Skip to content

fix(api): require a Tencent vector database name and type-check the provider - #44325

Merged
asukaminato0721 merged 1 commit into
langgenius:mainfrom
ShousenZHANG:refactor/type-check-tencent-vector
Oct 19, 2026
Merged

asukaminato0721 merged 1 commit into
langgenius:mainfrom
ShousenZHANG:refactor/type-check-tencent-vector

Conversation

@ShousenZHANG

Copy link
Copy Markdown
Contributor

Summary

Part of #36288, following #44237.

#44237 left the Tencent vector store out of type checking. Its 40 diagnostics are fixed here, and both tencent_vector.py and its unit tests are removed from api/pyrefly-local-excludes.txt.

The real problem behind 12 of them: TencentConfig.database was str | None, and every client call passes it as database_name. With TENCENT_VECTOR_DB_DATABASE unset, the config was accepted, and the first call, create_database_if_not_exists(database_name=None), built DatabaseRequest(database=None), which protobuf sends as an empty database name. The error then came back from the server instead of naming the setting.

What changed

  • TencentConfig.database is required. A validate_config model validator raises config TENCENT_VECTOR_DB_DATABASE is required when it is missing or empty, the same check and message the Baidu and AnalyticDB configs use. The factory passes TENCENT_VECTOR_DB_DATABASE or "", like the URL next to it, so the validator reports the setting. A new parametrized test, test_config_requires_a_database, covers None and ""; it fails on main.
  • Index lists are typed list[vdb_index.IndexField], since the collection mixes filter, vector and sparse indexes (4 diagnostics).
  • Upsert documents are typed list[document.Document | dict[str, Any]], the type upsert takes.
  • Vector dimension on load. describe_collection returns the index base types; the "vector" index is the VectorIndex _create_collection made, so it is cast, and the dimension is only taken when the SDK reports one (it is int | None), keeping the 1024 default otherwise.
  • SDK annotations. tcvectordb declares filter: Union[Filter, str] = None and encode_queries returns one or many sparse vectors. The two filter=None arguments get a targeted # pyrefly: ignore[bad-argument-type], and the single-query result is cast, each with a comment.
  • Unit tests. The fake SDK modules are types.ModuleType objects, and assigning attributes to them produced 19 diagnostics. They are now set through vars(module)[...], with the values unchanged. One assertion after _load_collection() becomes assert not vector._enable_hybrid_search, because the is True assertion just above it left the attribute narrowed.

Verification

  • Pyrefly 1.3.2, as pinned in uv.lock, on tencent_vector.py and test_tencent_vector.py: 40 diagnostics before, zero after.
  • The production check that dev/pyrefly-check-local runs (excluding tests/ and the remaining list), with the provider packages imported from this branch: passed. pyrefly check --config=tests/unit_tests/pyrefly.toml: passed.
  • These all passed:
    • ruff format --check and ruff check on the whole api/ tree, without the cache
    • lint-imports: 68 contracts kept
    • dev/lint_response_contracts.py --fail-on-mismatch
    • scripts/check_no_new_getattr.py, check_no_new_controller_sqlalchemy.py, check_no_new_session_mock.py, check_no_new_base_model.py, check_no_new_test_doubles.py and check_no_spec_mock.py
  • pytest on Python 3.13:
    • vdb-tencent unit tests: 7 passed; the 5 existing test IDs are unchanged, plus the two new cases
    • test_config_requires_a_database against main: both cases fail with DID NOT RAISE ValueError
    • all vector store unit tests (providers/vdb/*/tests/unit_tests and tests/unit_tests/core/rag/datasource/vdb): 709 passed

Screenshots

Before After
N/A: backend config validation and typing; 40 diagnostics in excluded files. N/A: zero diagnostics; both files back in make type-check.

Checklist

  • This change requires a documentation update, included: Dify Document
  • I understand that this PR may be closed in case there was no previous discussion or issues. (This doesn't apply to typos!)
  • I've verified the change and added or updated tests where meaningful regression risk justifies coverage.
  • I've updated the documentation accordingly.
  • I ran make lint && make type-check (backend) and vp staged (frontend) to appease the lint gods (backend only: ran the underlying ruff, import-linter, guard-script and Pyrefly commands listed above)

From Claude Code

…rovider

TencentConfig accepted a missing database, and every client call then
sent database_name=None, which failed on the server with an opaque
error. Require it with the same "config ... is required" error the other
vector stores use, and bring tencent_vector.py and its unit tests under
type checking (40 diagnostics).
Copilot AI balanced review requested due to automatic review settings October 19, 2026 03:23

Copilot AI left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

An organization owner can choose the repository organization as the billing source for reviews requested by members with a paid Copilot license. Learn about Copilot code review billing. Review the Copilot code review billing settings.

