langgenius/dify
1
0
Fork 0
mirror of https://github.com/langgenius/dify.git synced 2026-07-25 13:36:01 -04:00
Code Issues 775 Packages Projects Releases 10 Wiki Activity
11,776 Commits 644 Branches 175 Tags
feat/llm-ft-timeout
Commit Graph

150 Commits

Author SHA1 Message Date
L1nSn0w dd7be50645 docs(llm): trim first-token timeout comments to the essentials 
Keep only the non-obvious constraints (unit contract, daemon no-keep-alive
assumption, ContextVar fail-open threading, replace-not-narrow intent) and
cut prose that restates the code. The module docstring in
first_token_timeout.py stays the single canonical description of the path.
 2026-07-17 11:18:31 +08:00
L1nSn0w 8f1504bf95 chore(llm): improve timeout observability and pin the graphon transform contract 
Name the configured window in the mid-stream stall message so a read
timeout after the first token is traceable to first_token_timeout_ms; log
invalid values ignored at the pop point; state honestly in _read_timeout_for
that the budget replaces (not narrows) the operator read timeout; pin with a
test that FirstTokenTimeoutError passes graphon's _transform_invoke_error
unchanged, and cover the converter's defensive drop.
 2026-07-17 11:18:31 +08:00
L1nSn0w d865d3d703 refactor(llm): switch first-token timeout config to milliseconds 
Rename the completion_params key to first_token_timeout_ms so the unit is
explicit in the DSL, and lower the default from 60s to 10s with a 100ms-10min
range: users who enable this gate are latency-sensitive, and 60s rarely
matches that intent. The ms->s conversion happens exactly once, at the pop
point in _normalize_completion_params; everything downstream (ModelInstance
field, ContextVar, httpx read timeout) stays in seconds.
 2026-07-17 11:18:31 +08:00
L1nSn0w d77df0b0a4 feat(llm): enforce first-token timeout in the plugin transport 
The timeout is carried to the plugin-daemon transport through a ContextVar
(core.plugin.impl.first_token_timeout) and applied as httpx's per-request
read timeout in BasePluginClient._stream_request. The daemon withholds the
response headers until the model's first token, so the read timeout measures
time-to-first-token directly; a ReadTimeout before the first line surfaces as
FirstTokenTimeoutError (dify-local, subclassing graphon's InvokeError), while
a later stall stays a plain transport error. A non-positive budget disables
the gate.
 2026-07-17 11:18:31 +08:00
Byron.wang ab3e4daa95 refactor: explicit DB session propagation across backend paths (#38559) 
Co-authored-by: WH-2099 <wh2099@pm.me>
 2026-07-15 06:48:28 +00:00
zyssyz123 925f97be20 feat: daily sync (#38593) 
Co-authored-by: 盐粒 Yanli <mail@yanli.one>
Co-authored-by: yyh <yuanyouhuilyz@gmail.com>
Co-authored-by: Joel <iamjoel007@gmail.com>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: QuantumGhost <obelisk.reg+git@gmail.com>
Co-authored-by: yyh <92089059+lyzno1@users.noreply.github.com>
Co-authored-by: 盐粒 Yanli <yanli@dify.ai>
 2026-07-09 10:35:15 +00:00
非法操作 6edce14e88 fix: can't debug model plugins (#38500) 2026-07-07 13:03:32 +00:00
Jashwanth Reddy Gummula dd0c4a2296 fix: resolve 36288 mypy errors (#37850) 
Co-authored-by: WH-2099 <wh2099@pm.me>
 2026-07-07 06:37:20 +00:00
Ryuta KOBAYASHI 64d72c6fb9 feat(api): pass app_id to model plugins for provider-side cost attribution (#35859) 
Co-authored-by: WH-2099 <wh2099@pm.me>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
 2026-07-06 08:54:16 +00:00
Harsh Kashyap 78ca1a8a76 fix(api): resolve plugin external user ids in backwards invocation (#38098) 
Co-authored-by: Harsh Kashyap <Harsh23Kashyap@users.noreply.github.com>
