fanhuadesenlinnn/daily_stock_analysis
1
0
Fork 0
mirror of https://github.com/ZhuLinsen/daily_stock_analysis synced 2026-09-20 10:53:33 +08:00
Code Issues Packages Projects Releases Wiki Activity
990 Commits 425 Branches 143 Tags
972c31465654d86c52c59abfdb8414b82808f50f
Commit Graph

45 Commits

Author SHA1 Message Date
Elvis Wang
972c314656 feat: Web/API 指数入口与共享 canonical 去重基础 (#2312) 
* feat: Web/API 指数入口与共享 canonical 去重基础

为 Issue #2303 Phase 2 PR1 落地 Web/API 指数入口适配：

- API 使用 parse_analysis_target 构造结构化 AnalysisTarget 并贯通到 pipeline
- API 与 TaskQueue 去重按 asset_type 分支（指数用 canonical_id，个股用 legacy code）
- BatchTaskAcceptedResponse 追加可选 rejected 字段，未登记 CSI 单股 400、批量仅该目标失败
- TaskInfo 固化 dedupe_key，避免指数与同码个股折叠及 _analyzing_stocks 残留
- Web 移除 assetType=index 全局过滤，Chat 名称识别保护指数 canonical
- 补齐 Pipeline 指数 DecisionSignal market_override=cn 真实分支测试

* fix: 收敛指数 canonical 身份与批量响应契约

PR #2312 review 修复：报告 meta 补充 asset_type 隐藏指数自选；/analyze 在解析前限制非空原始 token；is_single 统一驱动 metadata/409/单任务 202；HomePage 三元计数继续后续 chunk。

* fix: avoid double space in index news search query

* fix: preserve canonical index identity

* fix: validate legacy task asset type

* fix: preserve canonical index identity in chat

* fix: preserve index identity across chat backends
 2026-09-01 19:25:15 +08:00
青玉案
d0e66a1dc3 feat: 新闻检索为空时在报告中如实标注 (#2229) 
* feat: 新闻检索为空时在报告中如实标注

消息面章节此前是「有内容才渲染」，检索一条没拿到时整段直接消失，
读报告的人无从判断是确实没新闻，还是检索静默失败了（搜索源限流、
未配置可用渠道等）。这把「抓取失败」呈现成了「确实没有新闻」。

- src/analyzer.py: AnalysisResult 新增 news_result_count，默认 None
- src/core/pipeline.py: 把 Step 4 已算好的计数交给结果对象
  （此前只进了 diagnostic context snapshot，报告层拿不到）
- src/notification.py: news_lines 为空且计数为 0 时，渲染明确提示，
  并说明结论未纳入新闻维度证据
- tests: 新增 5 条用例，含两条负例——计数为 None 时不得报警
  （那是未配置搜索渠道，不是失败）、拿到新闻时行为与改动前一致

不触碰任何检索路径，纯展示层增量。

* fix: 把新闻缺失提示放进真实渲染路径，并独立于模型输出判定

按 review 三条意见修正：

P1-1 提示只存在于 generate_daily_report，而正常流程从不调用它——
_send_single_stock_notification 与聚合报告走的是 dashboard / brief /
single_stock。原实现对所有标准 REPORT_TYPE 都不生效。
改为抽出共享判定 _empty_news_disclosure，四个渲染器统一接入。

P2 检索零命中但模型按 schema 写出了 market_sentiment / hot_topics 时，
原 elif 分支被跳过，报告会展示模型生成的情绪判断却隐瞒无新闻证据。
改为独立判定 news_result_count == 0，与模型是否产出文字无关。

P1-2 补 docs/CHANGELOG.md [Unreleased] 条目，并在
docs/data-source-stability.md 的「用户可见提示建议」一节记录该行为，
含 None / 0 / >0 三态语义表。

测试从 5 条增至 10 条，新增覆盖 dashboard、brief、single_stock 三个真实
渲染器，以及「模型有输出但检索为空」这一最糟组合。39 passed

* fix: 把新闻零命中披露覆盖到模板链路与企业微信入口

按 review 指出的 blocker 修正。此前只接了字符串拼接分支，遗漏两类活路径：

1. REPORT_RENDERER_ENABLED=true 时，generate_dashboard_report /
   generate_brief_report / generate_wechat_dashboard 会先 return render(...)，
   模板链路一路不渲染披露；
2. generate_wechat_dashboard 的非模板 fallback 从未接入，而 pipeline 在
   企业微信非 brief 场景会直接调用它。

后果是同一份分析结果在部分渠道披露、在另一些渠道沉默。

改法不再逐点打补丁，而是抽出单一事实来源：

- 新增 src/services/empty_news.py 持有判定与中英文案
- src/notification.py 的 _empty_news_disclosure 改为委托该模块
- src/services/report_renderer.py 为每条结果预计算 empty_news_disclosure，
  三个平台模板共用
- templates/report_markdown.j2 / report_brief.j2 / report_wechat.j2 各加渲染分支
- generate_wechat_dashboard 的 fallback 正文接入披露

新增 6 条回归测试：模板链路三个平台各一条、企业微信入口一条，
外加两条负例（未执行检索时模板与企业微信均不得提示）。

本文件测试 10 → 16 全过；全量 5824 passed，9 个既有失败与本 PR 无关
（干净 main 上同样失败，属测试顺序依赖）。

* fix: 修正计数源头的两处缺口（自查发现）

按 review 的 merge-base..HEAD 方法自查全链路，发现此前几轮都只盯着渲染出口，
从未核对计数源头，而源头本身在两条路径上是错的：

1. src/core/pipeline.py: news_result_count 只在 intel_results 非空时赋值，
   搜索服务整体失败（正是所有搜索源限流全挂的场景）时停留在 None，
   语义为「未执行检索」，于是本 PR 想解决的头号场景反而不提示。
   改为检索一发起即置 0。

