yangdx
541653ef6d
Merge branch 'main' into rerank
2025-07-14 18:11:11 +08:00
yangdx
9afe578fe7
Update webui assets
2025-07-14 17:56:51 +08:00
zrguo
c2da2fbe12
build
2025-07-14 17:19:28 +08:00
Daniel.y
1a760870df
Merge pull request #1765 from bun781/main
...
Add Cloudflare Workers AI support for serverless LLM/embedding
2025-07-14 17:15:58 +08:00
zrguo
2d7f87e936
Update settings.ts
2025-07-14 16:39:42 +08:00
zrguo
bbd91d3a18
Update operate.py
2025-07-14 16:37:25 +08:00
zrguo
4e425b1b59
Revert "update from main"
...
This reverts commit 1d0376d6a9
2025-07-14 16:29:00 +08:00
zrguo
1d0376d6a9
update from main
2025-07-14 16:27:49 +08:00
zrguo
c9cbd2d3e0
Merge branch 'main' into rerank
2025-07-14 16:24:29 +08:00
zrguo
ef2115d437
Update token limit
2025-07-14 15:53:48 +08:00
Daniel.y
e943591e68
Merge pull request #1779 from danielaskdd/fix-postgres-timezone
...
Fix: Resolve timezone handling problem in PostgreSQL storage
2025-07-14 04:20:38 +08:00
yangdx
7e988158a9
Fix: Resolve timezone handling problem in PostgreSQL storage
...
- Changed timestamp columns to naive UTC
- Added datetime formatting utilities
- Updated SQL templates for timestamp extraction
- Simplified timestamp migration logic
2025-07-14 04:12:52 +08:00
al1kss
e44e7296b2
fix: apply automatic ruff formatting
2025-07-14 00:28:45 +06:00
Daniel.y
375bfd57a4
Merge pull request #1778 from danielaskdd/add-ollama-num-ctx
...
feat: Refine summary logic and add dedicated Ollama num_ctx config
2025-07-14 02:13:08 +08:00
yangdx
b03bb48e24
feat: Refine summary logic and add dedicated Ollama num_ctx config
...
- Refactor the trigger condition for LLM-based summarization of entities and relations. Instead of relying on character length, the summary is now triggered when the number of merged description fragments exceeds a configured threshold. This provides a more robust and logical condition for consolidation.
- Introduce the `OLLAMA_NUM_CTX` environment variable to explicitly configure the context window size (`num_ctx`) for Ollama models. This decouples the model's context length from the `MAX_TOKENS` parameter, which is now specifically used to limit input for summary generation, making the configuration clearer and more flexible.
- Updated `README` files, `env.example`, and default values to reflect these changes.
2025-07-14 01:55:04 +08:00
yangdx
e8b3dfcf90
Bump api verion to 0182
2025-07-14 00:29:48 +08:00
Daniel.y
47588fa9c9
Merge pull request #1777 from danielaskdd/fix-postgres-field-len
...
Increase field lengths for entity and file paths for PostgreSQL
2025-07-14 00:27:44 +08:00
yangdx
157fb4c871
Increase field lengths for entity and file paths for PostgreSQL
...
- Expand entity_name length to 512 chars
- Increase source/target ID lengths
- Convert file_path to TEXT type
- Add migration logic
2025-07-14 00:24:54 +08:00
Daniel.y
568a809957
Merge pull request #1776 from danielaskdd/fix-redisk-conn-pool
...
Increase max length limits for Milvus storage fields
2025-07-13 23:26:08 +08:00
yangdx
187a623125
Increase max length limits for Milvus storage fields
...
- Extended entity_name max_length to 512
- Increased entity_type max_length to 128
- Expanded file_path limits to 1024
- Raised src_id/tgt_id limits to 512
2025-07-13 23:13:45 +08:00
Daniel.y
77b8fb9b77
Merge pull request #1775 from danielaskdd/fix-redisk-conn-pool
...
Hotfix: Resolves connection pool bugs for Redis
2025-07-13 22:57:18 +08:00
yangdx
6730a89d7c
Hotfix: Resolves connection pool bugs for Redis
...
- The previous implementation of the shared Redis connection pool had a critical issue where any Redis storage instance would disconnect the global shared pool upon closing. This caused `ConnectionError` exceptions for other instances still using the pool.
- This commit resolves the issue by introducing a reference counting mechanism in `RedisConnectionManager`.
2025-07-13 22:54:34 +08:00
yangdx
f185b3fb38
Optimize async task limits for graph processing
...
- Increased concurrency for graph operations
- Renamed variables for clarity
- Updated status messages
2025-07-13 21:51:19 +08:00
yangdx
ab805b35c4
Update doc: concurrent explain
2025-07-13 21:50:30 +08:00
Daniel.y
7945d7de59
Merge pull request #1774 from danielaskdd/fix-keyed-lock-error
...
