opensource
/
dify
mirror of https://github.com/langgenius/dify.git
2
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
5,638 Commits 288 Branches 160 Tags 483 MiB
Commit Graph

325 Commits

Author SHA1 Message Date
yihong 22258fb0bf
fix: filter bug for keywork cause code can not reach (#11666) 
Signed-off-by: yihong0618 <zouzou0208@gmail.com>
 2024-12-15 17:12:06 +08:00
yihong 36cb25b341
fix: support mdx files close #11557 (#11565) 
Signed-off-by: yihong0618 <zouzou0208@gmail.com>
 2024-12-12 13:37:56 +08:00
Jiang 0d04cdc323
Lindorm vdb (#11574) 
Co-authored-by: jiangzhijie <jiangzhijie.jzj@alibaba-inc.com>
 2024-12-12 09:43:27 +08:00
Jyong 9b7adcd4d9
update tidb batch get endpoint to basic mode (#11426) 2024-12-06 17:06:46 +08:00
Jyong d7c1f43b49
fix tidb full-text-search vector missed (#11337) 2024-12-04 16:13:23 +08:00
Jyong c58d2fce89
roll back rerank topn setting (#11297) 2024-12-03 17:34:56 +08:00
yihong e686f12317
fix: better handle error (#11265) 
Signed-off-by: yihong0618 <zouzou0208@gmail.com>
 2024-12-03 09:15:38 +08:00
-LAN- 9601102885
fix(word_extractor): Fix type error and remove stream in ssrf_proxy (#11241) 
Signed-off-by: -LAN- <laipz8200@outlook.com>
 2024-12-02 10:24:03 +08:00
Cling_o3 f9c2aa7689
feat: add retireval_top_n to config in env (#11132) 2024-11-30 11:14:45 +08:00
kazuya-awano 2d6865d421
Ensure consistent float type for cached embedding return values (#10185) 2024-11-29 09:18:41 +08:00
yihong d7160ee563
fix: typo in upstashVector if id is always true, also fix some type hint (#11183) 
Signed-off-by: yihong0618 <zouzou0208@gmail.com>
 2024-11-28 14:05:25 +08:00
-LAN- 9789905a1f
chore(*): Removes debugging print statements (#11145) 
Signed-off-by: -LAN- <laipz8200@outlook.com>
 2024-11-26 22:03:19 +08:00
Bowen Liang 6c8e208ef3
chore: bump minimum supported Python version to 3.11 (#10386) 2024-11-24 13:28:46 +08:00
yihong ed55de888a
fix: rules should not be None for in (#10977) 
Signed-off-by: yihong0618 <zouzou0208@gmail.com>
 2024-11-22 23:04:20 +08:00
AkisAya cb0c55daa7
fix weight rerank of knowledge retrieval (#10931) 2024-11-21 17:53:20 +08:00
yihong 58a9d9eb9a
fix: better WeightRerankRunner run logic use O(1) and delete unused code (#10849) 
Signed-off-by: yihong0618 <zouzou0208@gmail.com>
 2024-11-19 20:12:13 +08:00
Zane 14f3d44c37
refactor: improve handling of leading punctuation removal (#10761) 2024-11-18 21:32:33 +08:00
8bitpd 873e9720e9
feat: AnalyticDB vector store supports invocation via SQL. (#10802) 
Co-authored-by: 璟义 <yangshangpo.ysp@alibaba-inc.com>
 2024-11-18 19:29:54 +08:00
Bowen Liang 51db59622c
chore(lint): cleanup repeated cause exception in logging.exception replaced by helpful message (#10425) 2024-11-15 15:41:40 +08:00
Jyong 0b2d51d859
add the index field for elasticsearch (#10592) 2024-11-12 21:43:16 +08:00
-LAN- a1543b7da0
fix(extractor): temporary file (#10543) 2024-11-11 17:31:27 +08:00
Leo.Wang c9f785e00f
Feat/tools/gitlab (#10407) 2024-11-08 09:53:03 +08:00
Bowen Liang 574c4a264f
