AiDOPWarehouse/doc/_tmp_scan_blueprints.py

# -*- coding: utf-8 -*-
"""扫描交付文档目录下各模块蓝图设计方案：列出标题层级与内嵌图片清单。"""

import os
import shutil
import sys
import tempfile

sys.stdout.reconfigure(encoding="utf-8", errors="replace")

from docx import Document
from docx.oxml.ns import qn

ROOT = r"C:\Users\skygu\OneDrive\Projects\AIDOP\项目\项目管理\产互联项目管理\交付文档"
OUT = os.path.join(os.path.dirname(os.path.abspath(__file__)),
                   "_tmp_blueprint_scan.txt")

TARGETS = []
for dirpath, _, files in os.walk(ROOT):
    for fn in files:
        if fn.endswith(".docx") and "蓝图设计方案" in fn and "Tmp" not in dirpath \
                and "更新" not in dirpath and "(1)" not in fn:
            TARGETS.append(os.path.join(dirpath, fn))
TARGETS.sort()

tmpdir = tempfile.mkdtemp()
lines = []
for path in TARGETS:
    rel = os.path.relpath(path, ROOT)
    local = os.path.join(tmpdir, os.path.basename(path))
    try:
        shutil.copy2(path, local)
        doc = Document(local)
    except Exception as e:
        lines.append(f"\n### {rel}  [打不开: {type(e).__name__}]")
        continue

    imgs = [p for p in doc.part.package.parts
            if p.partname.startswith("/word/media/")]
    total_kb = sum(len(p.blob) for p in imgs) / 1024
    lines.append("\n" + "=" * 78)
    lines.append(f"### {rel}")
    lines.append(f"    段落 {len(doc.paragraphs)}　表格 {len(doc.tables)}　"
                 f"图片 {len(imgs)} 张（{total_kb:,.0f} KB）")

    # 标题结构
    lines.append("    --- 标题 ---")
    for p in doc.paragraphs:
        sname = p.style.name if p.style is not None else ""
        t = p.text.strip()
        if not t:
            continue
        lvl = None
        if sname.startswith("Heading"):
            try:
                lvl = int(sname.split()[-1])
            except ValueError:
                lvl = 1
        elif sname.startswith("标题"):
            lvl = 1
        if lvl and lvl <= 3:
            lines.append("    " + "  " * (lvl - 1) + f"[H{lvl}] {t[:70]}")

    # 图片：找出图片所在段落的上下文（图题）
    lines.append("    --- 图片上下文 ---")
    n = 0
    for i, p in enumerate(doc.paragraphs):
        if p._p.findall(".//" + qn("a:blip")):
            n += 1
            cap = ""
            for j in range(i + 1, min(i + 4, len(doc.paragraphs))):
                c = doc.paragraphs[j].text.strip()
                if c:
                    cap = c[:70]
                    break
            prev = ""
            for j in range(i - 1, max(i - 4, -1), -1):
                c = doc.paragraphs[j].text.strip()
                if c:
                    prev = c[:70]
                    break
            lines.append(f"      img{n:>2}  上文: {prev}")
            lines.append(f"            下文: {cap}")

with open(OUT, "w", encoding="utf-8") as f:
    f.write("\n".join(lines))
shutil.rmtree(tmpdir, ignore_errors=True)
print(f"扫描 {len(TARGETS)} 份蓝图 -> {OUT}")
for t in TARGETS:
    print("   ", os.path.relpath(t, ROOT))