2. _analyze_with_agent: Agent 模式自行调用 search_stock_news 完成检索，
   却从不回写计数，该路径下零命中永远静默。改为按检索结果回写 0 或实际条数。

渲染层再周全，源头数据不对则全部落空。

新增 2 条测试锁住这两处语义（18 passed，此前 16）。
全量 5826 passed，9 个既有失败与本 PR 无关。

* fix: disclose missing news search configuration

* chore: remove unrelated agent guidance

* test: run all empty news tests directly

* fix: preserve empty news disclosure across reports

* fix: 让 Agent 模式的新闻披露跟随实际消费的证据

原问题：agent_arch=multi 等受支持的 Agent 配置下，报告可能声称「未纳入新闻
面证据」而分析其实用了新闻，或反过来该提示而不提示。

根因：news_result_count 取自 executor.run() 结束后为持久化情报补打的一次
search_stock_news()。真实情报由 IntelAgent 通过 search_comprehensive_intel
取得，两者不等价，因此披露与真实证据链可能相反。

修复点：新增 src/agent/news_evidence.py，以运行期证据作用域收集 Agent 搜索
工具的真实返回条数；搜索渠道不可用为 None（未执行检索），可用则从 0 起步、
拿到多少算多少。pipeline 在 executor.run() 前后开启并读取该作用域，事后的
持久化补查不再回写计数。

回归风险：工具在 ThreadPoolExecutor 中执行，runner.py 以
contextvars.copy_context() 提交，故作用域中必须是可变累加器对象，换成不可变
值会让父线程读不到；已加回归测试锁住该机制。原 test_agent_path_records_count
断言的正是被修复的错误行为，已替换为反向断言。

Refs #2225

* fix: 让新闻披露以实际证据为准而非搜索命中数

原问题：本地已落库的资讯池或社交情绪进入 news_context 参与分析后，报告仍可能
声称「未配置搜索渠道，本次分析未纳入新闻面证据」或「零命中」。

根因：news_context 由三路来源拼成——实时检索、社交情绪（美股）、本地资讯池，
但只有实时检索会更新 news_result_count。披露断言的是「结论有没有用到新闻面
证据」，而计数只是「搜索命中了几条」，两者是不同命题，后两路参与时必然失真。

修复点：AnalysisResult 新增 news_evidence_present，由 news_context 是否非空
得出，pipeline 两条路径共用 src/services/empty_news.news_evidence_present()
这一个判定函数。披露改为先看有无证据；确无证据时才用计数解释原因
（None=未配置渠道，0=检索零命中）。历史重建同步恢复该字段。

回归风险：旧记录没有该字段，按计数回退推断，与该记录当时的报告表现一致，不会
追溯改变旧报告；已有用例锁住。review 只点名了本地资讯池，社交情绪属同一缺陷类，
本次一并修复并加测试。另加源码断言：任一 pipeline 路径改回只传计数即失败。

Refs #2225

* fix: 按来源登记新闻证据，不让零命中占位文本冒充证据

原问题：普通分析链路在「搜索已执行但一条证据都没拿到」时，报告不再显示零命中
披露——正是本 PR 要修的核心场景，反而比改动前更差。

根因：src/search_service.py 的 format_intel_report() 即使所有维度失败或为空，
也会输出「【XX 情报搜索结果】」标题和每个维度的「未找到相关信息」占位文本，
整段永远非空。上一版把拼好的 news_context 整段交给 news_evidence_present()
判定，于是 news_result_count == 0 时 evidence 被翻成 true，披露被吞掉，错误
状态还会经 to_dict() 持久化，继续影响历史、详情 API 与 Web。

修复点：判定改为按来源逐个登记——实时检索的真实命中数、社交情绪内容、本地
资讯池内容，任一为真才算有证据；两条 pipeline 路径都不再传拼好的整段。
news_evidence_present() 的契约随之改为接收各来源，并在文档串里写明为什么不能
传整段。

回归风险：新增反例用真实的 format_intel_report() 产出占位文本（不用 mock），
断言其不得被判成证据、且报告必须出现零命中披露。另有源码断言：谁把整段
news_context 交回判定函数即失败。上一版两条测试实际在保护该缺陷（一条名为
「任何非空 context 都算证据」，一条要求必须传入 news_context），已一并纠正。

Refs #2225

---------

Co-authored-by: Mach-Chan <zz-b240@zz-b240deMacBook-Air.local>
 2026-08-22 21:27:04 +08:00
lyl2104626211
68fc575f15 feat: 支持 Web/API 按市场触发大盘复盘 (#2074) 
* feat: add request-scoped market review regions

* fix: keep server default runtime-resolved

* fix: trace market review regions through task lifecycle

---------

Co-authored-by: zhulinsen <42829555+ZhuLinsen@users.noreply.github.com>
 2026-07-25 22:14:15 +08:00
zhulinsen
6da1360a0b feat: 重建市场结构上下文 (#1981) 
* feat: rebuild market structure context

* fix(review-feedback-1981): apps/dsa-web/src/components/report/MarketStructureCard.tsx 新增用户可见界面，但

