### What problem does this PR solve?
- ping
- token
- log level
### Type of change
- [x] Refactoring
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Refactor**
* System endpoints consolidated under /api/v1/system: ping, health
check, and token management moved to the centralized API surface.
* Token management unified at /api/v1/system/tokens with
list/create/delete behavior.
* **Documentation**
* API reference updated to reflect the new /api/v1/system paths.
* **Tests**
* Client fixtures and test utilities updated to use
/api/v1/system/tokens; one unit test for health/oceanbase status
removed.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Signed-off-by: Jin Hai <haijin.chn@gmail.com>
### What problem does this PR solve?
As title.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Bug Fixes**
* Standardized the query parameter used when listing documents so
listings behave consistently across the web and client interfaces.
* Clarified the error message shown when a required dataset ID is
missing to give clearer guidance to users.
* **Tests**
* Updated test coverage to reflect the standardized dataset identifier
usage.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Signed-off-by: Jin Hai <haijin.chn@gmail.com>
…
### What problem does this PR solve?
Closes#13857
Parent-child chunking was introduced in v0.23.0 but is only configurable
through the web UI. Users managing datasets programmatically cannot
enable it via the HTTP API or Python SDK because `ParserConfig` uses
`extra="forbid"`, rejecting the `children_delimiter` field at
validation.
### What does this PR change?
Adds a `parent_child` nested config to `ParserConfig`, following the
same pattern as `raptor` and `graphrag`:
```json
"parser_config": {
"parent_child": {
"use_parent_child": true,
"children_delimiter": "\n"
}
}
```
- api/utils/validation_utils.py — new ParentChildConfig model, added to
ParserConfig
- api/utils/api_utils.py — naive defaults + flatten to
children_delimiter for the execution layer
- api/apps/services/dataset_api_service.py — flatten on the update path
- test/testcases/configs.py — updated DEFAULT_PARSER_CONFIG
-
test/testcases/test_http_api/test_dataset_management/test_create_dataset.py
— 4 valid + 2 invalid test cases
No changes to the execution layer (rag/app/naive.py, rag/nlp/search.py).
Existing UI flow via ext is unaffected.
### 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):
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Added parent-child chunking configuration for dataset creation and
updates with new `use_parent_child` toggle and customizable
`children_delimiter` setting to specify how parent chunks are split into
child chunks.
* **Documentation**
* Updated HTTP and Python API references with parent-child chunking
configuration details and examples.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
### What problem does this PR solve?
Implements automatic adjustment of knowledge base chunk recall weights
based on user feedback (upvotes/downvotes). When users upvote or
downvote a response, the system locates the corresponding knowledge
snippets and adjusts their recall weight to improve future retrieval
quality.
**Closes #12670**
**How it works:**
1. User upvotes/downvotes a response via `POST /thumbup`
2. System extracts chunk IDs from the conversation reference
3. For each referenced chunk:
- Reads current `pagerank_fea` value from document store
- Increments (+1) for upvote or decrements (-1) for downvote
- Clamps weight to [0, 100] range
- Updates chunk in ES/Infinity/OceanBase
4. Future retrievals score these chunks higher/lower based on
accumulated feedback
**Files changed:**
- `api/db/services/chunk_feedback_service.py` - New service for updating
chunk pagerank weights
- `api/apps/conversation_app.py` - Integrated feedback service into
thumbup endpoint
- `test/testcases/test_web_api/test_chunk_feedback/` - Unit tests
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Chat message feedback now updates per-chunk relevance weights
(feature-flag gated), with configurable weighting and atomic updates
across storage backends.
* **Bug Fixes**
* Stricter validation for message feedback inputs and more robust
handling of feedback transitions.
* **Tests**
* Expanded test coverage for chunk-feedback behavior, weighting
strategies, storage backends, and thumb-flip scenarios.
* **Chores**
* CI workflow extended to run the new chunk-feedback web API tests.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Co-authored-by: mkdev11 <YOUR_GITHUB_ID+MkDev11@users.noreply.github.com>
Co-authored-by: mkdev11 <MkDev11@users.noreply.github.com>
### What problem does this PR solve?
Refactor version API to RESTful style. Python and go server API also
updated.
### Type of change
- [x] Refactoring
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
## Release Notes
* **Refactor**
* Migrated core API endpoints to the `/api/v1/` namespace for improved
consistency and organization.
* Standardized system version, search, and chat list endpoints under the
new API versioning structure.
* **New Features**
* Added MinIO region configuration support, allowing specification of
storage engine regional settings via environment variables or
configuration files.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Signed-off-by: Jin Hai <haijin.chn@gmail.com>
### What problem does this PR solve?
Add validation logic for parser_config.