chore(lint): Use logging.exception instead of logging.error (#10415) 2024-11-07 21:13:02 +08:00
Jyong 1024fc623e
fix the ssrf of docx file extractor external images (#10237) 2024-11-04 15:22:07 +08:00
Jiang 0c9e79cd67
Add Lindorm as a VDB choice (#10202) 
Co-authored-by: jiangzhijie <jiangzhijie.jzj@alibaba-inc.com>
 2024-11-04 09:10:26 +08:00
Shili Cao b61baa87ec
fix: avoid unexpected error when create knowledge base with baidu vector database and wenxin embedding model (#10130) 2024-10-31 21:34:23 +08:00
Jyong dad041c49f
fix issue: query is none when doing retrieval (#10129) 2024-10-31 21:25:00 +08:00
omr 11ca1bec0b
fix: optimize unique document filtering with set (#10082) 2024-10-31 16:32:58 +08:00
zhuhao 7433095240
chore: use dify_config.TIDB_SPEND_LIMIT instead of constant value (#10038) 2024-10-30 15:43:07 +08:00
Jyong 9ebd453b87
add rerank check when doing mutil-retrieval (#9998) 2024-10-30 11:17:39 +08:00
powerfool 878d13ef42
Added OceanBase as an option for the vector store in Dify (#10010) 2024-10-29 21:10:18 +08:00
Jyong 5580bcf870
add tidb spend limit config (#9999) 2024-10-29 17:51:13 +08:00
roadgoat19 c8ef9223e5
feat: couchbase integration (#6165) 
Co-authored-by: crazywoola <427733928@qq.com>
Co-authored-by: Elliot Scribner <elliot.scribner@couchbase.com>
Co-authored-by: crazywoola <100913391+crazywoola@users.noreply.github.com>
Co-authored-by: Bowen Liang <bowenliang@apache.org>
 2024-10-29 15:00:23 +08:00
Jyong f47177ecb4
add top_k for es full text search (#9963) 2024-10-28 23:04:54 +08:00
virgosoy 17cacf258e
fix: wrong element object (#9868) 2024-10-25 22:32:41 +08:00
Jyong 18106a4fc6
add tidb on qdrant type (#9831) 
Co-authored-by: Zhaofeng Miao <522856232@qq.com>
 2024-10-25 13:57:03 +08:00
Zixuan Cheng 88dec6ef2b
Added description for .ppt, specify the reason for unstructured.io (#9452) 
Co-authored-by: crazywoola <427733928@qq.com>
 2024-10-24 22:13:06 +08:00
Jyong 5f11fe521d
remove unstructured pdf extract (#9794) 2024-10-24 18:13:05 +08:00
Jyong 3e9d271b52
nltk security issue and upgrade unstructured (#9558) 2024-10-23 16:23:55 +08:00
ice yao ceb2c4f3ef
chore: reuse existing test functions with upstash vdb (#9679) 2024-10-23 10:42:11 +08:00
Zven 8e7a752b2a
feat: add upstash as a new vector database provider (#9644) 2024-10-23 09:16:35 +08:00
-LAN- 5f12c17355
fix(core): use CreatedByRole enum for role consistency (#9607) 2024-10-22 13:03:50 +08:00
Bowen Liang 4d9160ca9f
refactor: use dify_config to replace legacy usage of flask app's config (#9089) 2024-10-22 11:01:32 +08:00
-LAN- e61752bd3a
feat/enhance the multi-modal support (#8818) 2024-10-21 10:43:49 +08:00
ice yao 2155bba5b0
fix: update mismatch vector type (#9462) 2024-10-18 08:21:41 +08:00
zhuhao b90ad587c2
refactor: move the embedding to the rag module and abstract the rerank runner for extension (#9423) 2024-10-17 19:12:42 +08:00
zhuhao 86594851cb
refactor: update the default values of top-k parameter in vdb to be consistent (#9367) 2024-10-16 16:00:21 +08:00
Jyong 50635e9c15