* fix(review-feedback-1981): apps/dsa-web/src/components/report/MarketStructureCard.tsx 新增用户可见界面，但

* fix(review-feedback-1981): Scope in-flight ranking keys to each fetcher manager

* fix(review-feedback-1981): 修正无证据情况下的个股层级判定，并按仓库规范补充可访问的 Web 页面截图、更新 PR 描述事实

* fix(review-feedback-1981): apps/dsa-web/src/components/report/MarketStructureCard.tsx 新增用户可见界面，但

* fix(review-feedback-1981): 按 AGENTS.md 为新增市场结构卡片补充可直接查看的截图，并同步 PR 描述中的 diff 与 CI 事实

* fix(review-feedback-1981): apps/dsa-web/src/components/report/MarketStructureCard.tsx 新增用户可见界面，但

* fix(review-feedback-1981): 按仓库规范补充可直接审查的市场结构卡片截图，并同步 PR 描述中的 diff 与 CI 事实

* fix(review-feedback-1981): apps/dsa-web/e2e/market-structure-card-visual.spec.ts:276 在

* fix(review-feedback-1981): apps/dsa-web/e2e/market-structure-card-visual.spec.ts 将本 PR

* fix(review-feedback-1981): 同步当前 Head 的 diff/CI 结论，并按仓库规范在 PR 描述或评论中附可访问的市场结构卡片截图

* fix(review-feedback-1981): apps/dsa-web/e2e/market-structure-card-visual.spec.ts:261 新增 async 空对象解构

* fix(review-feedback-1981): 按仓库规范在 PR 描述或评论中附市场结构卡片的可访问截图，并同步当前 diff 统计和 CI 结果

* fix(review-feedback-1981): 消除当前运行结果对历史持久化成功的依赖，补充对应回归测试，并补齐可访问的 UI 截图及更新失真的 PR 描述

* fix(review-feedback-1981): src/utils/data processing.py 提取榜单时丢弃了上游 boards / concept boards 的

* fix(review-feedback-1981): 修正 PR 描述并补齐可访问的视觉证据

* fix(review-feedback-1981): 关闭市场结构前置数据获取可能阻塞主分析的问题，并同步修正 PR 描述、范围清单和视觉证据

* fix(review-feedback-1981): 将 PR 描述、完整范围、当前 CI 证据、视觉证据及回滚表述同步到最新 Head

* fix(review-feedback-1981): PR 描述与最新 Head 不一致：完整 diff 包含 38 个文件及 src/services/analysis

* fix(review-feedback-1981): 按最新 Head 同步 PR 描述，并补充可访问的市场结构卡片截图证据

* fix(review-feedback-1981): 统一即时响应与历史响应的 details.raw result 契约并补最终 API 回归测试，同时更新 PR 描述和可访问的视觉证据

* fix(review-feedback-1981): 将 PR 描述、完整范围、CI 结果及可访问的 UI 截图同步到最新 Head，消除描述与实际改动的实质性不一致

* fix(review-feedback-1981): 同步 PR 描述与最新 Head，并补充可访问的 Web 页面截图

* fix(review-feedback-1981): 同步 PR 描述并按仓库规范附上可访问的 Web 页面截图

* fix(review-feedback-1981): 代码结论 ：不可。已按 origin/main...76b5b596fdff 的完整 38 文件 diff 重新建立基线。上次 6 个

* fix(review-feedback-1981): 代码结论 ：不可。最新修复已正确收窄无成分股、无龙头证据的原生榜单路径：该路径现在返回 edge/partial，不再直接输出
 2026-07-12 11:30:59 +08:00
zhulinsen
282efa18e1 fix: preserve zero sentiment score alignment (#1900) 
* fix: preserve zero sentiment score alignment

* fix(review-feedback-1900): 修正 docs/CHANGELOG.md 中与本 PR 无关的条目，避免把未包含在当前 diff 的功能/修复写入本 PR and
 2026-07-04 09:23:18 +08:00
zhulinsen
83b8154d0a fix: 评分建议过度归为持有，需收敛 hold/watch/buy/sell… (#1885) (#1887) 
* fix(issue-1885): [bug]-评分建议过度归为持有，需收敛-hold/watch/buy/sell

* fix(review-feedback-1887): 澄清/修正评分与 action 的契约一致性问题，并补齐高风险路径的验证说明

* fix(review-feedback-1887): 补一句排除说明，避免和 AGENTS 对外部依赖/运行时配置变更需说明兼容、迁移与回退路径的要求产生歧义

* fix(review-feedback-1887): CI 当前未通过：结构化事实显示 backend-gate:failure，PR 描述中的本地验证也写明 test:TIMEOUT
 2026-07-03 20:24:13 +08:00
zhulinsen
03fe7a00a9 feat: 补齐大盘报告与 Web 个股页的概念板块展示 (#1764) (#1778) 
* feat(issue-1764): feat:-补齐大盘报告与-web-个股页的概念板块展示

* fix(review-feedback-1778): 确认历史报告 fallback 路径是否完整透传 conceptRankings，否则会出现新字段在部分历史详情/API 场景中静默丢失

* fix: render market review industry and concept rankings side-by-side

* fix: refine related board display and concept ranking cache
 2026-06-27 23:07:18 +08:00
zhulinsen
ef9371003c feat: harden JP/KR suffix market support (#1817) 
* feat: harden JP/KR suffix market support

* fix: define yfinance index quote metadata

* fix(review-feedback-1817): Handle JP 4-digit bare-code lookup in analysis paths