Refactor the processing flow. Before change, validation logics and
update logics are mixed up - some validation logis executes followed by
some update logic executes and then another such
"validation-and-then-update" which is not good. After change, all
validation logic executes firstly. Update logic will be executed after
ALL validation logic executed.
Validation logic for parameters (that come from front end) will be
checked using Pydantic. For validation logic that depends on data from
DB, they will be in separate methods.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
- [x] Refactoring
### What problem does this PR solve?
Implement UpdateDataset and UpdateMetadata in GO
Add cli:
UPDATE CHUNK <chunk_id> OF DATASET <dataset_name> SET <update_fields>
REMOVE TAGS 'tag1', 'tag2' from DATASET 'dataset_name';
SET METADATA OF DOCUMENT <doc_id> TO <meta>
### Type of change
- [ ] Refactoring
### What problem does this PR solve?
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
- [x] Refactoring
---------
Co-authored-by: Zhichang Yu <yuzhichang@gmail.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
### What problem does this PR solve?
This PR fixes a race in batch document parsing where overlapping parse
requests for the same document could clear/rewrite chunk state and make
previously parsed content appear lost. It adds an atomic per-document
parse guard so only one parse can run at a time for that document (Fixes
#13864 ).
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Previously, `apikey_required` called
`request.headers.get('Authorization').split()[1]` without checking for
None or insufficient parts, causing an unhandled AttributeError or
IndexError (500) instead of a proper 403 JSON response.
This applies the same guarding pattern already used by `token_required`
in the same file.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
- [x] Refactoring
Add REST APIs to dynamically query and modify log levels at runtime for
both Python (Flask) and Go servers.
Changes:
- common/log_utils.py: add set_log_level() and get_log_levels()
functions
- admin/server/routes.py: add GET/PUT /api/v1/admin/log_levels endpoints
- api/apps/system_app.py: add GET/PUT /api/{version}/system/log_levels
endpoints
- internal/logger/logger.go: add GetLevel() and SetLevel() with atomic
level support
- internal/handler/system.go: add GetLogLevel, SetLogLevel, Health
handlers
- internal/router/router.go: route /health to systemHandler
- internal/admin/handler.go: add GetLogLevel, SetLogLevel handlers
- internal/admin/router.go: add /api/v1/admin/log_level routes
### 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] New Feature (non-breaking change which adds functionality)
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
### What problem does this PR solve?
Enable reading Tag Set tags via API (expose tag_kwd field). The result
of the queried list chunks is as shown below:
<img width="1422" height="818" alt="image"
src="https://github.com/user-attachments/assets/abd1960a-fe34-489e-9d72-525f8e574938"
/>
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
Co-authored-by: heyang.why <heyang.why@alibaba-inc.com>
### What problem does this PR solve?
fixes issue #13799 where team members get model not authorized when
running RAG on an admin-shared knowledge base after the admin changes
the KB embedding model (for example to bge-m3).
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Allow create datasets with parse_type == 1/None and chunk_method, or
parse_type == 2 and pipeline_id.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Allow create dataset with resume chunk_method.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
Problem
The /file2document/convert endpoint ran all file lookups, document
deletions, and insertions synchronously inside the
request cycle. Linking a large folder (~1.7GB with many files) caused
504 Gateway Timeout because the blocking DB loop
held the HTTP connection open for too long.
Fix
- Extracted the heavy DB work into a plain sync function _convert_files
- Inputs are validated and folder file IDs expanded upfront (fast path)
- The blocking work is dispatched to a thread pool via
get_running_loop().run_in_executor() and the endpoint returns 200
immediately
- Frontend only checks data.code === 0 so the response change
(file2documents list → True) has no impact
Fixes#13781
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
### What problem does this PR solve?
Searches /search API to RESTFul
### Type of change
- [x] Documentation Update
- [x] Refactoring
Co-authored-by: Jin Hai <haijin.chn@gmail.com>
Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
### What problem does this PR solve?
CI isn't stable, try to fix it.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
---------
Signed-off-by: Jin Hai <haijin.chn@gmail.com>
### What problem does this PR solve?
Fix
migrate_add_unique_email-silently-skips-unique-constraint-when-non-unique-user_email-index-exists.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Files /file API to RESTFul style.
### Type of change
- [x] Documentation Update
- [x] Refactoring
---------
Co-authored-by: writinwaters <cai.keith@gmail.com>
Co-authored-by: Liu An <asiro@qq.com>
### What problem does this PR solve?
Saving dataset settings failed with validation error 101 (Extra inputs
are not permitted)
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
1. Split dataset api to gateway and service, and modify web UI to use
restful http api.