Fix/economical knowledge retrieval (#9396) 2024-10-16 15:13:45 +08:00
zhuhao cd7ab6231f
refactor: Add an enumeration type and use the factory pattern to obtain the corresponding class (#9356) 2024-10-15 12:51:13 +08:00
ice yao d15ba3939d
Add Volcengine VikingDB as new vector provider (#9287) 2024-10-13 21:26:05 +08:00
zhuhao d97d3ff5fc
chore: add abstract decorator and output log when query embedding fails (#9264) 2024-10-12 23:58:41 +08:00
Shili Cao 2ec6ffe478
feat:support baidu vector db (#9185) 2024-10-12 23:24:17 +08:00
Jyong 42b02b3a5f
Fix/agent external knowledge retrieval (#9241) 2024-10-11 19:21:03 +08:00
Jyong 80b62d50f5
Fix/add es num_candidates (#9225) 2024-10-11 16:04:23 +08:00
Jyong cabdb4ef17
fix retrieval resource positon missed (#9155) 
Co-authored-by: Bowen Liang <liangbowen@gf.com.cn>
 2024-10-11 10:32:42 +08:00
Bowen Liang 240b66d737
chore: avoid implicit optional in type annotations of method (#8727) 2024-10-09 14:36:43 +08:00
Aurelius Huang 4585cffce1
fix: Compatible with special characters in pg full-text search. (#8921) 
Co-authored-by: Aurelius Huang <cm.huang@aftership.com>
 2024-09-30 16:32:23 +08:00
Jyong 9d221a5e19
external knowledge api (#8913) 
Co-authored-by: Yi <yxiaoisme@gmail.com>
 2024-09-30 15:38:43 +08:00
Zhaofeng Miao 369e1e6f58
feat(website-crawl): add jina reader as additional alternative for website crawling (#8761) 2024-09-30 09:57:19 +08:00
Bowen Liang 74f58f29f9
chore: bump ruff to 0.6.8 for fixing violation in SIM910 (#8869) 2024-09-29 00:29:59 +08:00
ice yao 27e33fb15c
chore: fix wrong VectorType match case (#8857) 2024-09-28 10:54:04 +08:00
zhuhao 55e6123db9
feat: add min-connection and max-connection for pgvector (#8841) 2024-09-27 18:16:20 +08:00
8bitpd 4c1063e1c5
fix: AnalyticdbVector retrieval scores (#8803) 2024-09-27 12:05:21 +08:00
zhuhao 008e0efeb0
refactor: update delete method as an abstract method (#8794) 2024-09-26 16:36:21 +08:00
crazywoola bf64ff215b
fix: . is missing in file_extension (#8736) 2024-09-25 10:09:20 +08:00
omr 8fd297f8b4
fix: redundant check for available_document_count (#8491) 2024-09-22 13:39:41 +08:00
Jyong 7e611ffbf3
multi-retrival use dataset's top-k (#8416) 2024-09-14 21:48:44 +08:00
Bowen Liang a1104ab97e
chore: refurish python code by applying Pylint linter rules (#8322) 2024-09-13 22:42:08 +08:00
Bowen Liang 6613b8f2e0
chore: fix unnecessary string concatation in single line (#8311) 2024-09-13 14:24:49 +08:00
-LAN- 08c486452f
fix: score_threshold handling in vector search methods (#8356) 2024-09-13 14:24:35 +08:00
Jyong 49cee773c5
fixed score threshold is none (#8342) 2024-09-13 10:21:58 +08:00
Bowen Liang 40fb4d16ef
chore: refurbish Python code by applying refurb linter rules (#8296) 2024-09-12 15:50:49 +08:00
Bowen Liang c69f5b07ba
chore: apply ruff E501 line-too-long linter rule (#8275) 
Co-authored-by: -LAN- <laipz8200@outlook.com>
 2024-09-12 14:00:36 +08:00
takatost 56c90e212a
fix(workflow): missing content in the answer node stream output during iterations (#8292) 
Co-authored-by: -LAN- <laipz8200@outlook.com>
