first commit

2026-07-20 13:16:17 +08:00
commit 7affeaa253
235 changed files with 42351 additions and 0 deletions
--- a/image_test/inspect_xml.py
+++ b/image_test/inspect_xml.py
@@ -0,0 +1,100 @@
+"""检查 signed.docx 中签名图片对应的 drawing XML，看实际显示尺寸（EMU）。
+
+目的:
+    字节已经证明一致（e2e_test.py），但用户仍看到"空白"。
+    本脚本读 document.xml，找到图片所在段落的:
+      - wp:extent (图片显示尺寸 EMU)
+      - 段落 pPr/spacing (line 行高)
+      - 段落所在表格单元格 tcW (单元格宽度)
+    对比图片实际像素 vs 显示尺寸 vs 段落行高，定位空白的视觉来源。
+
+EMU 换算:
+    1 inch = 914400 EMU = 2.54 cm
+    1 cm   = 360000 EMU
+"""
+import sys
+import zipfile
+import xml.etree.ElementTree as ET
+from pathlib import Path
+
+BASE_DIR = Path(__file__).resolve().parent
+DOCX_PATH = BASE_DIR / "output" / "e2e" / "signed.docx"
+
+W = 'http://schemas.openxmlformats.org/wordprocessingml/2006/main'
+WP = 'http://schemas.openxmlformats.org/drawingml/2006/wordprocessingDrawing'
+A = 'http://schemas.openxmlformats.org/drawingml/2006/main'
+
+
+def emu_to_cm(emu):
+    try:
+        return int(emu) / 360000.0
+    except (TypeError, ValueError):
+        return None
+
+
+def main():
+    if not DOCX_PATH.exists():
+        print(f"docx 不存在: {DOCX_PATH}")
+        print("请先运行: python image_test/e2e_test.py")
+        sys.exit(1)
+
+    with zipfile.ZipFile(DOCX_PATH) as z:
+        xml_bytes = z.read('word/document.xml')
+
+    root = ET.fromstring(xml_bytes)
+
+    print("=== 段落列表（含 spacing / drawing）===\n")
+    for i, p in enumerate(root.iter(f'{{{W}}}p'), 1):
+        pPr = p.find(f'{{{W}}}pPr')
+        spacing_info = ""
+        if pPr is not None:
+            sp = pPr.find(f'{{{W}}}spacing')
+            if sp is not None:
+                line = sp.get(f'{{{W}}}line')
+                rule = sp.get(f'{{{W}}}lineRule')
+                before = sp.get(f'{{{W}}}before')
+                after = sp.get(f'{{{W}}}after')
+                if line or before or after:
+                    parts = []
+                    if line:
+                        cm = emu_to_cm(int(line) / 20 * 12700) if rule == 'exact' else None
+                        parts.append(f"line={line}({rule})->{cm:.2f}cm" if cm else f"line={line}({rule})")
+                    if before:
+                        parts.append(f"before={before}({int(before)/20:.1f}pt)")
+                    if after:
+                        parts.append(f"after={after}({int(after)/20:.1f}pt)")
+                    spacing_info = " | ".join(parts)
+
+        text = "".join(t.text or '' for t in p.iter(f'{{{W}}}t'))
+        has_drawing = len(list(p.iter(f'{{{WP}}}inline'))) > 0
+        marker = "[图]" if has_drawing else "   "
+
+        print(f"[段{i:02d}] {marker} text={text[:40]!r}")
+        if spacing_info:
+            print(f"        spacing: {spacing_info}")
+
+        for inline in p.iter(f'{{{WP}}}inline'):
+            extent = inline.find(f'{{{WP}}}extent')
+            if extent is not None:
+                cx = extent.get('cx')
+                cy = extent.get('cy')
+                print(f"        drawing extent: cx={cx}({emu_to_cm(cx):.2f}cm) cy={cy}({emu_to_cm(cy):.2f}cm)")
+
+        tc_parent = None
+        parent = p
+        while parent is not None:
+            parent_iter = list(root.iter())
+            break
+        for tc in root.iter(f'{{{W}}}tc'):
+            if p in list(tc.iter(f'{{{W}}}p')):
+                tcPr = tc.find(f'{{{W}}}tcPr')
+                if tcPr is not None:
+                    tcW = tcPr.find(f'{{{W}}}tcW')
+                    if tcW is not None:
+                        w = tcW.get(f'{{{W}}}w')
+                        t = tcW.get(f'{{{W}}}type')
+                        print(f"        cell width: w={w} type={t}")
+
+
+if __name__ == '__main__':
+    main()