mescloud/_apk_work/insert_manual_images.py

# -*- coding: utf-8 -*-
"""Insert 8.28 screenshots back into 9.19 markdown manuals."""
import json, re
from pathlib import Path

ROOT = Path(r"D:\产线\大洋\大洋操作手册汇总9.19")
IMG = ROOT / "images"
IMG_RE = re.compile(r"^\s*!\[.*?\]\(images/.*?\)\s*$")
CAP_RE = re.compile(r"^\s*<p class=\"imgcap\">.*$</s>$")

FASCICLES = [
    "1、系统管理操作手册.md",
    "2、设备管理操作手册.md",
    "3、物料管理操作手册.md",
    "4、质量管理操作手册.md",
    "5、生产管理操作手册.md",
    "6、ERP对接操作手册.md",
    "7、生产线管理操作手册.md",
    "8、PDA操作说明操作手册.md",
    "9、MES系统客户端员工扫码操作手册.md",
]


def norm(s):
    s = (s or "").replace(" ", "").replace("　", "")
    s = s.replace("→", "").replace("—", "").replace("-", "")
    s = re.sub(r"^\d+(\.\d+)*\.?", "", s)
    return s


def parse_sections(lines):
    """Return list of {level,title,start,end} covering body after heading."""
    secs = []
    for i, line in enumerate(lines):
        m = re.match(r"^(#{2,6})\s+(.*)$", line)
        if m:
            secs.append({"level": len(m.group(1)), "title": m.group(2).strip(), "start": i})
    for i, sec in enumerate(secs):
        sec["end"] = secs[i + 1]["start"] if i + 1 < len(secs) else len(lines)
    return secs


def strip_old_images(text):
    lines = text.splitlines()
    out = []
    i = 0
    while i < len(lines):
        line = lines[i]
        if IMG_RE.match(line) or line.startswith('<p class="imgcap">') or line.strip() == "*（界面截图取自 8.28 手册，现场如有改动以实际为准。）*":
            i += 1
            # skip following blank
            if i < len(lines) and not lines[i].strip():
                i += 1
            continue
        if line.strip() == "## 原手册界面截图（8.28）":
            # drop this section until next heading
            i += 1
            while i < len(lines) and not re.match(r"^## ", lines[i]):
                i += 1
            continue
        out.append(line)
        i += 1
    return "\n".join(out).rstrip() + "\n"


def score(sec, cap, heading, body=""):
    title = sec["title"]
    tn, cn, hn = norm(title), norm(cap), norm(heading)
    blob = tn + norm(body)
    s = 0
    if cn and (cn in tn or tn in cn):
        s += 60
    if hn and (hn in tn or tn in hn):
        s += 45
    hay = (title + " " + body).upper()
    for code in set(re.findall(r"OP\d+", (cap + heading).upper())):
        if code and code in hay:
            s += 80
    keys = re.findall(r"[\u4e00-\u9fff]{2,}", cap + " " + heading)
    for k in keys:
        if k in title or k in body[:800]:
            s += 12
    return s


def image_md(item):
    cap = item.get("caption") or "界面截图"
    rel = item["file"]
    if not rel.startswith("images/"):
        rel = "images/" + rel
    return "\n![%s](%s)\n\n*（界面截图取自 8.28 手册，现场如有改动以实际为准。）*\n" % (cap, rel)


def insert_into(md_name, items):
    path = ROOT / md_name
    text = strip_old_images(path.read_text(encoding="utf-8"))
    lines = text.splitlines()
    secs = parse_sections(lines)
    if not secs:
        path.write_text(text + "\n## 原手册界面截图（8.28）\n" + "".join(image_md(x) for x in items), encoding="utf-8")
        return len(items), 0
    buckets = {id(sec): [] for sec in secs}
    unmatched = []
    for item in items:
        best, best_s = None, 0
        for sec in secs:
            body = "\n".join(lines[sec["start"]:sec["end"]])
            sc = score(sec, item.get("caption", ""), item.get("heading", ""), body)
            if sc > best_s or (best is not None and sc == best_s and sec["level"] > best["level"]):
                best, best_s = sec, sc
        # special: login / home screens
        cap = item.get("caption") or ""
        if "登录" in cap:
            for sec in secs:
                if "登录" in sec["title"] or "安装" in sec["title"]:
                    best, best_s = sec, max(best_s, 80)
        if "主功能" in cap or "我的界面" in cap:
            for sec in secs:
                if "界面" in sec["title"]:
                    best, best_s = sec, max(best_s, 80)
        if best is None or best_s < 12:
            unmatched.append(item)
        else:
            buckets[id(best)].append(item)
    # insert from bottom so indexes stay valid
    for sec in reversed(secs):
        imgs = buckets[id(sec)]
        if not imgs:
            continue
        # insert before sec['end'] (start of next heading), after trailing blanks of this section
        pos = sec["end"]
        while pos > sec["start"] + 1 and pos <= len(lines) and (pos == len(lines) or not lines[pos - 1].strip()):
            pos -= 1
        block = "".join(image_md(x) for x in imgs).splitlines()
        lines[pos:pos] = [""] + block + [""]
    text = "\n".join(lines).rstrip() + "\n"
    if unmatched:
        extra = ["", "## 原手册界面截图（8.28）", "", "以下截图未能自动对上新目录，仍予保留。", ""]
        for x in unmatched:
            extra.extend(image_md(x).splitlines())
        text = text.rstrip() + "\n" + "\n".join(extra) + "\n"
    path.write_text(text, encoding="utf-8")
    return len(items) - len(unmatched), len(unmatched)


def main():
    total_m, total_u = 0, 0
    for md in FASCICLES:
        stem = md[:-3]
        man = IMG / stem / "manifest.json"
        if not man.exists():
            print("no manifest", stem)
            continue
        items = json.loads(man.read_text(encoding="utf-8"))
        # keep only files that exist
        ok = []
        for it in items:
            p = ROOT / it["file"]
            if p.exists():
                ok.append(it)
        m, u = insert_into(md, ok)
        total_m += m
        total_u += u
        print(stem, "matched", m, "appendix", u, "total", len(ok))
    print("ALL matched", total_m, "appendix", total_u)


if __name__ == "__main__":
    main()