From b8de25ba8af98c452d9460e8fc238b06cdb1b261 Mon Sep 17 00:00:00 2001 From: Taitres <920643082@qq.com> Date: Sat, 1 Aug 2026 11:22:16 +0800 Subject: [PATCH 1/2] =?UTF-8?q?Update=20plugin=20PDF=E5=A4=84=E7=90=86=20v?= =?UTF-8?q?1.0.1?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit - Initial commit - chore: add pdf-lib dependency - feat: configure all PDF features in plugin.json - feat: add TypeScript types for PDF services - feat: add Sidebar component - feat: add FileUpload component - feat: add Settings component - feat: add feature routing in App.tsx - feat: add PDF processing services to preload - feat: add PDF compress feature - feat: add PDF merge feature - feat: add PDF split feature - feat: add PDF watermark feature - feat: add extract images feature - feat: add PDF to image feature - feat: add PDF to Word feature (mock) - feat: add PDF to PPT feature (mock) - feat: add PDF to Excel feature (mock) - feat: add PDF edit feature (placeholder) - chore: final build and cleanup - fix: 13 bugs - services APIs, React types, settings JSON, watermark centering - fix: use getPathForFile for all FileUpload features, embed CJK font for Chinese watermarks - fix: Chinese watermark with subset font, extractImages stream decode, compression quality slider, pdfToImage via system ImageMagick - chore: archive audit-and-repair - chore: fix archive metadata paths - fix: ensure output directory exists for compressPdf, mergePdfs, addWatermark - tweak: remove extractImages, fix splitPdf ranges, fix addWatermark syntax - chore: add .worktrees to gitignore - 1 - chore: remove accidentally tracked .worktrees and Windows residue - chore: gitignore .worktrees, .comet, %SystemDrive% to prevent future accidental tracking - docs: update ZTools developer documentation - tweak: 优化前端和操作逻辑，统一 UI 组件与交互 - docs: design doc for pure local PDF→Word/PPT/Excel conversion - fix: pdfcpu watermark 改为正确位置参数语法，CJK 字体自动检测 feat: 每个操作隔离到独立任务文件夹（useTaskFolder hook） - chore: archive local-pdf-to-word-mammoth change (branch deleted) - fix: pdfcpu 字体名映射 + 每次操作生成独立任务文件夹 - fix: 自动安装完整 CJK 字体到 pdfcpu，修复水印不可见 - docs: design doc for portable LibreOffice local PDF-to-Office conversion - docs: implementation plan for portable LibreOffice PDF-to-Office conversion - feat: update Services/SettingsData types for portable LO and remove cloud API fields - feat: add useLibreOffice hook for portable LO lifecycle - fix: add source fallback, unmount safety, useCallback wrappers to useLibreOffice - feat: add LONotInstalledBanner component for LO missing prompt - feat: add PdfToWord component with LO detection - feat: add PdfToPpt component with LO detection - feat: add PdfToExcel component with LO detection - feat: wire up PdfToWord/Ppt/Excel routes and onOpenSettings prop in App - feat: replace cloud API UI with LibreOffice management in Settings - test: update mock services and rewrite Settings tests for LO management UI - feat: implement downloadFile/cancelDownload/uninstallLibreOffice in services.js for portable LO - fix: update LO portable URL to 26.2.4 paf.exe, handle paf extraction via 7z/silent - fix: paf.exe extraction - use 7z with fallback, recursive soffice search, correct file extension - fix: setCustomPath now calls installLibreOffice, support directory paths for pre-extracted LO - fix: strip trailing slashes from path, call 7z directly via Node, better error logging per step - feat: bundle 7zr.exe in bin/, use as primary extraction method before 7z/Expand-Archive/silent - fix: replace 7zr.exe with 7za.exe (NSIS support) to extract paf.exe archives - fix: suppress LO first-run wizard — add SAL_USE_VCLPLUGIN=svp, pre-create profile, --nofirststartwizard, capture stdout - feat(ai): add DocumentSchema normalize and fixtures - fix(ai): drop out-of-scope convert scripts from Task 1 package.json - feat(ai): write Word via docx from DocumentSchema - feat(ai): write Excel and PPT