* fix(review-feedback-1817): Route 4-digit JP bases through API normalization

* fix(review-feedback-1817): 补齐外部 API/provider/运行时配置风险说明或证明该检测为误报
 2026-06-27 22:10:08 +08:00
mumu
0ab94a8c3e fix: support KR stock autocomplete suggestions (#1723) 
* fix: support KR stock autocomplete suggestions

* fix(review-feedback-1723): 补充自动补全 JP/KR 建议渲染截图、前后对比或无法截图说明

* fix(review-feedback-1723): 补充 Web 自动补全 JP/KR 建议渲染的截图或无法截图说明

* fix(review-feedback-1723): docs/review-evidence-kr-autocomplete-ui.md 是针对本 PR

* fix(review-feedback-1723): 修正 PR 描述与当前 head 的实际改动、验证范围不一致问题

* fix(review-feedback-1723): 修正 PR 描述与当前 diff 的实质性不一致，避免合入记录误导后续维护者

* fix(review-feedback-1723): 收敛 PR 描述与实际 diff 的契约不一致，避免用户误以为裸 005930 已会解析为 005930.KS

* fix(review-feedback-1723): 把描述、验证证据和兼容性说明收敛到当前 6 个改动文件真实范围，或补齐描述中声称的后端/历史链路实现与测试

* fix(review-feedback-1723): 收敛 PR 描述与实际 diff，避免把未实现的 bare-code/历史展示能力作为本 PR 已交付内容合入

* fix(review-feedback-1723): 补上相应真实代码与测试后再审

* fix(review-feedback-1723): 补齐所声称的后端实现与测试后再评审

* fix(review-feedback-1723): PR description 与当前完整 diff 不一致：描述声称新增 bare JP/KR code

* fix(review-feedback-1723): PR 描述与当前完整 diff 不一致

* fix(review-feedback-1723): 补充 Refs 1718，方便后续追踪

* fix(review-feedback-1723): 收敛描述与验证证据

* fix: resolve JP KR bare stock codes

* fix: backfill JP KR daily analysis context

* fix(review-feedback-1723): preserve persisted phase timing

* fix: align JP KR history lookup keys

* fix(review-feedback-1723): preserve raw code for snapshot lookups

* fix(review-feedback-1723): 收敛
 2026-06-20 10:39:40 +08:00
LouisHong
145ee6b5a5 fix: [issue #1652] stabilize run-flow live states and layout polish (#1682) 
* fix: polish run-flow topology layout and i18n

* fix: prevent task panel layout squeeze

* test: update run flow frontend coverage

* fix: isolate market context run flow

* fix: handle completed task flow refresh

* feat: show running run-flow nodes

* fix: show chip provider runs in flow

* fix: stabilize stock run flow live updates

* fix: refine run-flow history lookup and spacing

* fix: update run-flow edge states and layout gaps

* fix: polish run-flow task panel text
 2026-06-14 17:57:16 +08:00
LouisHong
efa41e0ada feat: [issue #1652] add live run-flow diagnostics (#1656) 
* feat: add live run-flow updates and diagnostics

* feat: update task status filter to include 'cancel_requested'

* feat: enhance run-flow diagnostics and task management with cancel-requested status handling

* fix: stabilize run flow live updates

* fix: align live run-flow review contracts

* fix: sanitize live run-flow events
 2026-06-12 20:58:58 +08:00
LouisHong
fa9d9b0084 feat: [issue #1652] add run-flow snapshot contract and APIs (#1653) 
* feat: add run-flow snapshot contract and APIs

* fix: address run-flow contract review notes

---------

Co-authored-by: mumu <42829555+ZhuLinsen@users.noreply.github.com>
 2026-06-10 22:50:35 +08:00
LouisHong
4f536026f8 fix: clarify runtime logging context (#1643) 
* fix: clarify market review logging context

* docs: update changelog for logging context
 2026-06-09 01:08:55 +08:00
Alfred
ff6494b5d7 feat: 新增建议动作 taxonomy 字段边界 P0 (#1631) 
* feat: add decision action taxonomy fields

* fix: tighten decision action fallback handling

* fix: align legacy decision action guard fallback

* fix: tighten legacy decision action word boundaries

* fix: align decision action compound parsing

* fix: clean decision action residue

* fix: align web action label contract

* fix: classify avoided sell actions as hold
 2026-06-08 07:24:07 +08:00
mumu
8909168a64 fix: run manual market review on non-trading days (#1627) 2026-06-07 09:39:00 +08:00
mumu
d704fe377d feat: add Web UI language switch (#1621) 
* feat: add Web UI language switch

* fix(review-feedback-1621): 补充 zh/en 切换后的登录页、首页、设置页和侧边导航截图或可访问构建预览证据

* fix(review-feedback-1621): 补对应回归测试

* fix(review-feedback-1621): Pin the smoke test's initial language

* fix(review-feedback-1621): Web UI 改动缺少截图或等价可视证据

* fix(review-feedback-1621): 补一个 storage accessor 抛错时 fallback 到 zh 或 navigator 语言的回归测试

* fix(review-feedback-1621): 修复前端 lint/build 失败并补齐 Web 验证证据

* fix(review-feedback-1621): apps/dsa-web/ ：本 PR 新增语言切换入口并改动

* fix(review-feedback-1621): 解决冲突

* fix(review-feedback-1621): 解决冲突，并补齐 Web UI 变更的可视证据与 Web 侧验证说明

* fix(review-feedback-1621): 解决冲突并重新确认 diff 与 CI 结果

* fix(review-feedback-1621): 解决冲突

* fix(review-feedback-1621): add a file-level test

* fix: complete Web UI language switch evidence

* fix(review-feedback-1621): 确认设置页英文字段 label 语义回归、Playwright smoke locale 稳定性，以及结构化检测命中的外部模型/API