@github-actions github-actions Bot added the size:M This PR changes 30-99 lines, ignoring generated files. label Oct 19, 2026
@ghfind-review ghfind-review Bot added the review: high ghfind author score; see https://ghfind.com label Oct 19, 2026
@github-actions

Copy link
Copy Markdown
Contributor

Pyrefly Diff

base → PR
--- /tmp/pyrefly_base.txt	2026-10-19 03:24:24.919492632 +0000
+++ /tmp/pyrefly_pr.txt	2026-10-19 03:24:16.123443185 +0000
@@ -372,44 +372,6 @@
   --> providers/vdb/vdb-tencent/tests/integration_tests/conftest.py:94:24
 ERROR Argument `FilterIndexConfig | None` is not assignable to parameter `filter_index_config` with type `FilterIndexConfig` in function `tcvectordb.rpc.model.collection.RPCCollection.__init__` [bad-argument-type]
   --> providers/vdb/vdb-tencent/tests/integration_tests/conftest.py:95:33
-ERROR Object of class `ModuleType` has no attribute `IndexType` [missing-attribute]
-  --> providers/vdb/vdb-tencent/tests/unit_tests/test_tencent_vector.py:99:5
-ERROR Object of class `ModuleType` has no attribute `MetricType` [missing-attribute]
-   --> providers/vdb/vdb-tencent/tests/unit_tests/test_tencent_vector.py:105:5
-ERROR Object of class `ModuleType` has no attribute `FieldType` [missing-attribute]
-   --> providers/vdb/vdb-tencent/tests/unit_tests/test_tencent_vector.py:106:5
-ERROR Object of class `ModuleType` has no attribute `Document` [missing-attribute]
-   --> providers/vdb/vdb-tencent/tests/unit_tests/test_tencent_vector.py:108:5
-ERROR Object of class `ModuleType` has no attribute `HNSWSearchParams` [missing-attribute]
-   --> providers/vdb/vdb-tencent/tests/unit_tests/test_tencent_vector.py:109:5
-ERROR Object of class `ModuleType` has no attribute `AnnSearch` [missing-attribute]
-   --> providers/vdb/vdb-tencent/tests/unit_tests/test_tencent_vector.py:110:5
-ERROR Object of class `ModuleType` has no attribute `Filter` [missing-attribute]
-   --> providers/vdb/vdb-tencent/tests/unit_tests/test_tencent_vector.py:111:5
-ERROR Object of class `ModuleType` has no attribute `KeywordSearch` [missing-attribute]
-   --> providers/vdb/vdb-tencent/tests/unit_tests/test_tencent_vector.py:112:5
-ERROR Object of class `ModuleType` has no attribute `WeightedRerank` [missing-attribute]
-   --> providers/vdb/vdb-tencent/tests/unit_tests/test_tencent_vector.py:113:5
-ERROR Object of class `ModuleType` has no attribute `HNSWParams` [missing-attribute]
-   --> providers/vdb/vdb-tencent/tests/unit_tests/test_tencent_vector.py:115:5
-ERROR Object of class `ModuleType` has no attribute `FilterIndex` [missing-attribute]
-   --> providers/vdb/vdb-tencent/tests/unit_tests/test_tencent_vector.py:116:5
-ERROR Object of class `ModuleType` has no attribute `VectorIndex` [missing-attribute]
-   --> providers/vdb/vdb-tencent/tests/unit_tests/test_tencent_vector.py:117:5
-ERROR Object of class `ModuleType` has no attribute `SparseIndex` [missing-attribute]
-   --> providers/vdb/vdb-tencent/tests/unit_tests/test_tencent_vector.py:118:5
-ERROR Object of class `ModuleType` has no attribute `BM25Encoder` [missing-attribute]
-   --> providers/vdb/vdb-tencent/tests/unit_tests/test_tencent_vector.py:120:5
-ERROR Object of class `ModuleType` has no attribute `document` [missing-attribute]
-   --> providers/vdb/vdb-tencent/tests/unit_tests/test_tencent_vector.py:122:5
-ERROR Object of class `ModuleType` has no attribute `enum` [missing-attribute]
-   --> providers/vdb/vdb-tencent/tests/unit_tests/test_tencent_vector.py:123:5
-ERROR Object of class `ModuleType` has no attribute `index` [missing-attribute]
-   --> providers/vdb/vdb-tencent/tests/unit_tests/test_tencent_vector.py:124:5
-ERROR Object of class `ModuleType` has no attribute `RPCVectorDBClient` [missing-attribute]
-   --> providers/vdb/vdb-tencent/tests/unit_tests/test_tencent_vector.py:126:5
-ERROR Object of class `ModuleType` has no attribute `VectorDBException` [missing-attribute]
-   --> providers/vdb/vdb-tencent/tests/unit_tests/test_tencent_vector.py:127:5
 ERROR Object of class `ModuleType` has no attribute `VectorType` [missing-attribute]
   --> providers/vdb/vdb-tidb-vector/tests/unit_tests/test_tidb_vector.py:77:5
 ERROR Object of class `ModuleType` has no attribute `Vector` [missing-attribute]

@github-actions

Copy link
Copy Markdown
Contributor

Pyrefly Type Coverage

Metric Base PR Delta
Type coverage 75.38% 75.37% -0.00%
Strict coverage 75.05% 75.05% -0.00%
Typed symbols 62,554 62,554 0
Untyped symbols 20,525 20,528 +3
Modules 3802 3802 0

@codecov

codecov Bot commented Oct 19, 2026 •

Copy link
Copy Markdown

Codecov Report

❌ Patch coverage is 97.43590% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 88.42%. Comparing base (96202aa) to head (718b7d0).
⚠️ Report is 4 commits behind head on main.