from DocumentSchema via Node - feat(ai): extract per-page PDF text with pdfjs in preload - feat(ai): OpenAI and Anthropic compatible JSON completion client - feat(ai): wire convertPdfAi orchestration into preload services - feat(ai): settings UI for OpenAI/Anthropic conversion config - feat(ai): local/AI mode toggle on PDF convert pages - docs: document AI convert mode and setup - test: exclude preload AI tests from jsdom suite (use test:ai) - fix(build): restore publicDir so vite copies public/ into dist - fix(ai): lazy-load convert stack, cancel toast, unique sheets, batch test - refactor(convert): drop convert.exe; PDF to Office is AI-only - fix(ai): set pdfjs workerSrc for Electron preload (asar-safe) - fix(ai): cMap via fs factory + vision fallback for scan PDFs - feat: remove PDF edit; local convert fallback without API - v0.9 - v0.9.5 - v0.9.6 - v0.9.7 - v0.9.8 - v0.9.9 - fix(0.9.9): path-added files get preview, selection, and drop-to-add - v1.0.0 - fix(0.9.9): import files from ZTools super panel onPluginEnter - v1.0.0 preview - v1.0.0 review - Add preload dependency installation hooks - fix: install preload deps in ci - chore: bump plugin version to 1.0.1 --- plugins/pdf-process/public/plugin.json | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/plugins/pdf-process/public/plugin.json b/plugins/pdf-process/public/plugin.json index cc7c64c4..76a5f1c8 100644 --- a/plugins/pdf-process/public/plugin.json +++ b/plugins/pdf-process/public/plugin.json @@ -5,7 +5,7 @@ "description": "PDF 压缩、合并、拆分、水印、格式转换等工具", "author": "Taitres", "homepage": "https://github.com/Taitres/pdf-process", - "version": "1.0.0", + "version": "1.0.1", "main": "index.html", "preload": "preload/services.js", "logo": "logo.png", From c7f01b6f89377b34f3c3f4453560259d013f5fd1 Mon Sep 17 00:00:00 2001 From: Taitres <920643082@qq.com> Date: Sat, 1 Aug 2026 11:27:08 +0800 Subject: [PATCH 2/2] =?UTF-8?q?Update=20plugin=20PDF=E5=A4=84=E7=90=86=20v?= =?UTF-8?q?1.0.1?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit - Initial commit - chore: add pdf-lib dependency - feat: configure all PDF features in plugin.json - feat: add TypeScript types for PDF services - feat: add Sidebar component - feat: add FileUpload component - feat: add Settings component - feat: add feature routing in App.tsx - feat: add PDF processing services to preload - feat: add PDF compress feature - feat: add PDF merge feature - feat: add PDF split feature - feat: add PDF watermark feature - feat: add extract images feature - feat: add PDF to image feature - feat: add PDF to Word feature (mock) - feat: add PDF to PPT feature (mock) - feat: add PDF to Excel feature (mock) - feat: add PDF edit feature (placeholder) - chore: final build and cleanup - fix: 13 bugs - services APIs, React types, settings JSON, watermark centering - fix: use getPathForFile for all FileUpload features, embed CJK font for Chinese watermarks - fix: Chinese watermark with subset font, extractImages stream decode, compression quality slider, pdfToImage via system ImageMagick - chore: archive audit-and-repair - chore: fix archive metadata paths - fix: ensure output directory exists for compressPdf, mergePdfs, addWatermark - tweak: remove extractImages, fix splitPdf ranges, fix addWatermark syntax - chore: add .worktrees to gitignore - 1 - chore: remove accidentally tracked .worktrees and Windows residue - chore: gitignore .worktrees, .comet, %SystemDrive% to prevent future accidental tracking - docs: update ZTools developer documentation - tweak: 优化前端和操作逻辑，统一 UI 组件与交互 - docs: design doc for pure local PDF→Word/PPT/Excel conversion - fix: pdfcpu watermark 改为正确位置参数语法，CJK 字体自动检测 feat: 每个操作隔离到独立任务文件夹（useTaskFolder hook） - chore: archive local-pdf-to-word-mammoth change (branch deleted) - fix: pdfcpu 字体名映射 + 每次操作生成独立任务文件夹 - fix: 自动安装完整 CJK 字体到 pdfcpu，修复水印不可见 - docs: design doc for portable LibreOffice local PDF-to-Office