2. Old KB releated APIs are commented.
### Type of change
- [x] Refactoring
---------
Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
### What problem does this PR solve?
Fixes [#13505](https://github.com/infiniflow/ragflow/issues/13505): Jira
incremental sync could miss updated issues after initial sync,
especially near time boundaries.
Root cause:
- Jira JQL uses minute-level precision for `updated` filters.
- Incremental windows had no overlap buffer, so boundary updates could
be skipped.
- Sync log cursor tracking used a backward-facing update for
`poll_range_start`.
- Existing-doc updates in `upload_document` lacked a KB ownership guard
for doc-id collisions.
What changed:
- Added Jira incremental overlap buffer (`time_buffer_seconds`,
defaulting to `JIRA_SYNC_TIME_BUFFER_SECONDS`) when building JQL
lower-bound time.
- Preserved second-level post-filtering to avoid duplicate reprocessing
while still catching boundary updates.
- Improved Jira sync logging to include start/end window and overlap
configuration.
- Updated sync cursor tracking in `increase_docs` to keep
`poll_range_start` moving forward with max update time.
- Added KB ID safety check before updating existing document records in
`upload_document`.
Verification performed:
- Python syntax compile checks passed for modified files.
- Manual verification flow:
1. Run full Jira sync.
2. Edit an already-indexed Jira issue.
3. Run next incremental sync.
4. Confirm updated content is re-ingested into KB.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
---------
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
Closes#1398
### What problem does this PR solve?
Adds native support for EPUB files. EPUB content is extracted in spine
(reading) order and parsed using the existing HTML parser. No new
dependencies required.
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
To check this parser manually:
```python
uv run --python 3.12 python -c "
from deepdoc.parser import EpubParser
with open('$HOME/some_epub_book.epub', 'rb') as f:
data = f.read()
sections = EpubParser()(None, binary=data, chunk_token_num=512)
print(f'Got {len(sections)} sections')
for i, s in enumerate(sections[:5]):
print(f'\n--- Section {i} ---')
print(s[:200])
"
```
### What problem does this PR solve?
using builtin model when parsing gave an error because it expects
fid==builtin. split_model_name_and_factory returns id=None. pr allows
the model to be accepted wheter with or without @Builtin
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Feat: Export Agent Logs.
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
---------
Co-authored-by: balibabu <assassin_cike@163.com>
### What problem does this PR solve?
Follow-up expose agent structured outputs in non-stream completions
#13389.
### Type of change
- [x] Documentation Update
- [x] Refactoring
---------
Co-authored-by: writinwaters <cai.keith@gmail.com>
### What problem does this PR solve?
1. Split dataset api to gateway and service, and modify web UI to use
restful http api.
2. Old KB releated APIs are commented.
### Type of change
- [x] Refactoring
### What problem does this PR solve?
Add_chunk supports add image.
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
### What problem does this PR solve?
1. Split dataset api to gateway and service, and modify web UI to use
restful http api.
2. Old KB releated APIs are commented.
### Type of change
- [x] Refactoring
### What problem does this PR solve?
Fix: model selecton rule in get_model_config_by_type_and_name
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Feat: Modify the style of the release confirmation box.
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
---------
Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
Co-authored-by: balibabu <assassin_cike@163.com>
Co-authored-by: 6ba3i <isbaaoui09@gmail.com>
### What problem does this PR solve?
Get user_id from canvas and record it.
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
## Summary
Fixes#13544: PostgreSQL startup crash because
`update_tenant_llm_to_id_primary_key()` unconditionally uses
MySQL-specific SQL.
- Split `update_tenant_llm_to_id_primary_key()` into
`_update_tenant_llm_to_id_primary_key_mysql()` and
`_update_tenant_llm_to_id_primary_key_postgres()`, dispatching on
`settings.DATABASE_TYPE`
- MySQL path: unchanged (existing `DATABASE()`, `SET @row = 0`,
`AUTO_INCREMENT`, `DROP PRIMARY KEY` logic)
- PostgreSQL path: uses `current_database()`, `ROW_NUMBER() OVER (ORDER
BY ...)` for sequential IDs, `CREATE SEQUENCE` + `nextval()` for
auto-increment, and `information_schema.table_constraints` to find the
PK constraint name
- Also fix `migrate_add_unique_email()`: MySQL-only
`information_schema.statistics` is replaced with `pg_indexes` on
PostgreSQL
## Test plan
- [ ] Start RAGFlow with `DB_TYPE=postgres` — startup should complete
without `function database() does not exist` error
- [ ] Start RAGFlow with `DB_TYPE=mysql` (default) — existing behaviour
unchanged, migration runs as before
- [ ] Fresh PostgreSQL install: verify `tenant_llm.id` column is created
as a serial primary key after migration
- [ ] Idempotency: running migration twice on PostgreSQL should be a
no-op (column already exists check passes)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: gambletan <gambletan@github>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
