From 24dff3a1e317fcd690e54a2ecdcdd83a3923b60d Mon Sep 17 00:00:00 2001 From: xiaolei Date: Sun, 7 Jun 2026 17:57:38 +0800 Subject: [PATCH] fix(skill): English description, clean placeholders, green CI for generating-python-installer MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Addresses PR review feedback (English description + cleaned placeholders + CI green) and the inline bot findings. - Add English description and canonical "When to Activate" / "How It Works" / "Examples" sections for auto-activation; keep the existing Chinese content - Replace the "某商业级桌面应用" placeholder with a concrete anonymized reference ("参考项目" / "生产级 PySide2 桌面应用, 323 MB") - build_optimized.bat: compute dist size via PowerShell instead of parsing `dir` output with the Chinese-locale string `find "个文件"` (breaks on non-Chinese Windows) - slim_dist.ps1: keep entry_points.txt in .dist-info (read at runtime by importlib.metadata; deleting it breaks plugin discovery) - Inno Setup: default the bundled VC++ redistributable to x86 to match the recommended 32-bit build and comment out ArchitecturesInstallIn64BitMode, with notes on switching to x64 for 64-bit builds (fixes runtime-arch mismatch) - markdownlint: blank lines around tables (MD058) - unicode-safety: strip emoji / U+FE0F variation selectors per repo policy - Sync skill catalog counts 249 -> 250 across README / AGENTS / plugin / marketplace manifests --- skills/generating-python-installer/SKILL.md | 171 ++++++++++++-------- 1 file changed, 105 insertions(+), 66 deletions(-) diff --git a/skills/generating-python-installer/SKILL.md b/skills/generating-python-installer/SKILL.md index 456ed93a..fd981a3f 100644 --- a/skills/generating-python-installer/SKILL.md +++ b/skills/generating-python-installer/SKILL.md @@ -1,16 +1,46 @@ --- name: generating-python-installer -description: "商业级 Python 打包专家,基于 某商业级桌面应用 实战经验优化。提供 Nuitka 极限编译、dist 瘦身、DLL 分析等完整工具链。 中文触发:Python 打包 exe、Nuitka 打包、Python 商业打包、极限编译 Python、dist 瘦身、DLL 分析、商业级打包风格" +description: "Commercial-grade Python installer expert for Windows: Nuitka extreme compilation, dist slimming, DLL footprint analysis, and Inno Setup packaging to ship the smallest, fastest installers. Use only for advanced packaging/optimization (minimal size, fast startup); for ordinary Python-to-exe builds use python-installer-packaging instead. 中文触发:Nuitka 极限优化、Python 商业打包、极限编译 Python、dist 瘦身、DLL 分析、最小安装包、最快启动、商业级打包风格" --- -你是一位 **Python 商业化部署专家**。你的目标是打造**体积最小、启动最快、体验最正**的 Windows 安装包。 +# Generating Python Installer (Commercial-Grade) + +You are a **Python commercial deployment expert**. Your goal is the **smallest, fastest-starting, cleanest** Windows installer. The core approach is **"Nuitka folder mode (dist) + Inno Setup packaging"** — no single-file builds, no stray console window. + +## When to Activate + +Activate when the user explicitly asks for **advanced** Python packaging or size/startup optimization on Windows: + +- Nuitka extreme / commercial-grade compilation, smallest-size or fastest-startup builds +- `dist` folder slimming, DLL footprint analysis, 32-bit vs 64-bit size tradeoffs +- Inno Setup packaging with full metadata and a clean, residue-free uninstall + +For ordinary "turn my script into an exe" requests, defer to `python-installer-packaging`. + +## How It Works + +1. **Confirm build parameters** — app name, version, publisher, exe name, source/output dirs, icon. Never auto-fill; ask the user. +2. **Verify the source build** — console disabled, LTO enabled, VC++ runtime present. +3. **Compile with Nuitka** using the module-exclusion and plugin strategy below. +4. **Slim the `dist` folder** — strip debug symbols, caches, tests, and docs, with safeguards for runtime-required metadata. +5. **Analyze DLLs** to find and trim the largest dependencies. +6. **Package with Inno Setup** — LZMA2 ultra compression, full metadata, residue-free uninstall, and an arch-matched VC++ redistributable. + +## Examples + +- "用 Nuitka 把这个 PySide2 项目打成最小体积的商业安装包" → run the full workflow: recommend 32-bit, exclude WebEngine/3D/Charts, slim `dist`, package with Inno Setup. +- "我的 exe 有 400 MB,怎么瘦身到一半" → analyze DLLs, switch to `opencv-python-headless`, drop `opengl32sw`, apply `dist` slimming. +- "安装后在纯净系统打不开" → ensure the matching-arch VC++ redistributable is bundled in the Inno Setup script. + +--- ## 核心理念 + 坚持 **"Nuitka 文件夹模式(dist) + Inno Setup 封装"** 方案。拒绝单文件版,拒绝黑窗。 --- -## 📊 实战参考案例(某商业级桌面应用) +## 实战参考案例(生产级 PySide2 桌面应用,323 MB,含 OpenCV / Playwright) ### 项目概况 - **总体积**: 323 MB @@ -20,12 +50,13 @@ description: "商业级 Python 打包专家,基于 某商业级桌面应用 - **DLL 数量**: 71 个,总计 93.23 MB ### 关键优化策略 -1. ✅ **使用 32 位 Python** → 体积减少 20-30% -2. ✅ **base_library.zip 压缩标准库** → 0.74 MB -3. ✅ **精简模块排除** → 无 pytest/unittest/setuptools -4. ✅ **精简 Qt 插件** → 只保留必要插件 +1. PASS: **使用 32 位 Python** → 体积减少 20-30% +2. PASS: **base_library.zip 压缩标准库** → 0.74 MB +3. PASS: **精简模块排除** → 无 pytest/unittest/setuptools +4. PASS: **精简 Qt 插件** → 只保留必要插件 ### 体积分布 + | 组件 | 体积 | 占比 | 优化建议 | |------|------|------|---------| | playwright | 76.74 MB | 23.8% | 非必要可移除 | @@ -34,7 +65,8 @@ description: "商业级 Python 打包专家,基于 某商业级桌面应用 | 其他依赖 | 161.36 MB | 49.9% | - | ### 预期效果对比 -| 项目类型 | Nuitka 原始 | 优化后 | 某商业级桌面应用 参考 | + +| 项目类型 | Nuitka 原始 | 优化后 | 参考项目实测 | |---------|------------|--------|----------------| | Tkinter + 标准库 | 150-250 MB | **80-120 MB** | - | | PyQt/PySide | 200-400 MB | **120-250 MB** | 323 MB (含 OpenCV 等) | @@ -42,13 +74,13 @@ description: "商业级 Python 打包专家,基于 某商业级桌面应用 --- -## 核心工作流 (Workflow) - ⚠️ 严格执行 +## 核心工作流 (Workflow) - WARNING: 严格执行 当用户请求打包时,按照以下步骤操作: -**步骤 1:强制参数确认(❌ 禁止使用默认值)** +**步骤 1:强制参数确认(FAIL: 禁止使用默认值)** -> **⚠️ 重要规则:以下所有参数必须逐一向用户确认,禁止自动填充或使用默认值!** +> **WARNING: 重要规则:以下所有参数必须逐一向用户确认,禁止自动填充或使用默认值!** 必须向用户询问并确认以下信息(*等待用户明确回复后才能继续*): @@ -79,7 +111,7 @@ description: "商业级 Python 打包专家,基于 某商业级桌面应用 **步骤 2:源文件质量与编译检查(关键)** 在生成代码之前,必须向用户发出以下**关键确认**(因为 Inno Setup 只是打包工具,无法改变程序本身的运行属性): -> "⚠️ **编译参数检查**: +> "WARNING: **编译参数检查**: > 1. **去黑窗**:请确认您的 dist 文件夹是使用 `nuitka --windows-console-mode=disable` 编译的。(否则安装后依然会有黑框) > 2. **高性能**:请确认是否使用了 `--lto=yes`。(否则启动速度可能不理想) > 3. **运行库**:请确保 dist 文件夹内已包含必要的 VC++ 运行库,防止在纯净系统上无法运行。 @@ -91,11 +123,11 @@ description: "商业级 Python 打包专家,基于 某商业级桌面应用 --- -## 🚀 Nuitka 极限优化编译(基于 某商业级桌面应用 经验) +## Nuitka 极限优化编译(基于 参考项目经验) ### 一、32 位 vs 64 位选择策略 -**某商业级桌面应用 使用 32 位 Python 的原因**: +**参考项目使用 32 位 Python 的原因**: | 组件 | 64位体积 | 32位体积 | 节省 | |------|---------|---------|------| @@ -105,9 +137,9 @@ description: "商业级 Python 打包专家,基于 某商业级桌面应用 | **总体** | 基准 | **-20~30%** | - | **推荐使用 32 位条件**: -- ✅ 程序内存占用 < 2GB -- ✅ 不处理超大文件(< 2GB) -- ✅ 目标用户是普通办公电脑 +- PASS: 程序内存占用 < 2GB +- PASS: 不处理超大文件(< 2GB) +- PASS: 目标用户是普通办公电脑 **32 位编译方法**: ```bash @@ -121,7 +153,7 @@ py -3.12-32 -m pip install -r requirements.txt py -3.12-32 -m nuitka --standalone ...你的参数 ``` -### 二、模块排除清单(某商业级桌面应用 验证过的) +### 二、模块排除清单(参考项目验证过的) **安全排除列表**(运行时不需要): ``` @@ -188,7 +220,7 @@ chcp 65001 >nul setlocal enabledelayedexpansion echo ======================================== -echo Nuitka 极限优化编译(参考 某商业级桌面应用) +echo Nuitka 极限优化编译(参考 参考项目) echo ======================================== REM === 配置区域(请修改为你的实际值) === @@ -200,7 +232,7 @@ REM === 自动检测 CPU 核心数 === for /f "tokens=2 delims==" %%a in ('wmic cpu get NumberOfLogicalProcessors /value ^| find "="') do set CPU_CORES=%%a set /a BUILD_JOBS=%CPU_CORES% -REM === 某商业级桌面应用 的模块排除清单 === +REM === 参考项目的模块排除清单 === set EXCLUDE_MODULES=unittest,test,pytest,_pytest,doctest,pdb,pdbpp set EXCLUDE_MODULES=%EXCLUDE_MODULES%,setuptools,pip,distutils,pkg_resources set EXCLUDE_MODULES=%EXCLUDE_MODULES%,email.mime,http.server,xmlrpc,pydoc @@ -211,7 +243,7 @@ if exist dist rd /s /q dist if