conversion - docs: implementation plan for portable LibreOffice PDF-to-Office conversion - feat: update Services/SettingsData types for portable LO and remove cloud API fields - feat: add useLibreOffice hook for portable LO lifecycle - fix: add source fallback, unmount safety, useCallback wrappers to useLibreOffice - feat: add LONotInstalledBanner component for LO missing prompt - feat: add PdfToWord component with LO detection - feat: add PdfToPpt component with LO detection - feat: add PdfToExcel component with LO detection - feat: wire up PdfToWord/Ppt/Excel routes and onOpenSettings prop in App - feat: replace cloud API UI with LibreOffice management in Settings - test: update mock services and rewrite Settings tests for LO management UI - feat: implement downloadFile/cancelDownload/uninstallLibreOffice in services.js for portable LO - fix: update LO portable URL to 26.2.4 paf.exe, handle paf extraction via 7z/silent - fix: paf.exe extraction - use 7z with fallback, recursive soffice search, correct file extension - fix: setCustomPath now calls installLibreOffice, support directory paths for pre-extracted LO - fix: strip trailing slashes from path, call 7z directly via Node, better error logging per step - feat: bundle 7zr.exe in bin/, use as primary extraction method before 7z/Expand-Archive/silent - fix: replace 7zr.exe with 7za.exe (NSIS support) to extract paf.exe archives - fix: suppress LO first-run wizard — add SAL_USE_VCLPLUGIN=svp, pre-create profile, --nofirststartwizard, capture stdout - feat(ai): add DocumentSchema normalize and fixtures - fix(ai): drop out-of-scope convert scripts from Task 1 package.json - feat(ai): write Word via docx from DocumentSchema - feat(ai): write Excel and PPT from DocumentSchema via Node - feat(ai): extract per-page PDF text with pdfjs in preload - feat(ai): OpenAI and Anthropic compatible JSON completion client - feat(ai): wire convertPdfAi orchestration into preload services - feat(ai): settings UI for OpenAI/Anthropic conversion config - feat(ai): local/AI mode toggle on PDF convert pages - docs: document AI convert mode and setup - test: exclude preload AI tests from jsdom suite (use test:ai) - fix(build): restore publicDir so vite copies public/ into dist - fix(ai): lazy-load convert stack, cancel toast, unique sheets, batch test - refactor(convert): drop convert.exe; PDF to Office is AI-only - fix(ai): set pdfjs workerSrc for Electron preload (asar-safe) - fix(ai): cMap via fs factory + vision fallback for scan PDFs - feat: remove PDF edit; local convert fallback without API - v0.9 - v0.9.5 - v0.9.6 - v0.9.7 - v0.9.8 - v0.9.9 - fix(0.9.9): path-added files get preview, selection, and drop-to-add - v1.0.0 - fix(0.9.9): import files from ZTools super panel onPluginEnter - v1.0.0 preview - v1.0.0 review - Add preload dependency installation hooks - fix: install preload deps in ci - chore: bump plugin version to 1.0.1 - chore: 统一根 package.json 版本号为 1.0.1 --- plugins/pdf-process/package-lock.json | 4 ++-- plugins/pdf-process/package.json | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/plugins/pdf-process/package-lock.json b/plugins/pdf-process/package-lock.json index a1bed261..c688d6d8 100644 --- a/plugins/pdf-process/package-lock.json +++ b/plugins/pdf-process/package-lock.json @@ -1,12 +1,12 @@ { "name": "pdf-process", - "version": "0.9.9", + "version": "1.0.1", "lockfileVersion": 3, "requires": true, "packages": { "": { "name": "pdf-process", - "version": "0.9.9", + "version": "1.0.1", "hasInstallScript": true, "dependencies": { "jpeg-js": "^0.4.4", diff --git a/plugins/pdf-process/package.json b/plugins/pdf-process/package.json index 44c77623..88e9e8ca 100644 --- a/plugins/pdf-process/package.json +++ b/plugins/pdf-process/package.json @@ -1,6 +1,6 @@ { "name": "pdf-process", - "version": "0.9.9", + "version": "1.0.1", "description": "", "type": "module", "scripts": {