Files with missing lines Patch % Lines
...vdb-tencent/src/dify_vdb_tencent/tencent_vector.py 93.33% 0 Missing and 1 partial ⚠️
Additional details and impacted files
@@           Coverage Diff           @@
##             main   #44325   +/-   ##
=======================================
  Coverage   88.41%   88.42%           
=======================================
  Files        5516     5516           
  Lines      330388   330445   +57     
  Branches    66084    66101   +17     
=======================================
+ Hits       292120   292188   +68     
+ Misses      33073    33068    -5     
+ Partials     5195     5189    -6     
Flag Coverage Δ
api 88.63% <97.43%> (+0.02%) ⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

@asukaminato0721
asukaminato0721 added this pull request to the merge queue Oct 19, 2026
@asukaminato0721 asukaminato0721 self-assigned this Oct 19, 2026
Merged via the queue into langgenius:main with commit 08a220f Oct 19, 2026
42 checks passed
Awhitter added a commit to Awhitter/dify that referenced this pull request Oct 25, 2026
* feat(agent): stream Agent V2 text output in plain runs (langgenius#43986)

Co-authored-by: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>

* refactor(api): read provider configuration cache rows via pydantic (langgenius#44055)

* refactor(web): use Select for the Studio app type filter (langgenius#44070)

* refactor(web): model app settings with form primitives (langgenius#44073)

* fix(dify-ui): show destructive colors on keyboard focus (langgenius#44079)

* ci: require TypeBase for new models (langgenius#44087)

Co-authored-by: Codex <codex@openai.com>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>

* ci: enable Ruff E721 (langgenius#44096)

Co-authored-by: Codex <codex@openai.com>

* ci: enable Ruff SIM210 (langgenius#44095)

Co-authored-by: Codex <codex@openai.com>

* fix(web): align API key dialog lifecycle and copy behavior (langgenius#44077)

* test(dify-ui): wait for time picker visibility before axe scan (langgenius#44105)

* ci: enable Ruff B006 (langgenius#44099)

Co-authored-by: Codex <codex@openai.com>

* ci: enable Ruff FURB152 (langgenius#44091)

Co-authored-by: Codex <codex@openai.com>

* ci: enable Ruff E711 (langgenius#44088)

Co-authored-by: Codex <codex@openai.com>

* fix(knowledge): only batch-enable segments whose indexing completed (langgenius#44017)

* fix(nodejs-client): preserve SSE field value whitespace (langgenius#44083)

* fix(api): stage hit-testing query log with flush instead of committing caller session (langgenius#44084)

* fix(dify-agent): treat a confirmed-missing Home Snapshot as a success on Enterprise delete (langgenius#44085)

Co-authored-by: D41910 <d41910@users.noreply.github.com>

* fix(nodejs-client): override default headers case-insensitively (langgenius#44090)

* fix(nodejs-client): allow empty keyword filters in list APIs (langgenius#44097)

* fix(nodejs-client): replace stale user fields in native upload forms (langgenius#44120)

* fix(api): accept MCP protocol version 2025-11-25 (langgenius#44066)

* fix(api): compare datasource credentials as JSON on MySQL (langgenius#44039)

* fix(structured_output): skip reasoning JSON when selecting output (langgenius#44005)

Co-authored-by: crazywoola <427733928@qq.com>

* ci: enable Ruff E722 (langgenius#44098)

Co-authored-by: Codex <codex@openai.com>

* ci: enable Ruff UP032 (langgenius#44092)

Co-authored-by: Codex <codex@openai.com>

* ci: enable Ruff B905 (langgenius#44094)

Co-authored-by: Codex <codex@openai.com>

* ci: enable Ruff SIM113 (langgenius#44102)

Co-authored-by: Codex <codex@openai.com>

* fix(deps): upgrade Next.js to 16.4.1 for October security fixes (langgenius#44127)

* docs: record overlay ownership and focus lessons (langgenius#44126)

* refactor(workflow): use route identity and Console Query for webhook URLs (langgenius#44135)

* refactor(workflow): own preview chat requests with Console Query (langgenius#44136)

* perf(rag): reduce knowledge retrieval post-processing overhead (langgenius#43665)

* fix(api): fix typos in logger warning and docstrings (langgenius#44134)

* refactor(api): enable strict typing for enterprise telemetry exporter tests (langgenius#44138)

* refactor(api): enable strict typing for prompt transform tests (langgenius#44139)

* ci: enable Ruff TID252 (langgenius#44108)

Co-authored-by: Codex <codex@openai.com>

* ci: enable Ruff B026 (langgenius#44100)

Co-authored-by: Codex <codex@openai.com>

* ci: enable Ruff SIM103 (langgenius#44110)

Co-authored-by: Codex <codex@openai.com>

* ci: enable Ruff FURB113 (langgenius#44109)

Co-authored-by: Codex <codex@openai.com>

* refactor(tools): remove redundant nested cast in MCP tool (langgenius#44132)

* refactor(api): use Pydantic to validate crawl cache key (langgenius#44131)

* fix(mcp): collect all tool discovery pages (langgenius#43842)

* fix(deps): override fflate to 0.7.5 for CVE-2026-45820 (langgenius#44148)

* refactor(web): let layouts own viewport height instead of root h-full chain (langgenius#44143)

* refactor(web): centralize AppIcon data and Avatar rendering (langgenius#44149)

