Commits

Commit history

1. Commits on May 9, 2026

   • feat: enhance visual asset tracking with enriched metadata, standardized source references, and improved manifest reporting.

     Jayleonc

     committed
     381d8d3View commit details

     Copy full SHA for 381d8d3

     Browse repository at this point
2. Commits on Apr 29, 2026

   • feat: implement XlsxProcessor and TextProcessor with enhanced file detection logic in workflow conversion pipeline

     Jayleonc

     committed
     3910966View commit details

     Copy full SHA for 3910966

     Browse repository at this point

   • feat: add /api/v1/convert-by-content endpoint for direct Markdown submission and normalization.

     Jayleonc

     committed
     d47e295View commit details

     Copy full SHA for d47e295

     Browse repository at this point
3. Commits on Nov 5, 2025

   • feat: add production deployment guide and configuration templates

     Show description for 88f128c
     Jayleonc

     committed
     88f128cView commit details

     Copy full SHA for 88f128c

     Browse repository at this point
4. Commits on Oct 11, 2025

   • Merge pull request #15 from Jayleonc/codex/standardize-and-clean-forma-python-project

     Show description for cc7f156
     Jayleonc

     authored
     cc7f156View commit details

     Copy full SHA for cc7f156

     Browse repository at this point

   • Expand OCR environment configuration coverage

     Jayleonc

     committed
     1001b98View commit details

     Copy full SHA for 1001b98

     Browse repository at this point

   • fix: 延长异步任务超时至10分钟，增加PDF文本提取失败时的备用解析逻辑

     Jayleonc

     committed
     5cf20b2View commit details

     Copy full SHA for 5cf20b2

     Browse repository at this point
5. Commits on Sep 22, 2025

   • Merge pull request #14 from Jayleonc/codex/create-deployment-plan-for-forma-python-service

     Show description for 397fecf
     Jayleonc

     authored
     397fecfView commit details

     Copy full SHA for 397fecf

     Browse repository at this point

   • Add Docker deployment tooling

     Jayleonc

     committed
     e58e68dView commit details

     Copy full SHA for e58e68d

     Browse repository at this point
6. Commits on Sep 18, 2025

   • refactor: 统一异步超时控制，增强worker管理，完善图像尺寸检查，优化Markdown清洗逻辑

     Jayleonc

     committed
     4552530View commit details

     Copy full SHA for 4552530

     Browse repository at this point
7. Commits on Sep 12, 2025

   • feat: add advanced OCR option and FAQ generation task with callback; add bold headers chunking tests

     Jayleonc

     committed
     275e3a5View commit details

     Copy full SHA for 275e3a5

     Browse repository at this point
8. Commits on Sep 11, 2025

   • feat: 新增Embedding配置；qa-v2移除全局整合，改为直接输出带分类的QA列表并优化层级提炼逻辑；chunk ID改为短Base64 UUID生成；更新README说明新流水线

     Jayleonc

     committed
     927b239View commit details

     Copy full SHA for 927b239

     Browse repository at this point

   • refactor: 重构层级知识提炼，支持并行处理与状态缓存优化

     Show description for 7564309
     Jayleonc

     committed
     7564309View commit details

     Copy full SHA for 7564309

     Browse repository at this point

   • Merge pull request #13 from Jayleonc/codex/implement-hierarchical-qa-generation-process

     Show description for fbe1368
     Jayleonc

     authored
     fbe1368View commit details

     Copy full SHA for fbe1368

     Browse repository at this point

   • feat: add hierarchical QA pipeline

     Jayleonc

     committed
     4b563b2View commit details

     Copy full SHA for 4b563b2

     Browse repository at this point

   • feat: 新增OCR配置类及高级OCR客户端，增强PDF/DOCX/PPTX处理器支持高级OCR和VLM重试

     Show description for c211f72
     Jayleonc

     committed
     c211f72View commit details

     Copy full SHA for c211f72

     Browse repository at this point
9. Commits on Sep 10, 2025

   • feat: 增加MarkdownCleaner用于清理Markdown并集成到转换流程和回调中，优化请求验证错误日志

     Jayleonc

     committed
     9f42cdeView commit details

     Copy full SHA for 9f42cde

     Browse repository at this point
10. Commits on Sep 8, 2025

    • refactor: 增加详尽日志，增强文件类型识别和错误处理，改进转换流程鲁棒性

      Jayleonc

      committed
      d2c30fcView commit details

      Copy full SHA for d2c30fc

      Browse repository at this point

    • Merge pull request #12 from Jayleonc/codex/refactor-forma-to-asynchronous-microservice

      Show description for 085e4f1
      Jayleonc

      authored
      085e4f1View commit details

      Copy full SHA for 085e4f1

      Browse repository at this point

    • feat: add async conversion service with callbacks

      Jayleonc

      committed
      48dd705View commit details

      Copy full SHA for 48dd705

      Browse repository at this point

    • -

      Jayleonc

      committed
      fa39edcView commit details

      Copy full SHA for fa39edc

      Browse repository at this point
11. Commits on Aug 27, 2025

    • feat: 集成 Marker 解析器，新增 PdfMarkerProcessor 并更新依赖和导入声明

      Jayleonc

      committed
      42a0027View commit details

      Copy full SHA for 42a0027

      Browse repository at this point

    • refactor: 优化PPTX复杂页处理，改用一次PDF转换多次渲染方案，提高性能和内容保真度

      Show description for b230cbc
      Jayleonc

      committed
      b230cbcView commit details

      Copy full SHA for b230cbc

      Browse repository at this point
12. Commits on Aug 26, 2025

    • refactor: 优化chunk策略，新增语义markdown分块，调整代码结构和注释

      Jayleonc

      committed
      42af258View commit details

      Copy full SHA for 42af258

      Browse repository at this point
13. Commits on Aug 22, 2025

    • refactor: 移除旧核心模块，重构文档转换、OCR、知识构建等功能至新包结构，完善模块接口和调用逻辑

      Jayleonc

      committed
      a4398ebView commit details

      Copy full SHA for a4398eb

      Browse repository at this point

    • refactor: 优化MarkdownChunker，支持多级标题递归分块和智能段落拆分；增强JSONL读取健壮性，跳过无效行；补充无用信息过滤规则

      Jayleonc

      committed
      e482396View commit details

      Copy full SHA for e482396

      Browse repository at this point
14. Commits on Aug 21, 2025

    • Merge pull request #11 from Jayleonc/codex/add-knowledge-base-command-for-batch-processing

      Show description for a7231bb
      Jayleonc

      authored
      a7231bbView commit details

      Copy full SHA for a7231bb

      Browse repository at this point

    • feat: add knowledge-base command

      Jayleonc

      committed
      c9a58a1View commit details

      Copy full SHA for c9a58a1

      Browse repository at this point

    • fix: 优化prompts.yaml提取原则顺序，调整配置读取环境变量及默认模型名称，添加LLM配置打印debug信息

      Jayleonc

      committed
      c694abeView commit details

      Copy full SHA for c694abe

      Browse repository at this point

    • docs: v1.1 优化知识分类提示，增强分类规则并规范输出格式

      Jayleonc

      committed
      d2ba140View commit details

      Copy full SHA for d2ba140

      Browse repository at this point

    • feat: 支持知识批量提炼，新增输出文件名参数，调整知识结构模型，优化知识构建流程与导出逻辑

      Jayleonc

      committed
      7e7b6fbView commit details

      Copy full SHA for 7e7b6fb

      Browse repository at this point

    • Merge pull request #10 from Jayleonc/codex/merge-phase-three-and-four-into-global-knowledge-synthesis

      Show description for 1d528f8
      Jayleonc

      authored
      1d528f8View commit details

      Copy full SHA for 1d528f8

      Browse repository at this point

    • feat: merge categorization and synthesis

      Jayleonc

      committed
      61f02a0View commit details

      Copy full SHA for 61f02a0

      Browse repository at this point

    • docs: 优化图像转录与知识提炼提示文案，提高准确性与规范性

      Jayleonc

      committed
      71ae43fView commit details

      Copy full SHA for 71ae43f

      Browse repository at this point

    • refactor: 优化技术图像分析提示语，增强PPTX复杂幻灯片解析的错误提示逻辑

      Jayleonc

      committed
      ffb0025View commit details

      Copy full SHA for ffb0025

      Browse repository at this point