* fix(review-feedback-1621): 修正 PR 描述与实际实现不一致，并补齐 Playwright smoke 验证结果或明确未执行原因

* fix: complete web language behavior

* fix(review-feedback-1621): src/services/task queue.py / src/services/analysis service.py /

* fix: honor market review request language

* fix(review-feedback-1621): Keep UI language from overriding report language and Use localized

* fix(review-feedback-1621): docs/full-guide.md / docs/full-guide EN.md 把 PR 评审证据、执行命令、沙箱

* fix(review-feedback-1621): 修正 PR 描述与当前实现不一致的问题，并补齐运行时模型配置相关的兼容性/迁移证据

* fix(review-feedback-1621): 收敛后再合入
 2026-06-06 21:45:20 +08:00
fzxiong
39f24df60b 代码健康修复：稳定测试、健康检查、错误响应与 schema 版本记录 (#1607) 
* fix(api): return json from root health endpoint

* test(web): stabilize vitest runtime setup

* chore(security): warn on public web exposure without auth

* refactor(api): centralize error response helpers

* chore(db): record baseline schema version

* refactor(web): extract portfolio formatting helpers

* fix(db): make schema baseline initialization idempotent

* docs: document health check and schema baseline changes
 2026-06-05 21:26:06 +08:00
mumu
b612fb23e4 fix: 结构化大盘复盘报告展示 (#1576) 
* feat: add dedicated market review report view

* fix(review-feedback-1576): Use localized labels for English market reviews

* fix: structure market review payload rendering

* fix(review-feedback-1576): preserve market labels when rendering combined reviews and Use the

* fix(review-feedback-1576): Render structured data for every combined market and Localize

* fix: dedupe market review rendering

* test: align market review language prompt expectation

* fix(review-feedback-1576): 澄清并补足 LLM/运行时配置相关检测命中的兼容性证据

* fix(review-feedback-1576): 解决冲突并重新确认 CI

* fix(review-feedback-1576): docs/CHANGELOG.md：[Unreleased] 新增内容中除 1555 大盘复盘展示外，还包含 Agent

* fix(review-feedback-1576): 修复

* fix(review-feedback-1576): 修复美股中文大盘复盘仍混入英文策略/市场标签的问题，并补齐对应回归测试
 2026-06-04 13:40:42 +08:00
Alfred
7e54c6c3ba feat: add analysis phase API plumbing (#1573) 2026-06-03 19:51:07 +08:00
Alfred
f42b395745 feat: add market phase summary metadata (#1532) 2026-05-31 09:24:58 +08:00
Alfred
0295f9d709 feat: expose analysis context pack overview (#1515) 
Refs #1389
 2026-05-30 09:20:11 +08:00
mumu
8a23124d01 feat: add #1391 Phase 2 run diagnostic summaries (#1444) 
* fix(issue-1412): [bug]-stock_list格式问题

* fix(review-feedback-1413): preserve exchange hint for dotted A-share inputs

* fix(review-feedback-1413): Keep normalized A-share codes usable by market routing and preserve

* fix(review-feedback-1413): Limit raw dotted codes to fetchers that can parse them

* fix(review-feedback-1413): Keep Tushare daily input normalized for ETF detection

* fix(review-feedback-1413): 澄清结构化检测中的外部模型/API 与运行时配置迁移风险

* fix(review-feedback-1413): 处理或明确确认该失败与本 PR 无关且已有维护者豁免依据

* fix: keep stock list input as bare codes

* docs: add phase-0 run diagnostics contract

* fix(review-feedback-1435): 修正描述并澄清/补齐运行时代码变更的验证证据

* fix(review-feedback-1435): 补齐前缀提示识别，并增加对应回归测试

* fix(review-feedback-1435): 修正

* fix(review-feedback-1435): 解决冲突并更新描述/验证记录后再合入

* fix(review-feedback-1435): 解决冲突，并在最终 head 上重新确认 python -m pytest tests/test a share fetcher code

* fix(review-feedback-1435): 修复并补齐回归覆盖后再复核最终 head

* fix(review-feedback-1435): data provider/baostock fetcher.py 的 convert stock code 只从 .SH/.SS/.SZ

* fix: preserve A-share exchange hints

* fix(review-feedback-1435): 修正 docs/run-diagnostics-p0.md 对 Tushare 本轮范围的矛盾描述

* feat: add phase 1 run diagnostics trace plumbing

* feat: add phase 2 run diagnostic summaries

* fix(review-feedback-1441): 打通 trace id 与数据源运行快照，改动目标明确

* fix(review-feedback-1444): 修复 Agent 模式新报告通过历史诊断 API 返回 unknown 的正确性问题

* fix(review-feedback-1444): Propagate diagnostics lookup errors instead of masking them

