github-featured/deepset-ai--haystack
2
0
Fork 0
Code Issues Pull Requests Deployments Models Agent Notes Packages Projects Releases Wiki Activity
6,004 Commits 128 Branches 321 Tags
toolset-refactor
Commit Graph

3182 Commits

Author SHA1 Message Date
anakin87 b05954b29f Merge branch 'main' into toolset-refactor 2026-08-11 10:18:23 +02:00
anakin87 3bc0b51fa6 clarifu 2026-08-11 10:18:20 +02:00
Dhanush Pillay 0c3a485a67 feat: Add link_format to PyPDFToDocument and PDFMinerToDocument (#10677) (#12273) 
Co-authored-by: Julian Risch <julian.risch@deepset.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
 2026-08-10 15:21:46 +00:00
anakin87 4eedc4c09d rm unnecessary method 2026-08-10 14:59:35 +02:00
Singaraj B 50eebe097b fix: serialize and deserialize typing.Literal (#12286) 2026-08-10 14:08:29 +02:00
Anas Khan b4996ae3c4 fix: divide DocumentMAPEvaluator average precision by all relevant documents (#12255) 
Signed-off-by: Anas Khan <83116240+anxkhn@users.noreply.github.com>
Co-authored-by: Julian Risch <julian.risch@deepset.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
 2026-08-10 11:29:00 +00:00
anakin87 18c78e659f reduce diff 2026-08-10 11:50:11 +02:00
anakin87 cb6094ec0f simplify 2026-08-10 10:46:32 +02:00
Simon Schampijer 93e211774e feat: make MockChatGenerator.response_fn tool-aware (#12265) 
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Co-authored-by: Sebastian Husch Lee <10526848+sjrl@users.noreply.github.com>
Co-authored-by: Sebastian Husch Lee <sjrl423@gmail.com>
 2026-08-10 10:01:35 +02:00
David S. Batista 902fd98484 fix: increasing telemetry event version to Pipeline run (3.x) (#12256) 2026-08-10 07:44:20 +02:00
anakin87 a3057fa570 more 2026-08-07 20:17:35 +02:00
anakin87 48c31143a0 align 2026-08-07 20:07:55 +02:00
anakin87 5a37f397c7 less changes 2026-08-07 19:56:28 +02:00
anakin87 725b405d73 clean uo 2026-08-07 18:33:20 +02:00
anakin87 0987b78c8a refactor 2026-08-07 17:45:37 +02:00
anakin87 bcbbd9a455 toolset refactor POC 2026-08-07 16:38:34 +02:00
Yu Le 07c5206898 fix: validate runtime top_k in MetaFieldRanker (#12262) 
Co-authored-by: David S. Batista <dsbatista@gmail.com>
 2026-08-07 13:38:19 +02:00
Sebastian Husch Lee 49f8d2d549 fix: make strict datetime filter comparison opt-in (#12257) 2026-08-07 09:03:19 +02:00
Bharadwaj Pendyala 7038ed9d32 fix: deep-copy document metadata in Markdown, CSV and hierarchical splitters (#12249) 2026-08-07 08:57:54 +02:00
David S. Batista a42f62a820 fix: InMemoryDocumentStore adding optional parameter filters to get_metadata_field_unique_values() (#12258) 2026-08-06 14:06:04 +00:00
Sebastian Husch Lee e99ca20fe7 fix: Fix State restoration in Agent from a before_run hook (#12253) 2026-08-06 10:14:58 +02:00
Sebastian Husch Lee aeeb9dc749 chore: Update name from ContextCompactionHook to CompactionHook (#12252) 2026-08-06 06:50:30 +00:00
Sebastian Husch Lee ab4244404a feat: Add tool result pruning compactor (#12243) 
Co-authored-by: Stefano Fiorucci <stefanofiorucci@gmail.com>
 2026-08-05 14:59:00 +00:00
manjunath bhaskar f1f19e39d2 fix: DocumentJoiner top_k=0 is treated as unset instead of returning … (#12217) 
Co-authored-by: David S. Batista <dsbatista@gmail.com>
 2026-08-05 14:40:45 +00:00
LHMQ878 1cd96fcd95 fix: don't strip title when it is a name, not a schema keyword (#12219) 
Co-authored-by: LHMQ878 <LHMQ878@users.noreply.github.com>
Co-authored-by: David S. Batista <dsbatista@gmail.com>
 2026-08-05 07:45:22 +00:00
Sebastian Husch Lee 86ed78ca72 chore: Moving some agent utils to utils file (#12238) 2026-08-05 06:17:10 +00:00
Sebastian Husch Lee c07313f9aa feat: Add OpenaiTokenCounter using their input tokens api (#12237) 2026-08-04 14:12:28 +00:00
Sebastian Husch Lee a96197c69d docs: Add docs for ContextCompactionHook and Compactors (#12233) 2026-08-04 12:24:55 +02:00
Sebastian Husch Lee 7010e1e818 Simplify pipeline state serialization logic (#12234) 
