chore: establish local platform baseline

2026-08-26 06:41:54 +08:00
commit 29bdf4900f
575 changed files with 95644 additions and 0 deletions
@@ -0,0 +1,121 @@
+param(
+    [Parameter(Mandatory = $true)]
+    [string]$Source,
+
+    [string]$Output = (Join-Path $PSScriptRoot '..\artifacts\requirements-source-inventory.md')
+)
+
+$ErrorActionPreference = 'Stop'
+
+Add-Type -AssemblyName System.IO.Compression.FileSystem
+
+$sourcePath = (Resolve-Path -LiteralPath $Source).Path
+$outputPath = [System.IO.Path]::GetFullPath($Output)
+$outputDirectory = Split-Path -Parent $outputPath
+if (-not (Test-Path -LiteralPath $outputDirectory)) {
+    New-Item -ItemType Directory -Path $outputDirectory | Out-Null
+}
+
+$stylesByID = @{}
+$archive = [System.IO.Compression.ZipFile]::OpenRead($sourcePath)
+try {
+    $entry = $archive.GetEntry('word/document.xml')
+    if ($null -eq $entry) {
+        throw 'DOCX中缺少word/document.xml'
+    }
+    $stream = $entry.Open()
+    try {
+        $reader = [System.IO.StreamReader]::new($stream, [System.Text.Encoding]::UTF8, $true)
+        try {
+            [xml]$document = $reader.ReadToEnd()
+        }
+        finally {
+            $reader.Dispose()
+        }
+    }
+    finally {
+        $stream.Dispose()
+    }
+
+    $stylesEntry = $archive.GetEntry('word/styles.xml')
+    if ($null -ne $stylesEntry) {
+        $stylesStream = $stylesEntry.Open()
+        try {
+            $stylesReader = [System.IO.StreamReader]::new($stylesStream, [System.Text.Encoding]::UTF8, $true)
+            try {
+                [xml]$stylesDocument = $stylesReader.ReadToEnd()
+            }
+            finally {
+                $stylesReader.Dispose()
+            }
+        }
+        finally {
+            $stylesStream.Dispose()
+        }
+        $stylesNamespace = [System.Xml.XmlNamespaceManager]::new($stylesDocument.NameTable)
+        $stylesNamespace.AddNamespace('w', 'http://schemas.openxmlformats.org/wordprocessingml/2006/main')
+        foreach ($styleNode in $stylesDocument.SelectNodes('/w:styles/w:style', $stylesNamespace)) {
+            $idAttribute = $styleNode.Attributes.GetNamedItem('styleId', 'http://schemas.openxmlformats.org/wordprocessingml/2006/main')
+            $nameNode = $styleNode.SelectSingleNode('./w:name', $stylesNamespace)
+            if ($null -eq $idAttribute -or $null -eq $nameNode) {
+                continue
+            }
+            $nameAttribute = $nameNode.Attributes.GetNamedItem('val', 'http://schemas.openxmlformats.org/wordprocessingml/2006/main')
+            if ($null -ne $nameAttribute) {
+                $stylesByID[$idAttribute.Value] = $nameAttribute.Value
+            }
+        }
+    }
+}
+finally {
+    $archive.Dispose()
+}
+
+$namespace = [System.Xml.XmlNamespaceManager]::new($document.NameTable)
+$namespace.AddNamespace('w', 'http://schemas.openxmlformats.org/wordprocessingml/2006/main')
+$paragraphs = $document.SelectNodes('/w:document/w:body/w:p', $namespace)
+if ($paragraphs.Count -eq 0) {
+    throw 'DOCX正文没有可追踪段落'
+}
+
+$hash = (Get-FileHash -LiteralPath $sourcePath -Algorithm SHA256).Hash.ToLowerInvariant()
+$lines = [System.Collections.Generic.List[string]]::new()
+$lines.Add('# 原始需求来源段落清单')
+$lines.Add('')
+$lines.Add('> 此文件由`scripts/extract-requirements.ps1`从原始DOCX机械提取，用于稳定引用来源，不替代原需求文档。`Pxxxx`是`word/document.xml`正文段落的零基顺序号；空段落保留编号但不输出正文。')
+$lines.Add('')
+$lines.Add("- 来源文件：$([System.IO.Path]::GetFileName($sourcePath))")
+$lines.Add(('- SHA-256：`{0}`' -f $hash))
+$lines.Add("- 正文段落总数：$($paragraphs.Count)")
+$lines.Add('')
+$lines.Add('## 非空段落')
+$lines.Add('')
+
+$nonEmpty = 0
+for ($index = 0; $index -lt $paragraphs.Count; $index++) {
+    $paragraph = $paragraphs[$index]
+    $textParts = $paragraph.SelectNodes('.//w:t', $namespace) | ForEach-Object { $_.InnerText }
+    $text = (($textParts -join '') -replace '[\r\n\t]+', ' ' -replace '\s{2,}', ' ').Trim()
+    if ([string]::IsNullOrWhiteSpace($text)) {
+        continue
+    }
+    $nonEmpty++
+    $styleNode = $paragraph.SelectSingleNode('./w:pPr/w:pStyle', $namespace)
+    $style = 'Normal'
+    if ($null -ne $styleNode) {
+        $styleAttribute = $styleNode.Attributes.GetNamedItem('val', 'http://schemas.openxmlformats.org/wordprocessingml/2006/main')
+        if ($null -ne $styleAttribute -and -not [string]::IsNullOrWhiteSpace($styleAttribute.Value)) {
+            $style = $styleAttribute.Value
+            if ($stylesByID.ContainsKey($style)) {
+                $style = $stylesByID[$style]
+            }
+        }
+    }
+    $safeText = $text.Replace('\', '\\').Replace('`', '\`')
+    $lines.Add(('- `P{0}` [{1}] {2}' -f $index.ToString('0000'), $style, $safeText))
+}
+
+$lines.Insert(7, "- 非空段落数：$nonEmpty")
+[System.IO.File]::WriteAllLines($outputPath, $lines, [System.Text.UTF8Encoding]::new($false))
+
+Write-Output "已提取 $nonEmpty/$($paragraphs.Count) 个段落：$outputPath"