* fix(review-feedback-1444): 补对应回归断言

* fix(review-feedback-1444): 解决冲突后再合入

* fix(review-feedback-1444): 解决冲突

* fix(review-feedback-1444): 解决冲突

* fix(review-feedback-1444): 解决

* fix(review-feedback-1444): 解决

* fix(review-feedback-1444): 解决冲突

* fix(review-feedback-1444): 解决

* fix(review-feedback-1444): 解决

* fix(review-feedback-1444): 解决冲突后再合入

* fix(review-feedback-1444): 解决冲突后再合入

* fix(review-feedback-1444): 解决冲突后再合入

* fix(review-feedback-1444): 解决冲突

* fix(review-feedback-1444): Derive news diagnostics from retrieval evidence

* fix(review-feedback-1444): 解决冲突后再合入

* fix(review-feedback-1444): 解决冲突并基于解决后的最终 diff 重新确认 docs/CHANGELOG.md、诊断链路和测试结果

* fix(review-feedback-1444): 解决冲突

* fix(review-feedback-1444): 解决冲突后再合入

* fix(review-feedback-1444): 解决冲突并重新跑阻断型 CI，尤其是 backend-gate 和相关诊断/API/history 回归

* fix(review-feedback-1444): preserve report timestamp when enriching task results

* fix: address run diagnostics review feedback

* fix: redact diagnostic copy text secrets

* fix(review-feedback-1444): 补一条多渠道部分失败的回归测试
 2026-05-26 00:05:24 +08:00
mumu
e0e5fa6870 feat: add #1391 Phase 1 run diagnostics trace plumbing (#1441) 
* fix(issue-1412): [bug]-stock_list格式问题

* fix(review-feedback-1413): preserve exchange hint for dotted A-share inputs

* fix(review-feedback-1413): Keep normalized A-share codes usable by market routing and preserve

* fix(review-feedback-1413): Limit raw dotted codes to fetchers that can parse them

* fix(review-feedback-1413): Keep Tushare daily input normalized for ETF detection

* fix(review-feedback-1413): 澄清结构化检测中的外部模型/API 与运行时配置迁移风险

* fix(review-feedback-1413): 处理或明确确认该失败与本 PR 无关且已有维护者豁免依据

* fix: keep stock list input as bare codes

* docs: add phase-0 run diagnostics contract

* fix(review-feedback-1435): 修正描述并澄清/补齐运行时代码变更的验证证据

* fix(review-feedback-1435): 补齐前缀提示识别，并增加对应回归测试

* fix(review-feedback-1435): 修正

* fix(review-feedback-1435): 解决冲突并更新描述/验证记录后再合入

* fix(review-feedback-1435): 解决冲突，并在最终 head 上重新确认 python -m pytest tests/test a share fetcher code

* fix(review-feedback-1435): 修复并补齐回归覆盖后再复核最终 head

* fix(review-feedback-1435): data provider/baostock fetcher.py 的 convert stock code 只从 .SH/.SS/.SZ

* fix: preserve A-share exchange hints

* fix(review-feedback-1435): 修正 docs/run-diagnostics-p0.md 对 Tushare 本轮范围的矛盾描述

* feat: add phase 1 run diagnostics trace plumbing

* fix(review-feedback-1441): 打通 trace id 与数据源运行快照，改动目标明确
 2026-05-25 21:01:40 +08:00
zbl-96
639796f982 fix: restore board linkage for compatible history snapshots (#1416) 
* fix: restore board linkage from compatible snapshots

* chore: drop local review artifact from pr

* fix: enrich in-memory status board details

* fix: merge partial fundamental snapshots

* fix: preserve fallback fields on empty snapshots

---------

Co-authored-by: ZhuLinsen <zhuls97@163.com>
 2026-05-24 19:20:25 +08:00
mumu
1f58b63936 fix(issue-1382): [bug]-/api/v1/analysis/status/{task_id} (#1385) 2026-05-21 22:29:40 +08:00
mumu
0a987086d3 feat: Web 个股分析支持选择策略 (#1331) 
* feat: support strategy selection in web analysis

* fix(review-feedback-1331): The worker receives the original skills list, while TaskInfo stores a

* fix(review-feedback-1331): 补充对应 Web/API 文档，或在 PR 中明确说明现有哪份文档已覆盖且本次无需更新的依据
 2026-05-17 22:59:49 +08:00
mumu
fb98a62938 fix: 大盘分析的历史执行记录丢失 (#1306) (#1307) 
* fix(issue-1306): 大盘分析的历史执行记录丢失

* fix(review-feedback-1307): also special-case the history detail/client path to surface the saved

* fix(review-feedback-1307): 补齐大盘复盘历史记录的 Web/API 契约闭环与对应文档落点

* fix(review-feedback-1307): Disable stock rerun for market-review history rows

* fix(review-feedback-1307): Disable follow-up for market-review history rows

* fix(review-feedback-1307): 避免大盘复盘历史被整体回测当作普通股票分析记录处理

* fix(review-feedback-1307): 澄清结构化检测提示中的外部模型/API 与运行时配置迁移风险是否为误报，或补齐对应兼容性证据

* fix(review-feedback-1307): 修复普通个股历史不回归问题
 2026-05-16 13:36:17 +08:00
mumu
d0bb619f5c fix: normalize report strategy price fields (#1264) 2026-05-11 23:14:45 +08:00
mumu
c325a14b8f Add web market review trigger (#1242) 
* feat: add web market review trigger
 2026-05-10 15:05:31 +08:00
mumu
04933583a9 fix: inject REPORT_LANGUAGE into workflow and add current_price to status API (#1013, #983) (#1028) 
* fix: inject REPORT_LANGUAGE into workflow and add current_price to status API (#1013, #983)
* fix: preserve status API snapshot change pct fallbacks
* fix: keep status change pct scoped to intraday data
 2026-04-10 20:36:28 +08:00
mumu
e8862d5331 feat: 让普通股票分析链路在尽量不破坏现有接口契约的前提下支持 LiteLLM 流式输出… (#865) (#924) 2026-04-04 14:16:51 +08:00
mumu
5d23998a48 fix: 在 api/v1/endpoints/analysis.py 的 SSE 生成器 except… (#967) (#971) 
* fix(issue-967): [bug]-sse-任务流断开时-cancellederror-被静默吞掉，缺少
 2026-04-03 22:11:35 +08:00
mumu
7f201f3b27 fix: restore #648/#649 agent capabilities on main (#822) 
* feat: bot commands & async dispatch — /ask, /chat, /history, /strategies, NL routing (#648)