### What problem does this PR solve?
Fixes#13285
When an LLM returns a transient error (e.g. overloaded) during parsing,
the task progress is set to -1. Previously, the progress could never be
updated again, leaving the document permanently stuck in FAIL status
even after the task successfully recovered and completed.
Three coordinated changes address this:
1. task_service.update_progress: relax the progress update guard to
accept prog >= 1 even when current progress is -1, so a task that
recovers from a transient failure can report completion.
2. document_service.get_unfinished_docs: include documents that are
marked FAIL (progress == -1) but still have at least one non-failed task
(task.progress >= 0) in the polling set, so their status can be
re-synced once a task recovers. Documents where all tasks have
permanently failed are excluded to avoid unnecessary polling.
3. document_service.update_progress: explicitly set document status to
RUNNING when not all tasks have finished, instead of preserving whatever
stale status (potentially FAIL) the document previously had.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
## Summary
Fix knowledge-base chat retrieval when no individual document IDs are
selected.
## Root Cause
`async_chat()` initialized `doc_ids` as an empty list when the request
did not explicitly select documents. That empty list was then forwarded
into retrieval as an active `doc_id` filter, effectively becoming
`doc_id IN []` and suppressing all chunk matches.
## Changes
- treat missing selected document IDs as `None` instead of `[]`
- keep explicit document filtering when IDs are actually provided
- add regression coverage for the shared chat retrieval path
## Validation
- `python3 -m py_compile api/db/services/dialog_service.py
test/unit_test/api/db/services/test_dialog_service_use_sql_source_columns.py`
- `.venv/bin/python -m pytest
test/unit_test/api/db/services/test_dialog_service_use_sql_source_columns.py`
- manually verified that chat completions again inject retrieved
knowledge into the prompt
---------
Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
### What problem does this PR solve?
Add delete all support for delete operations.
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
- [x] Documentation Update
---------
Co-authored-by: writinwaters <cai.keith@gmail.com>
## Summary
- Convert bare `open()` calls to `with` context managers or
`Path.read_text()`
- File handles leak if not properly closed, especially on exceptions
- Fixes in crypt.py, sequence2txt_model.py, term_weight.py,
deepdoc/vision/__init__.py
## Test plan
- [x] File operations work correctly with context managers
- [x] Resources properly cleaned up on exceptions
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
### What problem does this PR solve?
Fix https://github.com/infiniflow/ragflow/issues/13388
Call get_flatted_meta_by_kbs in dify retrieval. Remove get_meta_by_kbs.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
## Summary
- scope normal document-list metadata lookups to the current page's
document IDs
- keep the `return_empty_metadata=True` path dataset-wide because it
needs full knowledge of docs that already have metadata
- add unit tests for both paged listing paths and the unchanged
empty-metadata behavior
## Why
`DocumentService.get_list()` and the normal `get_by_kb_id()` path were
calling `DocMetadataService.get_metadata_for_documents(None, kb_id)`,
which loads metadata for the entire dataset on every page request.
That becomes especially problematic on large datasets. The metadata scan
path paginates through the full metadata index without an explicit sort,
while the ES helper only switches to `search_after` beyond `10000`
results when a sort is present. In practice this can lead to unnecessary
full-dataset metadata work, slower document-list loading, and unreliable
`meta_fields` in list responses for large KBs.
This change keeps the existing empty-metadata filter behavior intact,
but scopes normal list responses to metadata for the current page only.
### What problem does this PR solve?
Previously, when an Agent component was configured with structured
output, the non-streaming /agents/{agent_id}/completions API never
returned the structured field in its response.
The root cause: the non-streaming code path only collected message
events to build full_content, then returned the workflow_finished
payload — which only contains the output of the last component in the
execution path (typically a Message component).
Any structured output set by upstream components (e.g., Agent or LLM)
was silently discarded.
This PR fixes the non-streaming handler to iterate node_finished events
and collect structured output from intermediate components.
If any component produced a non-empty structured value, it is included
in the final response under data.structured. The streaming path is
unaffected, as it already exposes node_finished events to the caller.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