 2024-09-12 13:59:48 +08:00
Bowen Liang 0f14873255
chore: cleanup ruff flake8-simplify linter rules (#8286) 
Co-authored-by: -LAN- <laipz8200@outlook.com>
 2024-09-12 12:55:45 +08:00
Bowen Liang 292220c596
chore: apply pep8-naming rules for naming convention (#8261) 2024-09-11 16:40:52 +08:00
Jyong bb3002b173
revert page column (#8217) 2024-09-10 18:21:22 +08:00
Bowen Liang 2cf1187b32
chore(api/core): apply ruff reformatting (#7624) 2024-09-10 17:00:20 +08:00
Jyong af92f19291
filter excel empty sheet (#8194) 2024-09-10 14:55:08 +08:00
Nam Vu 2d7954c7da
Fix variable typo (#8084) 2024-09-08 13:14:11 +08:00
ChengZi 2060db8e11
fix: change milvus init args from (host, port) to (url, token) (#8019) 
Signed-off-by: ChengZi <chen.zhang@zilliz.com>
 2024-09-06 17:32:48 +08:00
Byeongjin Kang d489b8b3e0
feat: return page number of pdf documents upon retrieval (#7749) 2024-09-05 16:43:26 +08:00
Jyong 0e71f6db84
fix spliter length missed (#7987) 2024-09-04 21:47:12 +08:00
KVOJJJin 14af87527f
Feat:remove estimation of embedding cost (#7950) 
Co-authored-by: jyong <718720800@qq.com>
 2024-09-04 14:41:47 +08:00
Sumkor 571415d1a4
fix: split text keep separator (#7930) 2024-09-04 12:59:10 +08:00
Jyong d8b6c053a2
fix rerank model value is empty string (#7937) 2024-09-03 21:25:21 +08:00
Jyong 01581dd35f
improve the notion table extract (#7925) 2024-09-03 17:52:07 +08:00
Jyong 6f33351eb3
ignore linked images when image id is none (#7890) 2024-09-02 19:37:05 +08:00
Jyong 60001a62c4
fixed chunk_overlap is None (#7703) 2024-08-27 16:38:06 +08:00
Kenn 122ce41020
feat: rewrite Elasticsearch index and search code to achieve Elasticsearch vector and full-text search (#7641) 
Co-authored-by: haokai <haokai@shuwen.com>
Co-authored-by: crazywoola <100913391+crazywoola@users.noreply.github.com>
Co-authored-by: Bowen Liang <bowenliang@apache.org>
Co-authored-by: wellCh4n <wellCh4n@foxmail.com>
 2024-08-27 11:43:44 +08:00
Qin Liu 162faee4f2
fix: set score_threshold to zero if it is None for MyScale vectordb (#7640) 
Co-authored-by: crazywoola <100913391+crazywoola@users.noreply.github.com>
 2024-08-27 09:47:16 +08:00
Jyong 7ae728a9a3
fix nltk averaged_perceptron_tagger download and fix score limit is none (#7582) 
Co-authored-by: crazywoola <100913391+crazywoola@users.noreply.github.com>
 2024-08-26 15:14:05 +08:00
张皮皮 f29685f8a1
fix score_threshold is none, return all top K documents (#7581) 2024-08-23 16:59:34 +08:00
Byeongjin Kang 0223fc6fd5
feat: add pgvector full_text_search (#7396) 2024-08-20 11:01:13 +08:00
mochi ba79088ffc
Fix SQL parser Error in MyScale vdb. (#7255) 2024-08-14 16:41:18 +08:00
miendinh f104b930cf
feat: support elasticsearch vector database (#3558) 
Co-authored-by: miendinh <miendinh@users.noreply.github.com>
Co-authored-by: crazywoola <100913391+crazywoola@users.noreply.github.com>
Co-authored-by: crazywoola <427733928@qq.com>
 2024-08-13 17:36:20 +08:00
Bowen Liang ccb6ddd840
chore: bump Ruff to 0.5.7 (#7174) 2024-08-12 10:24:48 +08:00
Joe 425174e82f