* fix(agent): fail closed on Landlock errors and require local tokens (langgenius#44152)

* chore(deps): bump mako from 1.3.12 to 1.4.2 in /api (langgenius#43607)

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

* chore(deps): bump multidict from 6.7.0 to 6.9.1 in /api (langgenius#43610)

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

* chore(deps): bump fsspec from 2026.3.0 to 2026.6.0 in /dify-agent (langgenius#43606)

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

* chore(deps): bump oauthlib from 3.3.1 to 4.0.0 in /api (langgenius#43257)

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

* chore(deps): bump pydantic-ai-slim from 2.31.0 to 2.53.0 in /api (langgenius#43725)

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

* ci: enable Ruff E712 (langgenius#44114)

Co-authored-by: Codex <codex@openai.com>

* refactor: migrate app tracing queries and harden monitoring configuration (langgenius#44150)

* ci: enable Ruff SIM108 (langgenius#44113)

Co-authored-by: Codex <codex@openai.com>

* refactor(api): remove redundant casts in tencent trace unit tests (langgenius#44158)

* fix(api): preserve timestamp ties in message pagination (langgenius#44023)

Co-authored-by: mac <mac@macdeMacBook-Pro.local>

* fix(web): keep the value selector of a Code node input when regenerating code (langgenius#44075)

* fix(e2e): share sandbox auth token across managed runtimes (langgenius#44164)

* chore: remove KnowledgeFS and New Knowledge from main (langgenius#44163)

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>

* ci: enable Ruff SIM105 (langgenius#44112)

Co-authored-by: Codex <codex@openai.com>

* refactor(web): retire the app detail store in favour of the query cache (langgenius#44166)

* fix(api): let a removed active member sign in with a re-invite (langgenius#43667)

Co-authored-by: Matt Van Horn <455140+mvanhorn@users.noreply.github.com>
Co-authored-by: Stream <Stream_2@qq.com>

* refactor(web): clarify App modes and scope workflow editor resources (langgenius#44171)

* chore: mark generated files as linguist-generated (langgenius#44180)

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>

* feat(api): let agents build, configure and ship Dify apps over openapi (langgenius#43067)

* ci: enable Ruff B007 (langgenius#44107)

Co-authored-by: Codex <codex@openai.com>

* ci: enable Ruff F821 (langgenius#44101)

Co-authored-by: Codex <codex@openai.com>

* ci: enable Ruff UP007 (langgenius#44106)

Co-authored-by: Codex <codex@openai.com>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>

* ci: enable Ruff N806 (langgenius#44116)

Co-authored-by: Codex <codex@openai.com>

* fix: accumulate workflow elapsed time across resumptions (langgenius#44161)

* refactor(web): migrate handwritten console API calls to the generated oRPC contract (langgenius#44169)

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>

* chore(deps): bump unstructured from 0.24.0 to 0.24.1 in /dify-agent (langgenius#44189)

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

* fix(web): normalize collapsible indicator rotation (langgenius#44193)

* fix(workflow): generate template-compatible snippet node IDs (langgenius#43838)

* fix(api): return protocol errors for unsupported MCP requests (langgenius#44197)

* fix(api): align hosted moderation provider validation (langgenius#44200)

* feat(api): add inner API for enterprise admin app list and detail (langgenius#44044)

Co-authored-by: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>

* fix(dify-ui): let portal order stack overlays so nested modals block clicks (langgenius#44214)

* fix(nodejs-client): collect workflow text_chunk and apply message_replace in toText (langgenius#44177)

* fix(rag): preserve newlines in recursive character fallback (langgenius#44179)

* refactor(web): remove unnecessary type assertions with a migration-only tsslint rule (langgenius#44219)

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>

* refactor: remove redundant TypeScript type aliases (langgenius#44225)

* refactor(web): replace straightforward explicit any types (langgenius#44228)

* refactor(dify-ui): give AlertDialog a header part and design defaults (langgenius#44223)

Co-authored-by: Claude <noreply@anthropic.com>

* refactor(web): remove unnecessary use prefixes (langgenius#44231)

* fix(web): clear Storybook and file preview lint suppressions (langgenius#44233)

* refactor(web): replace explicit any in test fixtures and mocks (langgenius#44235)

* refactor(dify-ui): compose dialogs from the Design Kit anatomy (langgenius#44221)

* refactor(web): exercise real frontend behavior in tests (langgenius#44243)

* chore(web): prefer vinext for frontend development (langgenius#44246)

* chore(contracts): upgrade @hey-api/openapi-ts to 0.99.0 (langgenius#44247)

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>

* refactor(dify-ui): scroll dialog bodies and give AlertDialog a body (langgenius#44248)

Signed-off-by: yyh <yuanyouhuilyz@gmail.com>
Co-authored-by: Stephen Zhou <hi@hyoban.cc>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>

* fix(web): align app and agent export feedback (langgenius#44251)

* refactor(web): migrate Notion page preview to Console Query (langgenius#44252)

* test(web): include untested sources in coverage (langgenius#44254)

* test(web): replace low-value suites with behavioral coverage (langgenius#44260)

* fix(api): preserve published triggers on draft sync (langgenius#44068)

Co-authored-by: mac <mac@macdeMacBook-Pro.local>

* fix: don't stringify list values for union-typed MCP tool parameters (langgenius#44182)