Co-authored-by: Harsh Kashyap <harshkashyap@Harshs-MacBook-Pro.local>
 2026-07-06 08:25:26 +00:00
chariri 904fadde20 refactor(api): remove member field compatibility (#37966) 
Co-authored-by: WH-2099 <wh2099@pm.me>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: Asuka Minato <i@asukaminato.eu.org>
 2026-07-06 07:00:43 +00:00
chariri c3b1508712 refactor(api): migrate workspace account endpoints to BaseModel (#37954) 2026-07-06 06:57:40 +00:00
yyh 1b3e8e9943 chore(agent-v2): sync daily changes (#38298) 
Signed-off-by: yyh <yuanyouhuilyz@gmail.com>
Co-authored-by: 盐粒 Yanli <mail@yanli.one>
Co-authored-by: Joel <iamjoel007@gmail.com>
Co-authored-by: zyssyz123 <916125788@qq.com>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: 盐粒 Yanli <yanli@dify.ai>
Co-authored-by: 林玮 (Jade Lin) <linw1995@icloud.com>
 2026-07-05 08:09:38 +00:00
非法操作 91cc50371b chore: use zstd for plugin model provider cache (#38382) 2026-07-03 08:58:30 +00:00
非法操作 059a1fe090 chore: compress large plugin_model obj in redis (#38374) 
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
 2026-07-03 07:32:07 +00:00
QuantumGhost f4ec608ef4 chore(api): upgrade graphon to v0.6.0, migrate HITL logic back to Dify (#38247) 
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
 2026-07-03 03:45:21 +00:00
Byron.wang 2f0c6c10bd refactor: thread explicit sessions through app retrieval paths (#38309) 
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
 2026-07-02 17:00:47 +00:00
WH-2099 7fb797c121 fix(api): keep provider refresh waiters single-flight (#38226) 2026-07-02 04:46:10 +00:00
Stephen Zhou 37bcb60c26 refactor(web): migrate plugin console contracts (#38252) 
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
 2026-07-01 15:34:52 +00:00
Pyuyi 200f8b800f fix(api): prevent plugin provider cache stampedes (#37388) 
Co-authored-by: VeraPyuyi <204892921+VeraPyuyi@users.noreply.github.com>
 2026-06-30 10:09:05 +00:00
wangxiaolei 0cc27dd401 chore: not use request.scoped session (#37421) 
Co-authored-by: WH-2099 <wh2099@pm.me>
 2026-06-22 19:38:24 +00:00
Evan 0fa43973b8 fix: clean unnecessary | None type annotations (#35557) (#36824) 
Co-authored-by: WH-2099 <wh2099@pm.me>
 2026-06-18 02:08:08 +00:00
Asuka Minato af99414fc1 chore: port isinstance to match case (#37271) 
Co-authored-by: WH-2099 <wh2099@pm.me>
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
 2026-06-18 00:18:03 +00:00
QuantumGhost f0b34bdeb4 feat(api): LLM polling support (#37462) 
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: WH-2099 <wh2099@pm.me>
 2026-06-17 23:34:33 +00:00
Evan a605e7f11e refactor: replace isinstance chains with match-case (#24487) (#37482) 2026-06-16 04:41:31 +00:00
Jingyi 9b74df21d0 feat(web): refine onboarding UI (#37433) 
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: yyh <yuanyouhuilyz@gmail.com>
Co-authored-by: Joel <iamjoel007@gmail.com>
Co-authored-by: hjlarry <hjlarry@163.com>
Co-authored-by: fatelei <fatelei@gmail.com>
Co-authored-by: Asuka Minato <i@asukaminato.eu.org>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: Xiyuan Chen <52963600+GareArc@users.noreply.github.com>
Co-authored-by: gigglewang <gigglewang@dify.ai>
Co-authored-by: Yunlu Wen <yunlu.wen@dify.ai>
Co-authored-by: chariri <w@chariri.moe>
Co-authored-by: Evan <2869018789@qq.com>
Co-authored-by: yyh <92089059+lyzno1@users.noreply.github.com>
 2026-06-15 08:47:15 +00:00
jashwanth_reddy_gummula 0ca39c96db refactor: replace if isinstance with match case (#37412) 2026-06-14 13:00:01 +00:00
Evan 99351d2f98 refactor: convert remaining isinstance chains to match/case (part 9) (#35902) (#37340) 2026-06-11 16:14:30 +00:00
