OnebotCatalog/onebot-data/tools/gen-pending-manual.py

# -*- coding: utf-8 -*-
"""
为待补分类(定位销/勾销缸/翻转缸/电缸)生成附图。
来源: Univer 官方图册 PDF（X盘 官方图册 / 选型表/spanner）。
输出: manual/<系列>/p<NN>.png + <系列>_manual.pdf
用法: PYTHONUTF8=1 python gen-pending-manual.py [--clean]
"""
import fitz, os, shutil, sys

SRC = r'X:/新开发产品-Research and Development/2_产品资料/Univer资料/官方图册'
SPANNER = r'X:/新开发产品-Research and Development/2_产品资料/Univer资料/选型表/spanner'
DST = r'D:/开发/OnebotCatalog/onebot-data/catalog/manual'
DPI = 110
CLEAN = '--clean' in sys.argv

# 系列 -> (目录, [pdf 名])；pdf 名不含扩展名
SPECS = {
    # ---- 定位销 ----
    'ULA': (SRC, ['ula40cten_en419', 'ula63cten_en407']),
    'ULB': (SRC, ['ulb32cten_en423']),
    'ULL': (SRC, ['ull50cten_en411']),
    'ULM': (SRC, ['ulm50cten_en410']),
    'ULO': (SRC, ['ulo50cten_en413']),
    'ULP': (SRC, ['ulp32cten_en404', 'ulp50cten_en405', 'ulp63cten_en406']),
    'ULR': (SRC, ['ulr50nken_en476', 'ulr63cten_en477']),
    'ULT': (SRC, ['ult50cten_en408']),
    'ULX': (SRC, ['ulx50cten_en425']),
    'ULY': (SRC, ['uly40cten_en426']),
    'ULZ': (SRC, ['ulz40cten_en428']),
    # ---- 勾销缸 ----
    'USP': (SRC, ['usp32cten_en355', 'usp50cten_en437', 'usp50ucten_en168']),
    'USL': (SRC, ['usl50cten_en171']),
    'USR': (SRC, ['usr40cten_en436']),
    'USS': (SRC, ['uss50cten_en433', 'uss63cten_en434']),
    'USSL': (SRC, ['ussl50cten_en358']),
    # ---- 翻转缸 ----
    'UAGP': (SRC, ['uagp075wcten_en580', 'uagp075bcten_en181', 'uagp080wcten_en174',
                   'uagp080wcten_en573', 'uagp120bcten_en175', 'uagp120wcten_en574',
                   'uagp150bcten_en178', 'uagp150wcten_en577', 'uagp155wcten_en583',
                   'uagp170bnken_en176', 'uagp170wcten_en575', 'uagp210bcten_en182',
                   'uagp210wcten_en581', 'uagp215bcten_en373', 'uagp215wcten_en584',
                   'uagp300bcten_en177', 'uagp300wcten_en576', 'uagp350bcten_en183',
                   'uagp350wcten_en582', 'uagp355bcten_en374', 'uagp355wcten_en585',
                   'uagp600bcten_en179', 'uagp600wcten_en578', 'uagp605bcten_en180',
                   'uagp605wcten_en579']),
    # LAGP 官方已改名 UAGP，图册在选型表/spanner（老版）
    'LAGP': (SPANNER, ['lagp075_en_c181', 'lagp120_en_c175', 'lagp150_en_c178',
                       'lagp155_en_c350', 'lagp170_en_c176', 'lagp210_en_c182',
                       'lagp215nwrev-00-0119en_en373', 'lagp300_en_c177']),
    # ---- 电缸 ----
    'LAE': (SRC, ['lae140_en_c220', 'lae225_en_c221', 'lae350_en_c222']),
}


def classify(page):
    t = page.get_text().upper()
    if 'ACCESSORIES' in t or 'CLAMPING ARMS' in t or 'BLADE ARMS' in t:
        return 'arm'
    if 'DIMENSIONS' in t:
        return 'dim'
    if 'CHARACTERISTICS' in t or 'CODIFICATION' in t:
        return 'char'
    return 'other'


def main():
    os.makedirs(DST, exist_ok=True)
    missing = []
    for series, (srcdir, pdfs) in SPECS.items():
        d = os.path.join(DST, series)
        if CLEAN and os.path.isdir(d):
            shutil.rmtree(d)
        os.makedirs(d, exist_ok=True)

        docs, all_pages = [], []
        for name in pdfs:
            p = os.path.join(srcdir, name + '.pdf')
            if not os.path.exists(p):
                missing.append(f'{series}: {name}')
                continue
            doc = fitz.open(p)
            docs.append((name, doc))
            for i in range(doc.page_count):
                all_pages.append((doc, i, classify(doc[i])))
        if not all_pages:
            print(f'[{series}] 无可用源 PDF')
            continue

        order = {'char': 0, 'dim': 1, 'arm': 2, 'other': 3}
        all_pages.sort(key=lambda x: order[x[2]])
        n = 1
        for doc, i, _k in all_pages:
            # 彩色输出 (官方图册本身是彩色的; 早期版本误用 csGRAY 转灰度,
            # 体积约翻倍但相对整个数据包可忽略)
            doc[i].get_pixmap(dpi=DPI).save(
                os.path.join(d, f'p{n:02d}.png'))
            n += 1

        out_pdf = fitz.open()
        for _name, doc in docs:
            out_pdf.insert_pdf(doc)
        outp = os.path.join(d, series + '_manual.pdf')
        out_pdf.save(outp, garbage=3, deflate=True)
        out_pdf.close()
        print(f'[{series}] {len(all_pages)} 页 / {len(pdfs)} PDF -> '
              f'{os.path.getsize(outp)/1024:.0f}KB')

    if missing:
        print('\n源缺失:')
        for m in missing:
            print('  ', m)
    print('完成')


if __name__ == '__main__':
    main()