Co-authored-by: Stream <Stream_2@qq.com>

* fix(cli): observe SSE read failures while consumers process events (langgenius#43943)

Co-authored-by: huyengiang101086-spec <270684046+huyengiang101086-spec@users.noreply.github.com>

* fix(skills): preserve PPTX presentation slide order (langgenius#43893)

Co-authored-by: LimbC-C <329404569+LimbC-C@users.noreply.github.com>

* fix(dify): retry failed archive deletion instead of reporting success (langgenius#44061)

Co-authored-by: Stream <Stream_2@qq.com>

* refactor(api): enable strict typing for tasks unit tests (langgenius#44255)

* refactor(api): enable strict typing for easy UI app config manager tests (langgenius#44244)

* test(web): consolidate suites around real feature behavior (langgenius#44274)

* fix(web): separate app dialog group and control labels (langgenius#44261)

Co-authored-by: Stephen Zhou <38493346+hyoban@users.noreply.github.com>

* fix(ops): bind tracing service accounts to the app workspace (langgenius#44208)

* refactor(api): enable strict typing for libs unit tests (langgenius#44276)

* refactor(ui): clarify dialog header close button anatomy (langgenius#44283)

* fix(plugins): align provider overlays and background update feedback (langgenius#44284)

* fix(web): merge parallel tool results into their own agent thoughts (langgenius#43843)

Co-authored-by: Stream <Stream_2@qq.com>

* test(web): exercise real workflow canvas controls and note actions (langgenius#44287)

* test(web): cover node context menu clicks and drop hooks-store facade mocks (langgenius#44288)

* fix(web): label the Notion page search field (langgenius#44290)

* ci: enable Ruff B903 (langgenius#44117)

Co-authored-by: Codex <codex@openai.com>

* refactor(web): migrate Mermaid diagrams to Pele (langgenius#44293)

* fix(tests): satisfy Ruff B903 for hit-test fake (langgenius#44299)

Co-authored-by: mac <mac@macdeMacBook-Pro.local>

* refactor(api): enable strict typing for configs unit tests (langgenius#44294)

Co-authored-by: Asuka Minato <i@asukaminato.eu.org>

* refactor(api): bring vector store provider sources under type checking (langgenius#44237)

Co-authored-by: Asuka Minato <i@asukaminato.eu.org>

* ci: enable Ruff PT011 (langgenius#44122)

Co-authored-by: Codex <codex@openai.com>

* ci: enable Ruff E402 (langgenius#44123)

Co-authored-by: Codex <codex@openai.com>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>

* refactor: migrate Dataset to TypeBase (langgenius#40817)

Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: Codex <codex@openai.com>

* ci: enable Ruff F841 (langgenius#44115)

Co-authored-by: Codex <codex@openai.com>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>

* fix: honor RTF Unicode fallback lengths in skill attachments (langgenius#43789)

Co-authored-by: LimbC-C <329404569+LimbC-C@users.noreply.github.com>

* chore(deps): bump patch and minor pnpm catalog dependencies (langgenius#44319)

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>

* chore(deps): drop unused postcss-selector-parser overrides (langgenius#43292)

* fix: preserve DOCX paragraph text in skill attachments (langgenius#43718)

Co-authored-by: LimbC-C <329404569+LimbC-C@users.noreply.github.com>

* feat(cli): let agents build Dify apps with difyctl and an embedded skill (langgenius#43819)

* refactor(api): enable strict typing for easy UI dataset and prompt template tests (langgenius#44320)

* fix(api): require a Tencent vector database name and type-check the provider (langgenius#44325)

* fix(web): open command menu instantly with a lighter backdrop (langgenius#44330)

* fix(cli): serialize token file updates (langgenius#44329)

Co-authored-by: mac <mac@macdeMacBook-Pro.local>

* fix(web): parse destructured and multi-line main signatures in Code node (langgenius#44046)

Co-authored-by: Stream <Stream_2@qq.com>

* refactor(web): simplify Mermaid preview layout and interactions (langgenius#44335)

* fix(web): stabilize Logs loading and conversation dialogs (langgenius#44338)

* fix(web): improve app access point accessibility (langgenius#44220)

Co-authored-by: lyzno1 <92089059+lyzno1@users.noreply.github.com>

* fix(vdb): restore TiDB on Qdrant full-text query filtering (langgenius#44339)

* fix(web): remove redundant plugin permission success toast (langgenius#44342)

* fix(web): handle default model form submission feedback (langgenius#44345)

* fix(web): save the shown timezone on new schedule trigger nodes (langgenius#43532)

* test(web): exercise variable assignment with real workflow stores (langgenius#44349)

* fix(api): initialize default LLMs during onboarding (langgenius#44350)

* fix: replace str(e) with generic messages in error responses (langgenius#43049)

Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>

* ci: enable Ruff B904 (langgenius#44121)

Co-authored-by: Codex <codex@openai.com>
Co-authored-by: hjlarry <hjlarry@163.com>

* refactor(api): enable strict typing for tool encryption tests (langgenius#44348)

* refactor(api): enable strict typing for data source auth tests (langgenius#44347)

* refactor(workflow): adopt Graphon 0.8 execution orchestration (langgenius#40277)

Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>

* test(web): verify canvas updates and debounced variable edits (langgenius#44361)

* refactor(api): enable strict typing for file fetcher, common controller and app factory tests (langgenius#44379)

* refactor(api): enable strict typing for unstructured extractor tests (langgenius#44380)

* fix(vdb): balance multi-keyword Qdrant results (langgenius#44352)

Co-authored-by: mac <mac@macdeMacBook-Pro.local>

* fix(cli): accept BOM-prefixed JSON input (langgenius#44386)

Co-authored-by: mac <mac@macdeMacBook-Pro.local>

* chore(deps): bump aiohttp from 3.14.3 to 3.14.4 in /api (langgenius#44387)

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: Yunlu Wen <yunlu.wen@dify.ai>

* refactor(api): enable strict typing for remaining libs unit tests (langgenius#44388)

* refactor(api): enable strict typing for storage unit tests (langgenius#44389)

* feat: add reusable network access group APIs (langgenius#41025)

Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: CodingOnStar <hanxujiang@dify.com>
Co-authored-by: hjlarry <hjlarry@163.com>
Co-authored-by: yyh <92089059+lyzno1@users.noreply.github.com>
Co-authored-by: yyh <yuanyouhuilyz@gmail.com>

* feat: confirm clearing default models and refresh values on open (langgenius#44354)

Co-authored-by: yyh <yuanyouhuilyz@gmail.com>

* fix(cli): change local YAML files only through a locked update (langgenius#44375)

* chore(i18n): sync translations with en-US (langgenius#44397)

Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>

* fix: retain model selector focus after clearing defaults (langgenius#44399)

* fix(dataset): batch segment and document cleanup with retries (langgenius#38519)

Co-authored-by: hjlarry <hjlarry@163.com>

* test(web): exercise workflow actions against real stores (langgenius#44381)

* fix(web): upgrade TanStack Hotkeys and isolate search commands (langgenius#44401)

* fix(web): parse Python Code node parameters with commas (langgenius#44337)

Co-authored-by: mac <mac@macdeMacBook-Pro.local>

* refactor(tooling): centralize lint and format maintenance (langgenius#44404)

* refactor(web): unify app metadata forms and dialog ownership (langgenius#44405)

* refactor(tooling): migrate remaining sources to TypeScript (langgenius#44410)

* fix(web): improve app deploy accessibility (langgenius#44378)

Co-authored-by: yyh <yuanyouhuilyz@gmail.com>

* fix(e2e): seed workspace defaults and refresh post-merge filters (langgenius#44412)

* chore: add frontend code owner backups and refresh coverage (langgenius#44413)

* fix(agent): finalize run as failed when run_started cannot be appended (langgenius#43959)

* refactor(web): use generated oRPC query keys (langgenius#44415)

* fix: keep archive deletion loading until completion (langgenius#44203)

* refactor(web): remove redundant annotation list success toasts (langgenius#44031)

* fix(web): apply log annotations to the displayed branch (langgenius#43238)

* feat: optimize access token ui (langgenius#44422)

* fix(agent): use Redis 6.0-compatible event replay (langgenius#44420)

* test(api): isolate Celery initialization app state (langgenius#44433)

* refactor(web): migrate console API helpers to generated contracts (langgenius#44430)

* refactor(web): query installation bootstrap status (langgenius#44382)

* fix(workflow): track human input forms by execution identity (langgenius#42819)

* refactor(api): enable strict typing for rate limit test fixtures (langgenius#44391)

* refactor(api): enable strict typing for core helper tests (langgenius#44390)

* chore(web): guard handwritten requests with native lint rules (langgenius#44436)

* fix(nodejs-client): reject streamed API errors in toText (langgenius#44199)

Co-authored-by: Stream <Stream_2@qq.com>

* fix(cli): load PowerShell module in rate-limit test (langgenius#44439)

* fix(api): write built-in document metadata in a single format (langgenius#44218)

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Stream <Stream_2@qq.com>

* feat(workflow): execute workflow tools as nested containers (langgenius#42820)

Co-authored-by: WH-2099 <wh2099@pm.me>

* feat(workflow): expose authorized nested tool traces (langgenius#42821)

Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>

* feat(workflow): add nested tool trace drilldown (langgenius#42822)

* fix(tracing): exclude called tool internals from external exports (langgenius#42823)

* chore(web): replace deprecated third-party APIs (langgenius#44466)

* fix(plugin): pin daemon dispatches to package identifiers (langgenius#44396)

* fix(web): enforce schema depth limits through nested arrays (langgenius#44295)

* feat: network access group add rbac check (langgenius#44464)

* fix(api): validate builtin tool credentials outside the create lock (langgenius#44371)

Co-authored-by: Stream <Stream_2@qq.com>

* fix(trigger): stop leaking internal exception text from webhook 500 responses (langgenius#44395)

Co-authored-by: Stream <Stream_2@qq.com>

* fix(web): preserve Studio list filters across navigation (langgenius#44429)

Co-authored-by: Stream <Stream_2@qq.com>

* test(web): wait for goto-anything focus (langgenius#44454)

Co-authored-by: mac <mac@macdeMacBook-Pro.local>

* test(web): exercise workflow configuration and controls with real stores (langgenius#44471)

* test(web): exercise workflow controls and access points through real owners (langgenius#44423)

