as partial patches.
- Preserve existing DSL when dsl is not present in the request.
- Update only specified user_canvas fields instead of saving the full
row.
- Add a regression test for settings updates preserving DSL.
Test:
`go test ./internal/service ./internal/handler`
Co-authored-by: Zhichang Yu <yuzhichang@gmail.com>
## Summary
- align the Go `/api/v1/thumbnails` endpoint with the frontend request
format for repeated `doc_ids`
- return thumbnail mappings for multiple documents instead of failing on
a single missing document
- preserve Python-compatible thumbnail formatting, including base64
thumbnail passthrough
### Summary
Fixes a bug in the Go chunk list handler where the available` query
parser rejected `false` and `0` even though they were documented as
supported values.`
This caused requests from the "Disabled" chunk filter to return HTTP 400
and broke the chunk list page when filtering disabled chunks.
### Summary
As title
This PR fixes dataset index task creation failing with unsupported data
type: entity.JSONMap when loading document chunking config.
#### issues:
```
2026/06/30 15:19:40 /home/infiniflow/Documents/development/ragflow/internal/dao/document.go:162
[error] unsupported data type: ragflow/internal/entity.JSONMap
```
#### Changes:
+ Adds the missing GORM type:longtext tag to ParserConfig in
DocumentDAO.GetChunkingConfig.
+ Adds a DAO regression test covering GetChunkingConfig joins across
document, knowledgebase, and tenant while scanning parser_config.
### Summary
As title
Before, it return `update success` but never insert or update any
metadata
fixed:
```go
_, err = s.docEngine.InsertMetadata(nil, []map[string]interface{}{
{
"id": docID,
"kb_id": doc.KbID,
"meta_fields": meta,
},
}, tenantID)
```
### What problem does this PR solve?
Closes#12962
MCPToolCallSessions created during agent execution (in `Agent.__init__`)
are never explicitly closed. Each session starts its own event loop
thread and opens an SSE/HTTP connection to the MCP server. When the
canvas goes out of scope, these threads and connections remain alive
indefinitely, accumulating over time and causing resource exhaustion
after prolonged use.
### Solution
1. Add a `Graph.close()` method that iterates all components, finds
MCPToolCallSessions held by Agent tools, and calls `close_sync()` on
each to properly shut down the event loop, thread, and connection.
2. Call `canvas.close()` in `finally` blocks after `canvas.run()`
completes in `canvas_service.py` and `canvas_app.py`.
3. Move MCP session cleanup to `finally` blocks in `test_tool` endpoint
(`mcp_server_app.py`) and `get_mcp_tools` (`api_utils.py`) to ensure
sessions are closed even on exceptions.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
---------
Co-authored-by: conflict-resolver <conflict-resolver@local>
Co-authored-by: Zhichang Yu <yuzhichang@gmail.com>
Replace fragile wall-clock timeout assertions with semantic checks for
deadline errors, retry suppression, and event ordering. Keep only
lower-bound timing checks where they prove backoff behavior. This
reduces CPU-load flakes without weakening regression coverage.
### Summary
This PR fixes incorrect dataset document counters in the Go service.
Several document creation paths inserted document records directly
through documentDAO.Create, bypassing the shared InsertDocument logic
that increments knowledgebase.doc_num. As a result, datasets could
contain documents while doc_num remained 0.
### Summary
```
/api/v1/chats/<chat_id>/sessions/<session_id>/messages/<msg_id> DELETE
/api/v1/chats/<chat_id>/sessions/<session_id>/messages/<msg_id>/feedback PUT
```
Migrates the chat session message delete and feedback APIs to the Go
server, matching the Python behavior for authorization, session
ownership checks, message/reference updates, and feedback validation.
### 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)
- [x] New Feature (non-breaking change which adds functionality)
- [x] Other (please describe):
## Summary
Agent (Canvas) runs previously did not surface token usage in the SSE
stream, and RAGFlow's own Langfuse generations for agent runs were
missing the prompt/completion split and the session/user correlation.
This made it impossible for an external caller (or Langfuse) to
reconcile an agent turn's cost with the upstream provider (e.g.