* feat: bot commands & async dispatch — /ask, /chat, /history, /strategies, NL routing

- CommandDispatcher with dispatch_async() and NL routing
- /ask multi-stock analysis (parallel execution, portfolio overlay)
- /chat free-form conversation with session persistence
- /history per-user session isolation (colon delimiter format)
- /strategies with category grouping and activation status
- Two-layer NL routing: regex pre-filter → LLM intent parsing
- Async webhook handler for FastAPI integration
- Feishu stream: capped ThreadPoolExecutor replaces unbounded threads
- DingTalk stream: direct async handler awaiting

* fix: address PR #648 review comments

- ask.py: simplify validate_args error message, add _merge_code_args() for comma+space separated codes
- chat.py: simplify validate_args error message
- feishu_stream.py: shutdown ThreadPoolExecutor in stop()
- test_ask_command.py, test_bot_dispatcher_async.py: replace sys.modules litellm hack with litellm_stub

* fix(bot): preserve ask strategy args and serialize Feishu replies

* fix(strategies): remove activate() side-effect from /strategies command

/strategies is a read-only listing command but was calling
sm.activate() which mutates the skill manager state.  Replace with
a config-based set lookup (configured_active) to derive ✅/⬜
status without any write operations.  This makes the command
purely read-only and eliminates the risk of altering subsequent
/ask or /chat agent skill sets.

(cherry picked from commit a472b824a8)

* feat: deep research agent & event monitor — /research command, alert rules, scheduler integration (#649)

* feat: deep research agent & event monitor — /research command, alert rules, scheduler integration

- ResearchAgent with 3-phase approach (decompose → research → synthesise)
- Token budget tracking for deep research queries
- /research bot command with aliases (/深研, /deepsearch)
- REST endpoint POST /api/v1/agent/research
- EventMonitor with PriceAlert/VolumeAlert rules
- Scheduler background task system for periodic alert polling
- EventMonitor config validation in system_config_service
- main.py: EventMonitor wiring into schedule mode
- Unsupported rule type rejection at config and runtime level

* fix: address PR #649 review comments

- events.py: fix to_dict_list docstring, validate created_at as float, raise ValueError on non-dict rules
- research.py: remove unused import uuid
- agent.py: replace nested ThreadPoolExecutor with asyncio.wait_for, use Field(default_factory=list), pass stock_code context
- scheduler.py: clamp minimum interval_seconds to 30s
- test_multi_agent.py: reduce sleep 3s→0.05s and timeout 1s→0.01s to speed up tests

* fix: normalize price alert direction and tighten research ticker regex

- events.py from_dict_list: .lower() on deserialized direction so
  _check_price matches correctly (fixes silent missed alerts for
  mixed-case config like {"direction":"Above"})
- research.py: restrict US ticker auto-detection to 1-4 chars
  (was 1-5); avoids misclassifying common words (MACRO, TREND)
  as stock codes in /research topic queries

(cherry picked from commit 28126db5bd)

* fix: reconcile recovered agent features with current main

* fix: address review feedback and backend gate failures

* fix: address remaining review feedback

* fix: restore bot agent gating and research timeout handling
 2026-03-27 21:29:50 +08:00
mumu
929521aefc fix: complete trading philosophy injection across analyzer paths (#810) 
* fix: complete trading philosophy injection across analyzer paths

* fix: preserve legacy prompts for default bull trend

* fix: tighten legacy prompt fallback rules

* fix: scope legacy prompt fallback to builtin bull trend

* fix: align notify flow with async analysis contract

* fix: expose notify controls to home dashboard hook
 2026-03-23 23:17:14 +08:00
smallwhite
3af9f60a99 feat: add notify toggle for push notifications in WebUI (#809) 
- Add notify field to AnalyzeRequest schema (default: true)
- Pass notify as execution param through task_queue to analysis_service
- Add notify checkbox next to analyze button in HomePage
- Add notify state and setter to stockPoolStore
 2026-03-23 22:05:05 +08:00
Alfred
e1dbee3318 feat: 在个股分析页展示个股所属板块及其与当日板块强弱的关系 (#669) (#800) 
* feat: show related boards in stock reports (#669)

* fix: preserve unavailable related board states

---------

Co-authored-by: mumu <42829555+ZhuLinsen@users.noreply.github.com>
 2026-03-22 20:52:40 +08:00
LouisHong
73d524a186 #730 [PR1] 首页股票分析输入框自动补全 AutoComplete (#775) 
* feat(autocomplete): rebuild web autocomplete mvp

* feat(autocomplete): add error boundary and runtime fallback

Add error boundary and runtime error handling to improve autocomplete stability:

- Add StockAutocompleteBoundary class component to catch render errors
- Add FallbackInput component for graceful degradation
- Add runtimeFallback state and error handling in useAutocomplete hook
- Wrap search logic with try-catch to prevent crashes
- Add comprehensive tests for error scenarios

This ensures the autocomplete degrades to a plain input when:
- Index loading fails (existing behavior)
- Runtime search throws (new)
- Component tree throws during render (new)

* feat(autocomplete): add input validation and optimize name resolution

- Add early rejection for obviously invalid mixed alphanumeric input
- Skip expensive AkShare/fuzzy fallback for non-CJK free text
- Enhance frontend validation with isObviouslyInvalidStockQuery
- Increase minimum query length from 1 to 2 characters
- Add comprehensive test coverage for validation logic
- Update E2E test placeholder text to reflect new capabilities

These changes address code review feedback about:
- Preventing abuse from invalid input patterns
- Improving performance by avoiding unnecessary network calls
- Ensuring consistent validation between frontend and backend

* fix(autocomplete): harden input validation and runtime guards

* update doc

* style(autocomplete): adjust suggestion item hover effect

- Reduce hover background opacity from 35% to 25% for lighter visual effect
- Brighten hover color in dark theme from #257280 to #2dcae6

* fix(autocomplete): do not auto-highlight first suggestion to preserve raw input on Enter

* test(home): align report fixture with report language metadata

* fix: unify task queue stock dedupe key
 2026-03-20 11:35:00 +08:00
mumu
ef328e906b feat: add configurable report language (#764) 
* feat: add configurable report language

* Fix review follow-ups for history and fallback localization

* fix: prefer .env report language at startup

* fix: address report language review feedback

* fix: address latest report language review feedback
 2026-03-19 22:55:17 +08:00
Alfred
274720234a fix: add A-share financial/dividend context outputs (#710) (#731) 
Co-authored-by: mumu <42829555+ZhuLinsen@users.noreply.github.com>
 2026-03-17 21:38:31 +08:00
mumu
d42952ef5e Improve analysis API async contracts and startup helpers (#656) 
* Improve analysis API async contracts and startup helpers

* Fix rebased analysis API contract regressions

* fix: align frontend report type union with full responses
 2026-03-15 20:54:04 +08:00
Alfred
c85423dfca [feat] 在分析报告中标记使用模型并完善展示链路 Issue #528 (#534) 
* feat: mark model used in reports and notifications (#528)

* refactor: centralize model_used normalization and raw_result parsing (#528)

---------

Co-authored-by: mumu <42829555+ZhuLinsen@users.noreply.github.com>
 2026-03-07 20:29:02 +08:00
Myrrolinz
026630adb4 fix: Date inconsistency & incorrect query id for db (fixes #322) (#363) 
* resolve empty analysis result list problem by endDate + 1

* fix bug 322 by using record id for db query

* address comments

* address potential issue and error
 2026-02-26 12:22:24 +08:00
Alfred
0780cfbd7d fix: BOT 与 WEB UI 股票代码大小写统一 (fixes #355) (#356) 
- 新增 canonical_stock_code()，统一入口处规范化
- BOT analyze: lower() -> strip().upper()
- Config/API/task_queue/main: 股票代码解析时应用规范化
- 历史记录与任务去重可正确识别同一股票
 2026-02-21 17:58:48 +08:00
mumu
8f30f9093d fix: report signal accuracy, data source reliability, and market revi… (#275) 
* fix: report signal accuracy, data source reliability, and market review quality #patch

- Fix signal emoji mismatch for compound advice and *ST markdown escaping
- Inject structured data (indices/stats/sectors) into market review
- Fix tushare return type and API endpoint; auto-inject realtime priority
- Add search cache (500-cap FIFO), limit supplement requests to 1
- Fix analysis API report=None and ReportStrategy type inconsistency
- Update CHANGELOG for v3.0.1 ~ v3.0.5

* fix:  F821 undefined name 'UnifiedRealtimeQuote'
 2026-02-08 18:21:33 +08:00
Krane
9847157980 Feature/React web support 新的WebUI (#256) 
* feat(web): add FastAPI & React web connect

* feat: add FastAPI package

* feat: system base struct

* feat: frontend base struct

* feat: 基础分析接口，历史记录接口

* fix: 使用正确的 FastAPI 方法定义，避免线程阻塞

* fix: 对接历史报告页、详情页

* fix: 修复接口问题

* fix: 网络配置 允许公网访问 跨域配置

* fix: 页面打包 & 提供 Server 静态访问

* fix: 优化dock栏样式

* fix: 优化图标样式

* fix: 修改部分配色

* fix: 删除垃圾文档

* fix: 删除垃圾代码

* fix: 驼峰转换工具

* fix: 历史记录列表滚动加载（分页）

* feat: 显示分析中任务

* fix: 调整布局

* fix: 优化 Market Sentiment 组件动效

* fix: 历史列表组件bug

* fix: FastAPI 日志配置

* feat: 新闻历史接口

* feat: 资讯列表

* feat: 优化布局

* fix: 修复页面元素变宽问题

* fix: 中文标题

* fix: 任务列表状态显示问题

* fix: 抽取日志配置

* fix: 优化报告价格显示

* fix: 修复编译错误

* fix: FastAPI 启动提取到 main.py

* fix: 补充新web-ui启动文档

* fix: 默认发送通知

* fix: FastAPI 模式适配 docker

* fix: FastAPI 模式文档

* Update api/v1/endpoints/stocks.py

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* fix: package-lock.json

* fix: 修改错别字

* fix: 更新文档说明

* fix: 更新文档说明

---------

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
 2026-02-05 20:56:38 +08:00