feat: update ops trace (#7102) 2024-08-09 15:22:16 +08:00
Jyong 12095f8cd6
extract docx filter comment element (#7092) 2024-08-08 16:53:29 +08:00
Jyong 169cde6c3c
add nltk punkt resource (#7063) 2024-08-08 14:23:22 +08:00
8bitpd 40c6f3c724
fix: add redis lock to AnalyticdbVector init (#6859) 
Co-authored-by: xiaozeyu <xiaozeyu.xzy@alibaba-inc.com>
 2024-08-07 17:32:06 +08:00
-LAN- df8f8c9a2c
feat(api/core/rag/datasource/vdb/analyticdb/analyticdb_vector.py): Checking config before init analyticdb (#7050) 2024-08-07 17:31:36 +08:00
Jyong 80c94f02e9
add vector field for other vectordb (#7051) 2024-08-07 17:14:03 +08:00
chenxu9741 72c75b75cf
feat: Add hyperlink parsing to the DOCX document. (#7017) 2024-08-07 16:01:14 +08:00
Bryan eece50acec
fix: tran list issue (#7009) 
Co-authored-by: libing <libing@healink.cn>
 2024-08-06 21:01:38 +08:00
Jyong 28d4e5b045
Fix/reranking mode is null (#7012) 2024-08-06 19:12:04 +08:00
yanghx c53875ce8c
fix #6902 .docx handles images within tables and handles cross-column tables (#6951) 2024-08-06 17:14:24 +08:00
Joe f97a51ce24
fix: reranking disable timer error (#6910) 2024-08-02 16:34:50 +08:00
Bowen Liang 6aa02f8c63
dep: bump pgvecto-rs client from 0.1.x to 0.2.x (#6891) 2024-08-02 15:51:23 +08:00
Jyong 44801df8f8
fix score threshold limit be None (#6900) 2024-08-02 12:10:51 +08:00
Jyong 8157fccf6d
delete weight_type (#6865) 2024-08-01 13:02:33 +08:00
-LAN- a98284b1ef
refactor(api): Switch to `dify_config` (#6750) 
Signed-off-by: -LAN- <laipz8200@outlook.com>
 2024-07-30 11:15:26 +08:00
Bowen Liang b414ea41d6
dep: initial support for Milvus 2.4.x (#6084) 2024-07-29 19:56:45 +08:00
Jyong cf258b7a67
add xlsx support hyperlink extract (#6722) 2024-07-26 19:26:52 +08:00
Yeuoly 79cb23e8ac
security/SSRF vulns (#6682) 2024-07-25 20:50:26 +08:00
灰灰 5e4ac11df3
fix: code block segmentation problem of markdown document (#6465) 2024-07-25 17:24:37 +08:00
tmuife 16b4f560cd
fix bugs(when using Oracle23ai as Vector DB) (#6658) 2024-07-25 17:07:14 +08:00
Jyong e4bb943fe5
Feat/delete single dataset retrival (#6570) 2024-07-24 12:50:11 +08:00
崔亮 0569c547ee
fix the issue of MILVUS_DATABASE has no effect. (#6424) 2024-07-23 15:03:55 +08:00
tmuife 06fc1bce9e
Add search by full text when using Oracle23ai as vector DB (#6559) 2024-07-23 15:03:21 +08:00
Sangmin Ahn 093b8ca475
fix: escape double quotation marks in the vector DB search query (#6506) 2024-07-23 15:02:25 +08:00
-LAN- cd7fa8027a
fix(api/core/model_manager.py): Avoid mutation during iteration. (#6536) 2024-07-22 22:58:22 +08:00
-LAN- 5e6fc58db3
Feat/environment variables in workflow (#6515) 
Co-authored-by: JzoNg <jzongcode@gmail.com>
 2024-07-22 15:29:39 +08:00
Jyong f38034e455
clean vector collection redis cache (#6494) 2024-07-21 15:09:09 +08:00
Waffle 2ba05b041f
refactor(myscale):Set the default value of the myscale vector db in DifyConfig. (#6441) 2024-07-19 10:57:45 +08:00
leoterry dc847ba145
Fix the vector retrieval sorting issue (#6431) 
Co-authored-by: weifj <“weifj@tuyuansu.com.cn”>