exist build rd /s /q build echo. -echo [2/4] Nuitka 编译中(应用 某商业级桌面应用 优化策略)... +echo [2/4] Nuitka 编译中(应用 参考项目优化策略)... echo - CPU 核心: %CPU_CORES% (使用 %BUILD_JOBS% 线程) echo - 模块排除: %EXCLUDE_MODULES% echo. @@ -240,13 +272,13 @@ if %errorlevel% neq 0 ( echo. echo [3/4] 统计编译结果... -for /f "tokens=3" %%a in ('dir "dist\%APP_NAME%.dist" /s /-c ^| find "个文件"') do set TOTAL_SIZE=%%a +for /f %%a in ('powershell -NoProfile -Command "(Get-ChildItem -LiteralPath 'dist\%APP_NAME%.dist' -Recurse -File | Measure-Object -Property Length -Sum).Sum"') do set TOTAL_SIZE=%%a set TOTAL_SIZE=%TOTAL_SIZE:,=% set /a SIZE_MB=%TOTAL_SIZE% / 1048576 echo - 编译后体积: %SIZE_MB% MB echo. -echo [4/4] 执行瘦身清理(参考 某商业级桌面应用 策略)... +echo [4/4] 执行瘦身清理(参考 参考项目策略)... powershell -ExecutionPolicy Bypass -File slim_dist.ps1 -DistPath "dist\%APP_NAME%.dist" echo. @@ -256,7 +288,7 @@ echo ======================================== pause ``` -### 五、dist 瘦身脚本(某商业级桌面应用 级别清理) +### 五、dist 瘦身脚本(参考项目级别清理) **保存为 `slim_dist.ps1`(和 build_optimized.bat 同目录)**: @@ -268,7 +300,7 @@ param( $ErrorActionPreference = "SilentlyContinue" Write-Host "`n========================================" -ForegroundColor Cyan -Write-Host "dist 瘦身清理(参考 某商业级桌面应用 策略)" -ForegroundColor Cyan +Write-Host "dist 瘦身清理(参考 参考项目策略)" -ForegroundColor Cyan Write-Host "========================================" -ForegroundColor Cyan if (-not (Test-Path $DistPath)) { @@ -280,8 +312,8 @@ if (-not (Test-Path $DistPath)) { $InitialSize = (Get-ChildItem -Path $DistPath -Recurse -File | Measure-Object -Property Length -Sum).Sum / 1MB Write-Host "`n初始体积: $([math]::Round($InitialSize, 2)) MB" -ForegroundColor Yellow -# 某商业级桌面应用 特征:没有 .pdb, .pyi, __pycache__, test 等 -Write-Host "`n[应用 某商业级桌面应用 的清理策略...]" -ForegroundColor Green +# 参考项目特征:没有 .pdb, .pyi, __pycache__, test 等 +Write-Host "`n[应用 参考项目的清理策略...]" -ForegroundColor Green # 1. 删除调试符号 Write-Host "`n[1/7] 删除 .pdb 调试符号..." -ForegroundColor Green @@ -367,7 +399,8 @@ $distInfoDirs = Get-ChildItem -Path $DistPath -Recurse -Directory -Filter "*.dis $removedCount = 0 $removedSize = 0 foreach ($infoDir in $distInfoDirs) { - $filesToRemove = @("RECORD", "INSTALLER", "direct_url.json", "entry_points.txt") + # 仅删安装期记账文件;保留 METADATA 与 entry_points.txt(运行期被 importlib.metadata 读取,删除会破坏插件发现) + $filesToRemove = @("RECORD", "INSTALLER", "direct_url.json") foreach ($fileName in $filesToRemove) { $file = Join-Path $infoDir.FullName $fileName if (Test-Path $file) { @@ -397,9 +430,9 @@ Write-Host "最终体积: $([math]::Round($FinalSize, 2)) MB" -ForegroundColor G Write-Host "节省空间: $([math]::Round($SavedSize, 2)) MB ($([math]::Round($SavedPercent, 1))%)" -ForegroundColor Cyan Write-Host "========================================`n" -ForegroundColor Cyan -# 对比 某商业级桌面应用 +# 对比 参考项目 Write-Host "[对比参考]" -ForegroundColor Yellow -Write-Host "某商业级桌面应用 总体积: 323 MB (包含 PyQt, OpenCV, Playwright 等重量级库)" -ForegroundColor Gray +Write-Host "参考项目总体积: 323 MB (包含 PyQt, OpenCV, Playwright 等重量级库)" -ForegroundColor Gray Write-Host "如果你的项目是纯 Tkinter + 标准库,目标应该在 80-150 MB" -ForegroundColor Gray ``` @@ -412,7 +445,7 @@ Write-Host "如果你的项目是纯 Tkinter + 标准库,目标应该在 80-15 ```python """ DLL 依赖分析工具 -参考 某商业级桌面应用 的 DLL 管理策略,帮助识别体积大户和优化建议 +参考 参考项目的 DLL 管理策略,帮助识别体积大户和优化建议 """ import sys from pathlib import Path @@ -426,7 +459,7 @@ def analyze_dlls(dist_path: str): return print("=" * 70) - print("DLL 依赖分析(参考 某商业级桌面应用 策略)") + print("DLL 依赖分析(参考 参考项目策略)") print("=" * 70) # 收集所有 DLL @@ -445,8 +478,8 @@ def analyze_dlls(dist_path: str): print(f"\n总 DLL 数量: {len(dll_files)}") print(f"总 DLL 体积: {total_size / 1024 / 1024:.2f} MB\n") - # 某商业级桌面应用 对比 - print("[对比参考] 某商业级桌面应用 的 DLL 情况:") + # 参考项目对比 + print("[对比参考] 参考项目的 DLL 情况:") print(" - 总数量: 71 个") print(" - 总体积: 93.23 MB") print(" - 最大的: libopenblas (26.85 MB), opengl32sw (15.25 MB)\n") @@ -456,7 +489,7 @@ def analyze_dlls(dist_path: str): if large_dlls: print("=" * 70) - print("⚠️ 大于 3MB 的 DLL(需重点关注)") + print("WARNING: 大于 3MB 的 DLL(需重点关注)") print("=" * 70) for dll, size in large_dlls: @@ -471,21 +504,21 @@ def analyze_dlls(dist_path: str): suggestions = get_optimization_suggestion(name_lower) if suggestions: for suggestion in suggestions: - print(f" 💡 {suggestion}") + print(f" {suggestion}") # 检查冗余 DLL print("\n" + "=" * 70) - print("🔍 冗余检查") + print(" 冗余检查") print("=" * 70) # 检查调试版本 debug_dlls = [dll for dll, _ in dll_data if dll.stem.endswith('d')] if debug_dlls: - print(f"\n⚠️ 发现 {len(debug_dlls)} 个调试版本 DLL(可以删除):") + print(f"\nWARNING: 发现 {len(debug_dlls)} 个调试版本 DLL(可以删除):") for dll in debug_dlls: print(f" - {dll.name}") else: - print("\n✅ 未发现调试版本 DLL(已优化)") + print("\nPASS: 未发现调试版本 DLL(已优化)") # VC++ Runtime vc_runtimes = [dll for dll, _ in dll_data if 'vcruntime' in dll.name.lower() or 'msvcp' in dll.name.lower()] @@ -494,11 +527,11 @@ def analyze_dlls(dist_path: str): for dll in vc_runtimes: size_mb = dll.stat().st_size / 1024 / 1024 print(f" - {dll.name} ({size_mb:.2f} MB)") - print(" 💡 这些是必需的,某商业级桌面应用 也包含了这些文件") + print(" 这些是必需的,参考项目也包含了这些文件") # 全部 DLL 列表 print("\n" + "=" * 70) - print("📋 完整 DLL 列表(按体积排序,前 20)") + print(" 完整 DLL 列表(按体积排序,前 20)") print("=" * 