呆萌闷油瓶 c4a8d79be9 perf(api): reduce workflow startup latency for chatflow (#36773) 
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
 2026-06-11 07:05:35 +00:00
呆萌闷油瓶 9ac71329a4 fix(plugin): align plugin list endpoint counts with live endpoint state (#37179) 
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
 2026-06-10 08:11:11 +00:00
盐粒 Yanli ba9975a083 feat(dify-agent): sync shell and back proxy updates (#37159) 
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
 2026-06-10 03:04:32 +00:00
zyssyz123 a80bba2c35 feat(agent): Agent Files / agent Cloud storage — api backend (ENG-589) (#37172) 
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
 2026-06-09 04:01:05 +00:00
bymle 24876bb05d fix: avoid duplicating lines when merging text for summarization (#37093) 
Co-authored-by: bymle <229636660+bymle@users.noreply.github.com>
 2026-06-05 07:02:57 +00:00
Tim Ren 0ce0127e7e fix(security): reject path traversal sequences before plugin daemon forward (GHSA-gvc6-fh3x-89xh) (#35796) 
Co-authored-by: Ido Shani <ido@zafran.io>
Co-authored-by: -LAN- <laipz8200@outlook.com>
 2026-05-25 16:17:39 +00:00
-LAN- 8d99326fb3 feat(plugin): cache plugin model providers by tenant (#36449) 
Co-authored-by: WH-2099 <wh2099@pm.me>
 2026-05-23 09:12:09 +00:00
Sven Weidenmann 4d8b6c7dc0 refactor: add missing @override decorator to remaining MCP, Jieba, embeddings, and misc subclasses (#36528) 2026-05-22 13:45:35 +00:00
Sven Weidenmann 82b0a03f5a refactor: add missing @override decorator to PluginModelRuntime (#36493) 2026-05-21 19:40:40 +00:00
-LAN- 19476109da chore(api): upgrade graphon to v0.3.0 (#35469) 
Signed-off-by: -LAN- <laipz8200@outlook.com>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: WH-2099 <wh2099@pm.me>
 2026-05-09 07:30:03 +00:00
Tim Ren 5cf741895f fix(plugin): preserve multi-value HTTP response headers (#35726) 
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-04-30 08:09:56 +00:00
99 3e876e173a chore(api): adapt Graphon 0.2.2 upgrade (#35377) 
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
 2026-04-18 11:16:24 +00:00
Yunlu Wen 3193e8a712 chore: reorg imports (#35308) 
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
 2026-04-16 08:50:02 +00:00
wdeveloper16 7e7b27fdec refactor: replace bare dict with dict[str, Any] in response converter… (#35212) 
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-04-14 19:45:04 +00:00
dataCenter430 83b242be7b refactor: replace bare dict with typed annotations in core plugin module (#35096) 2026-04-13 19:23:21 +00:00
wdeveloper16 4e0273bb28 refactor: replace bare dict with dict[str, Any] in provider entities and plugin client (#35077) 2026-04-13 17:09:25 +00:00
corevibe555 488fcd4f83 refactor(services): replace Union with | syntax in service layer (batch 2) (#34906) 
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
 2026-04-10 07:05:31 +00:00
dataCenter430 0bdd1267fb refactor: convert appmode plugin if/elif to match/case (#30001) (#34790) 2026-04-09 02:28:03 +00:00
zhangbububu 4fb3fab82d fix: add backward-compatible query param for decode_plugin_from_ident… (#34720) 2026-04-08 13:28:37 +00:00
corevibe555 c825d5dcf6 refactor(api): tighten types for Tenant.custom_config_dict and MCPToolProvider.headers (#34698) 2026-04-08 01:36:42 +00:00
99 8f9dbf269e chore(api): align Python support with 3.12 (#34419) 
Co-authored-by: Asuka Minato <i@asukaminato.eu.org>
 2026-04-02 05:07:32 +00:00
wangxiaolei 57f358a96b perf: use global httpx client instead of per request create new one (#34311) 
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
 2026-04-01 01:19:32 +00:00