* fix(agent): attribute nested tool activity to its source (langgenius#42824)

* fix(cli): respect cancellation during device authorization retries (langgenius#44315)

Co-authored-by: huyengiang101086-spec <270684046+huyengiang101086-spec@users.noreply.github.com>

* feat(cli): send a folder as one part per file (langgenius#44356)

Co-authored-by: yunlu.wen <yunlu.wen@dify.ai>

* fix(web): handle autofilled Web App login credentials (langgenius#44474)

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>

* fix(api): handle non-object JSON auth error bodies (langgenius#44470)

* refactor: migrate agent models to TypeBase (langgenius#40816)

Co-authored-by: Codex <codex@openai.com>

* refactor(api): type-check vector store provider unit tests (langgenius#44458)

* refactor(api): enable strict typing for utils unit tests (langgenius#44331)

* test(web): replace node action mocks with real canvas behavior (langgenius#44495)

* fix(web): enforce focus style ownership and simplify invite recipients (langgenius#44504)

* fix: improve customization settings accessibility (langgenius#44492)

* fix(web): label workspace default model selectors (langgenius#44500)

* refactor(ui): split toast stories and align app toaster naming (langgenius#44510)

* chore(weaviate): bump bundled Weaviate to 1.39.13 (CVE-2026-111943) (langgenius#44508)

* chore: bump Node.js runtime to 24.22.1 (langgenius#44511)

* feat(tracing): export Agent app and Workflow Agent execution trees (langgenius#44144)

Co-authored-by: yyh <yuanyouhuilyz@gmail.com>

* feat(tracing): configure Agent tracing and link workflow runs to traces (langgenius#44186)

Co-authored-by: Yansong Zhang <916125788@qq.com>

* fix(web): make document metadata panel scrollable (langgenius#43621)

* test: split Build exit scenarios to reduce timeout flakiness (langgenius#44519)

* chore(web): improve Billing settings accessibility (langgenius#44487)

Co-authored-by: yyh <92089059+lyzno1@users.noreply.github.com>

* fix(deps): upgrade certifi and pillow (langgenius#44524)

* refactor(api): detach console app trigger management (langgenius#44457)

* refactor(web): highlight markdown code blocks with @pierre/highlights (langgenius#44522)

* fix(api): enforce cloud app quotas on DSL imports (langgenius#44525)

* perf(api): skip saved prompts when loading chat history (langgenius#44521)

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>

* chore: improve Members settings accessibility (langgenius#44484)

* fix(web): preserve inline agent layout and clarify configure ownership (langgenius#44534)

Co-authored-by: hjlarry <hjlarry@163.com>

* fix: improve archived logs accessibility (langgenius#44535)

Co-authored-by: yyh <yuanyouhuilyz@gmail.com>

* fix: improve Preferences accessibility (langgenius#44539)

* fix: improve IP Policies accessibility (langgenius#44548)

* fix: preserve knowledge sidebar when adding files (langgenius#44552)

* fix: improve sort control accessibility and type safety (langgenius#44564)

* chore: type-check with tsrs (langgenius#44562)

* fix(api): stop leaking Redis result connections from fire-and-forget groups (langgenius#44551)

Co-authored-by: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>

* fix(api): upgrade NLTK to 3.10.4 (langgenius#44581)

* feat(agent): edit config prompts through the sandbox CLI (langgenius#44561)

* chore: port Document to TypeBase (langgenius#36084)

Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: Codex <codex@openai.com>

* feat: KnowledgeFS and New Knowledge (langgenius#44165)

Co-authored-by: Jyong <76649700+JohnJyong@users.noreply.github.com>
Co-authored-by: hjlarry <hjlarry@163.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: zhaohao1004 <zhaohao1004@gmail.com>
Co-authored-by: yyh <yuanyouhuilyz@gmail.com>
Co-authored-by: FFXN <lizy@dify.ai>

* refactor: migrate CustomizedSnippet to TypeBase (langgenius#40818)

Co-authored-by: Codex <codex@openai.com>

* fix(php-sdk): preserve custom API base URL paths (langgenius#44569)

* fix(api): combine UTF-16 surrogate pairs in RTF extraction (langgenius#44565)

Co-authored-by: jayachandra2003 <jayachandravennam.jc@email.com>

* chore(i18n): sync translations with en-US (langgenius#44566)

Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>

* refactor(ui): align floating positioning with Base UI (langgenius#44588)

* chore(i18n): sync translations with en-US (langgenius#44595)

Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>

* chore(deps): bump certifi from 2026.4.22 to 2026.10.19 in /dify-agent (langgenius#44462)

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

* fix(web): avoid redundant document enable success toast (langgenius#44594)

* test(web): replace workflow owner doubles with real canvas behavior (langgenius#44590)

* chore(release): prepare version 1.18.0 (langgenius#44546)