 2024-07-18 19:25:41 +08:00
Poorandy c8f5dfcf17
refactor(rag): switch to dify_config. (#6410) 
Co-authored-by: -LAN- <laipz8200@outlook.com>
 2024-07-18 18:40:36 +08:00
Jyong 0de224b153
fix wrong using of RetrievalMethod Enum (#6345) 2024-07-16 19:09:04 +08:00
tangyoha 0cbbaf3f68
fix: markdown proc will remove image (#5855) 2024-07-12 20:07:22 +08:00
Matri a9ee52f2d7
Fix/firecrawl parameters issue (#6213) 2024-07-12 12:59:50 +08:00
Waffle 7b225a5ab0
refactor(services/tasks): Swtich to dify_config witch Pydantic (#6203) 2024-07-12 12:25:38 +08:00
Zhuo Qiu 63e34e5227
feat: support MyScale vector database (#6092) 2024-07-11 15:21:59 +08:00
8bitpd 7c70eb87bc
feat: support AnalyticDB vector store (#5586) 
Co-authored-by: xiaozeyu <xiaozeyu.xzy@alibaba-inc.com>
 2024-07-09 13:32:04 +08:00
Aurelius Huang f546db5437
fix: document truncation and loss in notion document sync (#5631) 
Co-authored-by: Aurelius Huang <cm.huang@aftership.com>
 2024-07-05 11:48:17 +08:00
Joe e8b8f6c6dd
Feat/fix ops trace (#5672) 
Co-authored-by: takatost <takatost@gmail.com>
 2024-06-28 00:24:37 +08:00
Bowen Liang dcb72e0067
chore: apply flake8-comprehensions Ruff rules to improve collection comprehensions (#5652) 
Co-authored-by: -LAN- <laipz8200@outlook.com>
 2024-06-27 11:21:31 +08:00
Joe 4e2de638af
feat: add ops trace (#5483) 
Co-authored-by: takatost <takatost@gmail.com>
 2024-06-26 17:33:29 +08:00
Jyong 43335b5c87
delete the deprecated method (#5612) 2024-06-26 12:51:50 +08:00
Xiaoguang Sun 13fcd7a901
feat: Add program_name attribute to TiDB connection (#5499) 
Signed-off-by: Xiaoguang Sun <sunxiaoguang@gmail.com>
 2024-06-24 14:41:07 +08:00
tmuife 6a09409ec9
Add Oracle23ai as a vector datasource (#5342) 
Co-authored-by: walter from vm <walter.jin@oracle.com>
 2024-06-22 01:48:07 +08:00
Bowen Liang 39c14ec7c1
improve: unify Excel files parsing in either xls or xlsx file format by Pandas (#4965) 2024-06-20 16:14:49 +08:00
Bowen Liang aed56b1a8f
fix: Revert "feat: initial support for Milvus 2.4.x (#3795)" downgrading to 2.3.x for Linux arm64 installation failure (#5414) 2024-06-20 11:18:05 +08:00
quicksand 0105129fa8
fix bug: tencent vdb #5378 (#5408) 2024-06-20 10:37:39 +08:00
Bowen Liang c923684edd
chore: extract retrival method literal values into enum (#5060) 2024-06-19 16:05:27 +08:00
Bowen Liang bb33ffc332
feat: initial support for Milvus 2.4.x (#3795) 2024-06-19 13:55:44 +08:00
baojingyu d160d1ed02
feat: support opensearch approximate k-NN (#5322) 2024-06-19 12:44:33 +08:00
sino be3512aa57
fix: unable to reindex documents (#5276) 2024-06-17 17:19:43 +08:00
takatost 12c815c597
fix: ExtractSetting optional value missing None as default val (#5238) 2024-06-15 02:58:47 +08:00
Jyong ba5f8afaa8
Feat/firecrawl data source (#5232) 
Co-authored-by: Nicolas <nicolascamara29@gmail.com>
Co-authored-by: chenhe <guchenhe@gmail.com>
Co-authored-by: takatost <takatost@gmail.com>
 2024-06-15 02:46:02 +08:00
quicksand 4080f7b8ad
feat: support tencent vector db (#3568) 2024-06-14 19:25:17 +08:00