OpenRouter), because a single turn can issue several distinct LLM calls
(query rewriting / cross-language translation, multi-round tool
reasoning, nested sub-agents, and the final answer).
This PR introduces a per-run token usage sink so that **every** LLM call
in a run is aggregated and reported once, and enriches Langfuse
generations with the prompt/completion split plus session/user
attributes.
## What changes
### 1. Per-run token usage sink (`common/token_utils.py`)
- Adds two `contextvars`: `token_usage_sink` (a mutable per-run
accumulator) and `langfuse_run_attrs` (session_id/user_id for the run).
- Adds `record_run_token_usage(...)` (thread-safe via a lock, because
`thread_pool_exec` copies the context into worker threads that share the
sink dict) and `usage_from_response(...)` which extracts a
`{prompt_tokens, completion_tokens, total_tokens}` split from
OpenAI/OpenRouter-style responses.
### 2. Provider layer captures the prompt/completion split
(`rag/llm/chat_model.py`)
- `LiteLLMBase` and `Base` now store `self.last_usage`
(prompt/completion/total) for the most recent chat call, in both the
plain and tool-calling paths.
- Streaming requests set `stream_options.include_usage = True` (LiteLLM
path) so the authoritative usage arrives on the final chunk; this is
read even on the usage-only chunk that carries no `choices`.
- Fixes a multi-round accounting bug in `*_with_tools`: token totals
were **overwritten** by each round (`total_tokens = tol`) instead of
accumulated, undercounting multi-round tool conversations. Each round is
now committed to a running aggregate.
### 3. LLMBundle reports usage once, per call
(`api/db/services/llm_service.py`)
- New `_report_usage(total_tokens)` records the call's usage into the
active run sink and returns the prompt/completion/total split for
Langfuse. The split is only used when it is consistent with the
authoritative total; otherwise only the total is reported.
- All three chat entry points (`async_chat`, `async_chat_streamly`,
`async_chat_streamly_delta`) now emit `usage_details` with
`input`/`output`/`total` instead of total-only.
- `_start_langfuse_observation` now applies `session_id`/`user_id` from
the per-run context (`langfuse_run_attrs`) so agent-run generations are
correctly grouped, even though agent LLMBundles are constructed without
those attributes.
### 4. Canvas installs the sink and emits the aggregate
(`agent/canvas.py`)
- `Canvas.run()` installs a fresh `token_usage_sink` and
`langfuse_run_attrs` (from `user_id`/`session_id`) at the start of every
turn.
- `message_end` now includes an aggregated `usage` object:
`{prompt_tokens, completion_tokens, total_tokens, calls}` covering all
LLM calls in the run.
### 5. Pass session id into the run
(`api/db/services/canvas_service.py`)
- `completion()` forwards `session_id` to `Canvas.run()` for Langfuse
session correlation.
## Why a context variable
LLM calls in an agent run originate from many places that each build
their own `LLMBundle` (e.g. `cross_languages`/`keyword_extraction`
helpers, the Agent component, and nested sub-agents invoked as tools). A
run-scoped context variable is the only non-invasive chokepoint that
captures all of them exactly once, including nested agents (which run in
the same async context) and thread-pool tools (the executor copies the
context).
## Behavior / compatibility
- No public API or wire-format removal: `message_end` gains an
additional optional `usage` field; existing consumers are unaffected.
- When a provider does not return authoritative usage, behavior falls
back to the previous token estimate (total only, no split).
- Non-agent flows (Dataflow `Pipeline`, sync `Graph.run`) are untouched.
## Testing
- [x] Simple agent answer: `message_end.usage.total_tokens` matches
provider usage.
- [x] Agent with cross-language retrieval: aggregate equals the sum of
both provider calls.
- [x] Tool-calling agent (multi-round): total accumulates across rounds.
- [x] Nested agent (agent-as-tool): sub-agent tokens included in the
parent run total.
- [x] Langfuse: agent generations show input/output split and are
grouped by session/user.
---------
Co-authored-by: yzc <yuzhichang@gmail.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
## Problem
When building or updating a knowledge graph with a large number of
entities and edges, `set_graph()` in `rag/graphrag/utils.py` creates one
`asyncio` task per entity and one per edge, each calling
`embd_mdl.encode([single_name])` — a single-item HTTP request to the
embedding server.