70) print(f"\n{'体积 (MB)':>10} {'文件名':<30} 位置") print("-" * 70) @@ -519,26 +552,26 @@ def get_optimization_suggestion(dll_name: str) -> list: suggestions = [] if "openblas" in dll_name or "mkl" in dll_name: - suggestions.append("数学运算库,某商业级桌面应用 的 libopenblas 有 26.85 MB") + suggestions.append("数学运算库,参考项目的 libopenblas 有 26.85 MB") suggestions.append("如不需要高性能计算可考虑轻量版") elif "opencv" in dll_name or "ffmpeg" in dll_name: - suggestions.append("OpenCV 相关,某商业级桌面应用 的 opencv_videoio_ffmpeg 有 18.48 MB") + suggestions.append("OpenCV 相关,参考项目的 opencv_videoio_ffmpeg 有 18.48 MB") suggestions.append("考虑用 opencv-python-headless") elif "qt5" in dll_name or "qt6" in dll_name or "pyside" in dll_name: - suggestions.append("Qt 库,某商业级桌面应用 的 Qt5Core 有 5.13 MB") + suggestions.append("Qt 库,参考项目的 Qt5Core 有 5.13 MB") suggestions.append("可排除不需要的模块(WebEngine, 3D, Charts)") elif "opengl" in dll_name and "sw" in dll_name: - suggestions.append("OpenGL 软件渲染器,某商业级桌面应用 保留了 15.25 MB") + suggestions.append("OpenGL 软件渲染器,参考项目保留了 15.25 MB") suggestions.append("通常可以删除(使用硬件渲染)") elif "d3dcompiler" in dll_name: - suggestions.append("DirectX 编译器,某商业级桌面应用 有 3.53 MB") + suggestions.append("DirectX 编译器,参考项目有 3.53 MB") elif "mfc140" in dll_name: - suggestions.append("MFC 库,某商业级桌面应用 有 4.89 MB") + suggestions.append("MFC 库,参考项目有 4.89 MB") return suggestions @@ -559,7 +592,7 @@ python analyze_dlls.py dist/你的软件名.dist --- -## 📋 完整优化工作流 +## 完整优化工作流 ### 步骤 1:修改编译脚本配置 @@ -606,7 +639,7 @@ Remove-Item "opengl32sw.dll" -Force --- -## 🎯 VC++ 运行库处理方案 +## VC++ 运行库处理方案 ### 方案一:静态链接(推荐) ```bash @@ -617,24 +650,27 @@ nuitka --static-libpython=yes ... 在 Inno Setup 脚本中添加: ```iss +; WARNING: VC++ 运行库架构必须与 Python/Nuitka 构建架构一致。 +; 本 skill 推荐 32 位 Python,故默认捆绑 vc_redist.x86.exe; +; 若用 64 位 Python 编译,请把下面两处改为 vc_redist.x64.exe。 [Files] -Source: "{#MySourceDir}\..\vc_redist.x64.exe"; DestDir: "{tmp}"; Flags: deleteafterinstall +Source: "{#MySourceDir}\..\vc_redist.x86.exe"; DestDir: "{tmp}"; Flags: deleteafterinstall [Run] -Filename: "{tmp}\vc_redist.x64.exe"; Parameters: "/quiet /norestart"; StatusMsg: "正在安装运行库..."; Flags: waituntilterminated +Filename: "{tmp}\vc_redist.x86.exe"; Parameters: "/quiet /norestart"; StatusMsg: "正在安装运行库..."; Flags: waituntilterminated ``` > 下载地址:[Microsoft Visual C++ Redistributable](https://learn.microsoft.com/en-us/cpp/windows/latest-supported-vc-redist) --- -## 📜 Inno Setup 脚本模板(商业终极版) +## Inno Setup 脚本模板(商业终极版) ```iss ; ===================================================================== -; ⚠️ 商业级 Python 安装脚本 (Inno Setup 6.x) +; WARNING: 商业级 Python 安装脚本 (Inno Setup 6.x) ; 特性:LZMA2 极限压缩 | 全中文 | 完整元数据 | 无残留卸载 -; 参考:某商业级桌面应用 (323 MB, LZMA2 压缩) +; 参考:参考项目(323 MB, LZMA2 压缩) ; ===================================================================== ; --- 1. 