* Exclude generated Vercel deployment cache from source checks

---------

Signed-off-by: dependabot[bot] <support@github.com>
Signed-off-by: yyh <yuanyouhuilyz@gmail.com>
Co-authored-by: Yunlu Wen <yunlu.wen@dify.ai>
Co-authored-by: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: gongfpp <1360187968@qq.com>
Co-authored-by: yyh <92089059+lyzno1@users.noreply.github.com>
Co-authored-by: Asuka Minato <i@asukaminato.eu.org>
Co-authored-by: Codex <codex@openai.com>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: D41910 <158550716+D41910@users.noreply.github.com>
Co-authored-by: D41910 <d41910@users.noreply.github.com>
Co-authored-by: leilei3167 <imleilei123@gmail.com>
Co-authored-by: 非法操作 <hjlarry@163.com>
Co-authored-by: Shreyansh Jain <shreyanshjain646@gmail.com>
Co-authored-by: crazywoola <427733928@qq.com>
Co-authored-by: 呆萌闷油瓶 <253605712@qq.com>
Co-authored-by: Power <26699525@qq.com>
Co-authored-by: Eddy ZHANG <eddy.zhang24@gmail.com>
Co-authored-by: Lokesh Chauhan <lokesh0186@gmail.com>
Co-authored-by: Xiyuan Chen <52963600+GareArc@users.noreply.github.com>
Co-authored-by: 盐粒 Yanli <yanli@dify.ai>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: Azusa010 <m3171470760@gmail.com>
Co-authored-by: mika <spacexstarship01@outlook.com>
Co-authored-by: mac <mac@macdeMacBook-Pro.local>
Co-authored-by: Yueliup <127648758+Yueliup@users.noreply.github.com>
Co-authored-by: Stephen Zhou <hi@hyoban.cc>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Matt Van Horn <mvanhorn@users.noreply.github.com>
Co-authored-by: Matt Van Horn <455140+mvanhorn@users.noreply.github.com>
Co-authored-by: Stream <Stream_2@qq.com>
Co-authored-by: 林玮 (Jade Lin) <linw1995@icloud.com>
Co-authored-by: xjh <87483213+xiajiahang@users.noreply.github.com>
Co-authored-by: Temitope Kadri <147322025+TemitopeKadri@users.noreply.github.com>
Co-authored-by: lorenzozane <w2898683706@gmail.com>
Co-authored-by: Yann Roberto <165009858+shoyann@users.noreply.github.com>
Co-authored-by: David Allada <davidanilallada@gmail.com>
Co-authored-by: huyengiang101086-spec <huyengiang101086@gmail.com>
Co-authored-by: huyengiang101086-spec <270684046+huyengiang101086-spec@users.noreply.github.com>
Co-authored-by: LimbC-C <limbcc2333@gmail.com>
Co-authored-by: LimbC-C <329404569+LimbC-C@users.noreply.github.com>
Co-authored-by: CRYPTONIKAV <29835619+CRYPTONIKAV@users.noreply.github.com>
Co-authored-by: Stephen Zhou <38493346+hyoban@users.noreply.github.com>
Co-authored-by: Liu Ziming <liuziming.lzm@alibaba-inc.com>
Co-authored-by: Metehan Biçer <metehanbicer006@gmail.com>
Co-authored-by: mechmaster20 <srujan.gowni@iitgn.ac.in>
Co-authored-by: Denis Ermakov <ide404@gmail.com>
Co-authored-by: Wu Tianwei <30284043+WTW0313@users.noreply.github.com>
Co-authored-by: Juhyuk choi <246447183+JHC56@users.noreply.github.com>
Co-authored-by: -LAN- <laipz8200@outlook.com>
Co-authored-by: zyssyz123 <916125788@qq.com>
Co-authored-by: CodingOnStar <hanxujiang@dify.com>
Co-authored-by: yyh <yuanyouhuilyz@gmail.com>
Co-authored-by: Joel <iamjoel007@gmail.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>
Co-authored-by: Alexey Shabalin <a.shabalin@gmail.com>
Co-authored-by: Codada <108800975+Redestiny@users.noreply.github.com>
Co-authored-by: wangxiaolei <fatelei@gmail.com>
Co-authored-by: 青阳寻雪 <142769065+lulalalual@users.noreply.github.com>
Co-authored-by: Tianyu Cai <cai18633758993@gmail.com>
Co-authored-by: WH-2099 <wh2099@pm.me>
Co-authored-by: Miyar <14232275+gakugaku@users.noreply.github.com>
Co-authored-by: xhdnoah <40939515+xhdnoah@users.noreply.github.com>
Co-authored-by: Duda Nogueira <dudanogueira@gmail.com>
Co-authored-by: zhsama <torvalds@linux.do>
Co-authored-by: Mesrop Harutyunyan <104313591+mesrop2008@users.noreply.github.com>
Co-authored-by: Takuya <83127305+Takuya-ops@users.noreply.github.com>
Co-authored-by: Jyong <76649700+JohnJyong@users.noreply.github.com>
Co-authored-by: zhaohao1004 <zhaohao1004@gmail.com>
Co-authored-by: FFXN <lizy@dify.ai>
Co-authored-by: Winoren <158550716+Winoren@users.noreply.github.com>
Co-authored-by: Jaya Chandra Vennam <149901816+jayachandra2003@users.noreply.github.com>
Co-authored-by: jayachandra2003 <jayachandravennam.jc@email.com>
Co-authored-by: ZengLiangliang <zll21377105@126.com>
Co-authored-by: Alec Whitters <alecwhitters@Alecs-MacBook-Pro-26.local>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

review: high ghfind author score; see https://ghfind.com size:M This PR changes 30-99 lines, ignoring generated files.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants