Skip to content

ThinkReader 0.8.6

Choose a tag to compare

View all tags
@zhangzejing zhangzejing released this 15 Sep 15:05
· 8 commits to main since this release
v0.8.6
7b4a6f9

ThinkReader 0.8.6

  • 修复重复 /MediaBox 等可恢复 PDF 结构错误导致的导入失败，原 PDF 内容不变。
  • 文件树自动刷新外部删除和重命名；RAW PDF 改名后按内容恢复关联，解析使用临时输入副本。
  • RAW PDF 右键支持“重新导入”；成功、排队和解析中的文件置灰。
  • 随 App 提供 PDF.js 中文 CMap 与标准字体资源，修复部分中文 PDF 原文文字提取及批注定位基础缺失。

Validation / 验证

Implementation commit: dc52d49d5e7b3de9302dc3522aac165bdb99abeb.

65 backend test files: 400 passed, one opt-in real MinerU test skipped. Frontend build/regressions, Electron file-tree interactions, Chinese CMap regression, packaged resource hashes, frozen MCP UTF-8 and backend persistence smoke, and packaged Chinese Profile setup/restart passed.

A local eight-page Chinese PDF import and reader check passed without changing the original. Recovered Chinese source-text count increased from 2 to 7,412. This is not a measurement of MinerU OCR or Agent annotation accuracy. Real MinerU/Agent provider quality, clean-machine installation and installer upgrades were not tested. Windows binaries are unsigned.

旧论文已有的几何索引不会自动重建；若升级后旧论文仍有中文定位问题，需要通过 App 的相关服务重建派生索引，保留原 PDF、笔记和批注。

SHA-256

  • ThinkReader-Setup-0.8.6-win_x64.exe: C78E04750DDD90D9CEECA7A62201B890383CB9B0428C8D4C2D64C0D594952B70
  • ThinkReader-0.8.6-win_x64.exe: 7731117928350ABACC4F9E94CE0ADC71953DB6D0E95B8B78B9B51A96D861A2A5
  • ThinkReader-0.8.6-win_x64.zip: EEBDB38F41006F58157BB1B86697EC59CD85638345FFE037198EA7AFF4E2BDC9
Assets 5
Loading