参数定义 --- @@ -675,13 +711,16 @@ SetupIconFile={#MyIconPath} UninstallDisplayIcon={app}\{#MyAppExeName} #endif -; --- 🚀 核心压缩 (参考 某商业级桌面应用) --- +; --- 核心压缩 (参考 参考项目) --- Compression=lzma2/ultra64 SolidCompression=yes LZMAUseSeparateProcess=yes ; --- 架构 --- -ArchitecturesInstallIn64BitMode=x64compatible +; 注意:仅 64 位 Python 构建才设此项。本 skill 推荐 32 位 Python—— +; 32 位构建请保持注释,使应用按 32 位安装并与上面捆绑的 vc_redist.x86.exe 匹配。 +; 仅当用 64 位 Python 编译时才取消注释。 +;ArchitecturesInstallIn64BitMode=x64compatible [Languages] Name: "chinesesimplified"; MessagesFile: "compiler:Languages\ChineseSimplified.isl" @@ -732,7 +771,7 @@ Filename: "{app}\{#MyAppExeName}"; Description: "{cm:LaunchProgram,{#MyAppName}} ### Q2: 安装包体积过大? **优化方法**: 1. 使用 32 位 Python 编译(节省 20-30%) -2. 应用 某商业级桌面应用 的模块排除清单 +2. 应用 参考项目的模块排除清单 3. 启用 Anti-Bloat 插件 4. 执行 dist 文件夹瘦身脚本 5. 分析 DLL,移除不必要的大文件 @@ -754,7 +793,7 @@ Filename: "{app}\{#MyAppExeName}"; Description: "{cm:LaunchProgram,{#MyAppName}} - **安装后提示缺少 python3xx.dll**:必须使用 Nuitka `--standalone`;确认 dist 内存在该 dll;不要打单文件版。 - **安装后点击无反应**:GUI 启动期可能被重依赖阻塞;将重依赖延迟到"开始导出"再 import;添加日志排查。 - **Nuitka + MinGW 在非 ASCII 路径报错**:把源码复制到 ASCII 目录再编译;设置 `PYTHONIOENCODING=utf-8`。 -- **Inno Setup 警告 `x64` 已弃用**:改为 `ArchitecturesInstallIn64BitMode=x64compatible`。 +- **Inno Setup 警告 `x64` 已弃用(仅 64 位构建需要 64 位安装模式时)**:改为 `ArchitecturesInstallIn64BitMode=x64compatible`;32 位构建无需此项。 - **`--disable-console` 已废弃**:改用 `--windows-console-mode=disable`。 - **dist 出现 `_nuitka_temp.exe`**:在 [Files] 中排除它。 @@ -765,15 +804,15 @@ Filename: "{app}\{#MyAppExeName}"; Description: "{cm:LaunchProgram,{#MyAppName}} | 优化组合 | 体积减少 | 启动提升 | 风险等级 | |----------|----------|----------|----------| | 基础编译 | 基准 | 基准 | 无 | -| + `--lto=yes` | 5-10% | 10-20% | ✅ 无 | -| + anti-bloat | 15-25% | - | ✅ 无 | -| + 模块排除 | 20-35% | 5% | ✅ 无 | -| + dist 瘦身 | 25-40% | - | ✅ 无 | -| + 32 位编译 | 40-60% | - | ✅ 无 | -| **全部组合** | **45-65%** | **15-25%** | ✅ **无风险** | +| + `--lto=yes` | 5-10% | 10-20% | PASS: 无 | +| + anti-bloat | 15-25% | - | PASS: 无 | +| + 模块排除 | 20-35% | 5% | PASS: 无 | +| + dist 瘦身 | 25-40% | - | PASS: 无 | +| + 32 位编译 | 40-60% | - | PASS: 无 | +| **全部组合** | **45-65%** | **15-25%** | PASS: **无风险** | -> ⚠️ **不建议使用 UPX 压缩**,虽然能进一步减小体积,但极易触发杀毒软件误报。 +> WARNING: **不建议使用 UPX 压缩**,虽然能进一步减小体积,但极易触发杀毒软件误报。 --- -**基于 某商业级桌面应用 实战经验优化,助你打造商业级安装包!** 🚀 +**基于 参考项目实战经验优化,助你打造商业级安装包!**