For a graph with 17,000+ nodes and edges (real case reported in #16205),
this generates **34,000+ individual embedding API round-trips** instead
of ~266 batched calls at the default `_INSERT_BULK_SIZE=128`. The
asyncio gather over thousands of tasks makes the embedding server the
bottleneck; under load, a single slow/failed call aborts all remaining
tasks, causing the pipeline to stall and never complete.
Closes#16205. Related: #15921.
## Root Cause
```python
# Before (in set_graph, node loop):
tasks = [asyncio.create_task(graph_node_to_chunk(n, ...)) for n in nodes]
# Each task calls embd_mdl.encode([single_name]) — 1 HTTP call per node
```
`graph_node_to_chunk` checks the embed cache first, but the cache is
cold on first build, so every task makes a live API call.
## Fix
Pre-warm the embedding cache with batched calls before spawning tasks.
Each batch pre-warm calls `embd_mdl.encode(batch_of_128)` once,
populating the cache. Then every individual task hits the cache and
makes zero embedding API calls.
- Only encodes names not already in cache (no-op on warm cache / small
incremental updates)
- Uses existing project idioms: `thread_pool_exec`, `chat_limiter`,
`_INSERT_BULK_SIZE`, `get_embed_cache`, `set_embed_cache`
- Mirrors the `ENABLE_TIMEOUT_ASSERTION` timeout pattern from
`graph_node_to_chunk`
- Zero behavior change: per-task encode logic remains as a correct
fallback
## Result
| Graph size | Before | After |
|---|---|---|
| 17,576 edges | ~17,576 embedding calls → stall | ~138 batched calls |
| 17,509 nodes | ~17,509 embedding calls → stall | ~137 batched calls |
| **Total** | **~35,000 calls** | **~275 calls** |
---------
Co-authored-by: Oti_B <oti@mac.speedport.ip>
### Summary
The addition of the Context method to Go's testing.T provides
significant improvements for writing concurrent tests. It allows better
management of goroutines, ensuring they properly exit and preventing
issues like deadlocks and unfinished processes.
By using Context, errors and cancellations can be handled more
effectively, making tests more robust and easier to reason about. This
change also enables tighter integration between tests and the
application code, especially for systems that span multiple concurrent
components. Overall, it simplifies test code and enhances test stability
and maintainability.
More info: [golang/go#18368](https://github.com/golang/go/issues/18368)
Signed-off-by: blackflytech <blackflytech@outlook.com>
### Summary
`truncateText` in the `reduction` and `summarization` middlewares
truncates with `s[:maxLen]`, which slices by byte. When `maxLen` lands
inside a multi-byte character (common with CJK or other non-ASCII
content flowing through the agent), the string is cut mid-rune and the
tail byte(s) become invalid UTF-8. That broken text then goes into the
reduced context / summary prompt.
`TruncateToolResult` in the same `reduction` package already avoids this
by slicing on a rune boundary and even notes it in a comment. This PR
makes the two `truncateText` helpers do the same, so they stay
consistent with the existing helper.
Both functions keep their existing output shape (summarization still
appends `...`). Added a small unit test in each package covering ASCII
truncation and a CJK string, asserting the result stays valid UTF-8.
## Summary
TuShare required non-empty upstream input but filtered fetched news with
the static `keyword` param (default empty string), so agent-provided
keywords were ignored.
Use `self._param.keyword or ans` when filtering, matching how AkShare
uses upstream input for its query.
Fixes#16360
## Test plan
- [x] `test_tushare_filters_with_upstream_keyword_when_param_empty`
mocks the API and asserts only rows matching the upstream keyword are
returned
---------
Co-authored-by: yzc <yuzhichang@gmail.com>
Co-authored-by: Harsh Kashyap <harshkashyap@Harshs-MacBook-Pro.local>
### What problem does this PR solve?
Closes#16418.
`scholarly.search_pubs(...)` returns a **lazy generator**, but
`agent/tools/googlescholar.py` treated it as a re-iterable, bounded
list:
```python
scholar_client = scholarly.search_pubs(kwargs["query"], ...) # lazy generator
self._retrieve_chunks(scholar_client, ...) # (1) iterates -> exhausts it
self.set_output("json", list(scholar_client)) # (2) already empty -> []
```
1. **`json` output was always empty.** `_retrieve_chunks` iterates
`scholar_client`, exhausting the generator; `list(scholar_client)` then
returns `[]`.