Hotfix: prevent premature lock cleanup in multiprocess mode
2025-07-13 14:13:32 +08:00
yangdx
85cd1178a1
fix: prevent premature lock cleanup in multiprocess mode
...
- Change cleanup condition from count == 1 to count == 0 to properly
remove reused locks from cleanup list
- Fix RuntimeError: Attempting to release lock for xxxx more times than it was acquired
2025-07-13 13:51:48 +08:00
yangdx
03b40937f7
Reduce embedding concurrency limit from 16 to 8
2025-07-13 03:13:52 +08:00
yangdx
a2eeae9661
Fixes incorrect cleanup count
2025-07-13 02:38:36 +08:00
yangdx
582e952020
Disable direct logging by default for shared storage module
2025-07-13 01:58:50 +08:00
yangdx
cbf544b3c1
Remvoe redundant log message
2025-07-13 01:51:30 +08:00
yangdx
e4aef36977
Update webui assets
2025-07-13 01:36:25 +08:00
yangdx
fc7b0a9273
Improve query settings input experience in WebUI
2025-07-13 01:35:21 +08:00
yangdx
465757aa6a
Increase status card label width and dialog size
2025-07-13 01:14:12 +08:00
yangdx
efc359c411
Update webui assets
2025-07-13 00:57:41 +08:00
Daniel.y
0e52582bf2
Merge pull request #1773 from danielaskdd/cleanup-lock-in-healthcheck
...
Feat: Added reranker config and lock status to status card of WebUI
2025-07-13 00:52:57 +08:00
Daniel.y
9674ade611
Merge pull request #1772 from danielaskdd/rebuild-chunks-in-parallel
...
Optimize knowledge graph rebuild with parallel processing
2025-07-13 00:49:52 +08:00
yangdx
eb31ff0f90
Update i18n translation
2025-07-13 00:46:12 +08:00
yangdx
ab561196ff
Feat: Added reranker config and lock status to status card of WebUI
2025-07-13 00:41:54 +08:00
yangdx
0e3aaa318f
Feat: Add keyed lock cleanup and status monitoring
2025-07-13 00:09:00 +08:00
Daniel.y
9f3af332ec
Merge pull request #1771 from danielaskdd/namespace-keyed-lock
...
Refac: Generalize keyed lock with namespace support
2025-07-12 13:29:38 +08:00
yangdx
e4bf4d19a0
Optimize knowledge graph rebuild with parallel processing
...
- Add parallel processing for KG rebuild
- Implement keyed locks for data consistency
2025-07-12 13:22:56 +08:00
yangdx
a85d7054d4
fix: move node existence check inside lock to prevent race condition
...
Move knowledge_graph_inst.has_node check inside get_storage_keyed_lock
in _merge_edges_then_upsert to ensure atomic check-then-act operations
and prevent duplicate node creation during concurrent updates.
2025-07-12 12:22:32 +08:00
yangdx
2ade3067f8
Refac: Generalize keyed lock with namespace support
...
Refactored the `KeyedUnifiedLock` to be generic and support dynamic namespaces. This decouples the locking mechanism from a specific "GraphDB" implementation, allowing it to be reused across different components and workspaces safely.
Key changes:
- `KeyedUnifiedLock` now takes a `namespace` parameter on lock acquisition.
- Renamed `_graph_db_lock_keyed` to a more generic _storage_keyed_lock`
- Replaced `get_graph_db_lock_keyed` with get_storage_keyed_lock` to support namespaces
2025-07-12 12:10:12 +08:00
yangdx
f2d875f8ab
Update comments
2025-07-12 11:05:25 +08:00
yangdx
943ead8b1d
Bump api version to 0181
2025-07-12 05:59:13 +08:00
Daniel.y
b0ca25e5f1
Merge pull request #1768 from schmidt-marvin/main
...
fix(build): pyproject.toml setup
2025-07-12 05:44:48 +08:00
Daniel.y
ad7d7d0854
Merge pull request #1770 from danielaskdd/merge_lock_with_key
...
Refac: Optimize keyed lock cleanup logic with time and size tracking
2025-07-12 05:24:36 +08:00
yangdx
5ee509e671
Fix linting
2025-07-12 05:17:44 +08:00
yangdx
964293f21b
Optimize lock cleanup with time tracking and intervals
...
- Add cleanup time tracking variables
- Implement minimum cleanup intervals
- Track earliest cleanup times
- Handle time rollback cases
- Improve cleanup logging
2025-07-12 04:34:26 +08:00
yangdx
39965d7ded
Move merging stage back controled by max parallel insert semhore
2025-07-12 03:32:08 +08:00