Co-authored-by: David S. Batista <dsbatista@gmail.com>
 2026-08-04 12:16:47 +02:00
Muhammad Furqan d69eb08483 fix: support negative indices in _ToolsetWrapper.__getitem__ (#11760) 
Co-authored-by: Julian Risch <julian.risch@deepset.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-authored-by: Julian Risch <julianrisch@gmx.de>
 2026-08-04 12:02:07 +02:00
苏紫辰 a68ed4ac2b fix: cancel sibling tasks when concurrent retrieval fails (#11967) 
Co-authored-by: bogdankostic <bogdankostic@web.de>
 2026-08-04 11:19:03 +02:00
Aarkin Karnik 3ec9d08ace fix: normalize ISO date equality filters (#11963) 
Co-authored-by: Julian Risch <julian.risch@deepset.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
 2026-08-04 11:11:10 +02:00
David Pavlovschii fcc13465c5 fix: round-trip serialization of mixed-type lists in schema-based serialization (#12202) 
Co-authored-by: Sebastian Husch Lee <10526848+sjrl@users.noreply.github.com>
 2026-08-04 09:56:38 +02:00
Aditya Raut 876fe3d55a fix: restore auto-variadic socket state in remove_component (#12206) 
Co-authored-by: Sebastian Husch Lee <10526848+sjrl@users.noreply.github.com>
 2026-08-04 09:55:49 +02:00
David S. Batista 7d980a4bd0 fix: in memory doc store get_unique_metadata_values() pagination and case-insensitive + Mixin tests updated (#12168) 
Co-authored-by: Sebastian Husch Lee <10526848+sjrl@users.noreply.github.com>
 2026-08-04 09:28:02 +02:00
Praveen Kumar Mittal 1453504721 fix: JSONConverter raises KeyError on ByteStream sources without file_path (#12208) 2026-08-04 07:03:24 +00:00
Divya Narahari 9c504d6bba fix: return independent metadata dicts from normalize_metadata (#12107) 
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Co-authored-by: Stefano Fiorucci <stefanofiorucci@gmail.com>
 2026-08-03 16:32:29 +02:00
Sebastian Husch Lee fe858a6294 feat: add context compaction hook (#12196) 2026-08-03 15:59:15 +02:00
Stefano Fiorucci 4875a247fe fix: improve meta collisions in Document serde (#12214) 2026-08-03 09:55:50 +02:00
iridescentWen 6acc8805d4 docs: fix docstring parameter names in three components (#12204) 2026-07-31 13:28:30 +02:00
Praveen Kumar Mittal 5a0724f196 fix: MSGToDocument raises KeyError on ByteStream sources without file_path (#12207) 
Co-authored-by: David S. Batista <dsbatista@gmail.com>
 2026-07-31 12:19:03 +02:00
Tim 3f73aa5e60 Do not count whitespace runs as words in RecursiveDocumentSplitter fallback (#12139) 
Co-authored-by: Timur Rakhmatullin <174210871+TimurRakhmatullin86@users.noreply.github.com>
Co-authored-by: David S. Batista <dsbatista@gmail.com>
 2026-07-31 11:38:30 +02:00
Stefano Fiorucci e3aa021da7 feat: add AgentTool (#12186) 2026-07-30 18:09:06 +02:00
Sebastian Husch Lee f452038c40 feat: add token counters (#12195) 2026-07-30 18:00:01 +02:00
Otis Cui 18679f6286 fix: AnswerBuilder returned referenced documents in scrambled order (#11953) 2026-07-30 13:11:51 +02:00
vidigoat 1f460e6207 fix: round-trip serialization of Callable types with an explicit parameter list (#12122) 
Co-authored-by: David S. Batista <dsbatista@gmail.com>
 2026-07-30 12:47:47 +02:00
Loi Nguyen b53e614d1c fix: MarkdownHeaderSplitter silently drops a trailing header with no body text (#12064) 
Co-authored-by: Loi Nguyen <1948922+lntutor@users.noreply.github.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
Co-authored-by: Julian Risch <julian.risch@deepset.ai>
 2026-07-29 20:53:19 +02:00
Sebastian Husch Lee c42f9bc2ba fix: deferred annotation pickup in Agent (#12185) 2026-07-29 11:08:48 +00:00
Yu Le 984fc43b13 fix!: validate runtime SentenceWindowRetriever window size (#12183) 
Co-authored-by: Julian Risch <julian.risch@deepset.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
 2026-07-29 10:29:30 +00:00
Sebastian Husch Lee 66fb9d0ded fix!: resume a pipeline snapshot taken on a later loop visit (#12162) 
Co-authored-by: David S. Batista <dsbatista@gmail.com>
 2026-07-29 06:09:13 +00:00