2. **`top_n` was never applied.** Unlike `ArXiv`
(`max_results=self._param.top_n`), the unbounded generator was passed
straight to `_retrieve_chunks`, which has no internal limit — so the
tool kept paginating well past Top N (until an error, rate-limit/block,
or `COMPONENT_EXEC_TIMEOUT`).
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### Changes
- Materialize at most `top_n` results once with `itertools.islice`, and
reuse that list for both `_retrieve_chunks` and the `json` output.
- Add regression tests
(`test/unit_test/agent/component/test_googlescholar.py`, stubbing
`scholarly.search_pubs`) covering the `top_n` bound, the non-empty
`json` output, and the empty-query short-circuit.
Verified: against `main` the new tests fail with `assert 30 == 5` (top_n
ignored) and `assert 0 == 5` (empty json); with this fix all pass.
Backend-only.
---------
Co-authored-by: Zhichang Yu <yuzhichang@gmail.com>
## Summary
Add two new MCP tools to the RAGFlow MCP server:
1. **ragflow_list_datasets** - List all accessible datasets with IDs,
names, descriptions
2. **ragflow_list_chats** - List all accessible chat assistants with
IDs, names, descriptions
### Implementation
- Added `list_chats()` method to `RAGFlowConnector`
- Registered both tools in `list_tools()` and `call_tool()`
- Follows existing `ragflow_retrieval` pattern for error handling
### Usage via langchain-mcp-adapters
---------
Co-authored-by: saltsalt123 <saltsalt123@users.noreply.github.com>
Co-authored-by: yzc <yuzhichang@gmail.com>
### What problem does this PR solve?
Closes https://github.com/infiniflow/ragflow/issues/14571.
Adds CAJAL as a first-class local scientific-writing option in RAGFlow:
- registers `agnuxo/cajal-4b-p2pclaw` as a known Ollama chat model with
a 32K context setting
- adds a built-in “CAJAL scientific paper agent” template under the
existing agent template catalog
- preconfigures the agent for grounded scientific writing: retrieval
first, citation traceability, LaTeX-ready output, and explicit
limitations when evidence is missing
- adds unit coverage to ensure the template normalizes through RAGFlow’s
production template loader, keeps graph form data in sync, and exposes
the Ollama model option
Behavior/evidence gathered for the requested model:
- Hugging Face model metadata for `Agnuxo/CAJAL-4B-P2PCLAW` reports
`pipeline_tag=text-generation` and tags including `gguf`, `llama.cpp`,
`vllm`, `scientific-research`, `papers`, `academic-writing`, `latex`,
and `license:apache-2.0`.
- The model card documents CAJAL as a 4B scientific paper generation
model with 32K context, local inference, LaTeX/citation specialization,
and CPU-only support around 5 tok/s on Ryzen 7 5800X.
- Local CPU generation could not be completed on this machine because
the advertised Ollama model name is not currently resolvable from
Ollama’s registry: both
`https://registry.ollama.ai/v2/agnuxo/cajal-4b-p2pclaw/manifests/latest`
and
`https://registry.ollama.ai/v2/library/agnuxo/cajal-4b-p2pclaw/manifests/latest`
returned `404 Not Found`; the Hugging Face repo tree currently exposes
an 8.4 GB `model.safetensors` but no GGUF artifact in `main`. The
template therefore targets the documented Ollama model name for users
who have the local CAJAL deployment/model file available.
Verification run locally:
```bash
python3 -m pytest test/test_cajal_template_unit.py -q
# 3 passed in 0.34s
python3 - <<'PY'
import json, glob
for f in sorted(glob.glob('agent/templates/*.json') + ['conf/llm_factories.json']):
with open(f, encoding='utf-8') as fp: json.load(fp)
print('json_ok')
PY
# json_ok
python3 -m ruff check test/test_cajal_template_unit.py
# All checks passed!
git diff --check
```
`uv run pytest
test/testcases/test_web_api/test_agent_app/test_cajal_template_unit.py
-q` was also attempted first, but dependency setup failed before test
collection while building `ormsgpack==1.5.0` from uv with a package
metadata parse error. Clearing uv’s `ormsgpack` cache and retrying
reproduced the same build failure, so the focused unit test was run with
the system Python environment instead.
### 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: sxxtony <sxxtony@users.noreply.github.com>
Co-authored-by: yzc <yzc@users.noreply.github.com>
Co-authored-by: Zhichang Yu <yuzhichang@gmail.com>
…rialization
Agent components (llm.py, agent_with_tools.py, message.py) store
functools.partial objects as deferred streaming handles in their output
slots. When the canvas state gets serialized for SSE events, Redis
commits, or logging, these partials — plus non-copyable objects like
Langfuse clients — crash json.dumps and deepcopy.
Changes:
- canvas_app.py: add default=str to json.dumps for SSE event
serialization (lines 238, 296)
- canvas.py: wrap deepcopy calls in try/except to handle non-copyable
objects (Langfuse clients, etc.), add default=str to final json.dumps
- base.py: add default=str to ComponentParamBase.__str__ to handle
non-serializable objects in component parameters
Closes#14229
### 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)
- [ ] New Feature (non-breaking change which adds functionality)
- [ ] Documentation Update
- [ ] Refactoring
- [ ] Performance Improvement
- [ ] Other (please describe):
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: yzc <yuzhichang@gmail.com>
## Summary
- Treat `max_tokens=0` as unset (`or 8192`) when building model context
budgets, fixing agents that silently zeroed prompts when a vLLM model
had `max_tokens: 0` in tenant config
- Replace trailing same-role canvas history in `LLM._sys_prompt_and_msg`
instead of skipping the current user prompt
- Add `LLM.fit_messages()` validation after `message_fit_in` on agent
paths so empty user content fails fast with a clear error instead of
reaching vLLM
Fixes#16411
## Root cause
Agent canvas workflow called `message_fit_in` with `int(max_length *
0.97)`. When `max_length` was `0`, both system and user content were
trimmed to empty strings. The `[HISTORY STREAMLY]` log showing only
`{"role":"user","content":""}` matches this. A secondary bug skipped
appending the formatted user prompt when history ended with a `user`
role message.
## Test plan
- [x] Added `test/unit_test/agent/component/test_llm_prompt.py` for
role-replace, validation, and zero-budget fitting
- [x] Added
`test_message_fit_in_zero_budget_preserves_non_empty_messages` in
`test_generator_message_fit_in.py`
- [ ] CI unit tests
- [ ] Manual: agent canvas `begin → Retrieval → Agent → Message` with
vLLM Qwen3; confirm user message reaches LLM
Made with [Cursor](https://cursor.com)
---------
Co-authored-by: Taranum Wasu <taranumwasu@Taranums-MacBook-Pro.local>
Co-authored-by: Cursor <cursoragent@cursor.com>
Recovery PR for #16173 after the fork branch was accidentally reset
during rewrite-cleanup.
Cherry-picked onto current `main`:
- fix(common/time_utils): correct fallback timestamp and ISO-8601
normalization
- fix(common/time_utils): preserve zero timestamps and mark regression
tests
- test(common/time_utils): make fallback assertions deterministic
Supersedes closed#16173 — same branch
`Harsh23Kashyap/fix/time-utils-edgecases`, rebuilt per @yuzhichang
recovery steps in
https://github.com/infiniflow/ragflow/pull/16173#issuecomment-4829663835
---------
Co-authored-by: Harsh Kashyap <harshkashyap@Harshs-MacBook-Pro.local>
Co-authored-by: Cursor <cursoragent@cursor.com>
## Summary
Two changes to make Go build \& run independent of native libraries
(office_oxide, pdfium, pdf_oxide).
## 1. Make native libraries optional (build.sh + Go source)
## 2. Roll back tests.yml CI changes from PR #16391
This PR migrates the Box web OAuth flow from Python to Go for:
- POST /api/v1/connectors/box/oauth/web/start
- GET /api/v1/connectors/box/oauth/web/callback
- POST /api/v1/connectors/box/oauth/web/result
