| 123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350 |
- # -*- coding: utf-8 -*-
- """
- 脱敏 Markdown → 带高亮的 Word(.docx)→ Word COM 导出 PDF
- 替换掉浏览器打印路线(沙箱下 Chrome/Edge headless 不产出文件)
- """
- import io, os, re, sys, subprocess
- sys.stdout.reconfigure(encoding='utf-8')
- from docx import Document
- from docx.shared import Pt, RGBColor, Cm
- from docx.enum.text import WD_ALIGN_PARAGRAPH, WD_COLOR_INDEX
- from docx.enum.table import WD_TABLE_ALIGNMENT
- from docx.enum.section import WD_ORIENT
- from docx.oxml.ns import qn
- from docx.oxml import OxmlElement
- BASE = r'E:/cfc/docs/参考资料'
- OUT = os.path.join(BASE, '脱敏交付')
- os.makedirs(OUT, exist_ok=True)
- PURPLE = RGBColor(0x5B, 0x3E, 0x96)
- DARK = RGBColor(0x25, 0x23, 0x2B)
- GREY = RGBColor(0x5A, 0x54, 0x66)
- FONT_CN = '微软雅黑'
- FONT_EN = 'Segoe UI'
- MASK = [
- ('张学红', 'L 女士'),
- ('560822024', 'A-2401'),
- # 清理带编号/姓名的原始文件名,避免二次可识别
- ('`A-2401-L 女士.json`', '报告解析 JSON'),
- ('`A-2401-L 女士.pdf`', '原始检测报告 PDF'),
- ('`A-2401-L 女士`', '报告解析文件'),
- ]
- KEY_PHRASES = [
- '她不是吃得少,是吃得很讲究——但讲究的方向恰好全在压制铁吸收',
- '抓三件事——拆开早餐的铁吸收抑制',
- '她的肠道生态底子是好的,问题高度集中,可干预性很强',
- '这不是"菌群崩坏",是"关键岗位缺人"',
- '铁被早餐堵死了、双歧缺粮塌方了、早绝经让骨与雌激素两个窗口提前了',
- '不需要补乳杆菌',
- '不必补乳杆菌',
- '长双歧杆菌为第一靶点',
- '外源益生菌的**定植率很低**',
- '益生菌**点火**,瓜尔豆纤维**添柴**',
- '不能把"运动少"归因为懒、缺动力或自律不足',
- '必须临床区分,不可自行归因',
- '不得表述为"她正在抑郁"',
- '42 岁绝经 = 早绝经',
- '二选一时取瓜尔豆纤维',
- '她不吃肉,所以红肉不是她的问题',
- '菊粉或瓜尔豆纤维(二选一)→ 推荐瓜尔豆纤维',
- '与铁剂错开 2 小时',
- ]
- # ---------------- 基础排版工具 ----------------
- def set_run_font(run, size=10.5, bold=False, color=DARK, italic=False):
- run.font.size = Pt(size)
- run.font.bold = bold
- run.font.italic = italic
- run.font.color.rgb = color
- run.font.name = FONT_EN
- rPr = run._element.get_or_add_rPr()
- rFonts = rPr.find(qn('w:rFonts'))
- if rFonts is None:
- rFonts = OxmlElement('w:rFonts'); rPr.append(rFonts)
- rFonts.set(qn('w:eastAsia'), FONT_CN)
- rFonts.set(qn('w:ascii'), FONT_EN)
- rFonts.set(qn('w:hAnsi'), FONT_EN)
- def shade_cell(cell, hexcolor):
- tcPr = cell._tc.get_or_add_tcPr()
- shd = OxmlElement('w:shd')
- shd.set(qn('w:val'), 'clear'); shd.set(qn('w:color'), 'auto')
- shd.set(qn('w:fill'), hexcolor)
- tcPr.append(shd)
- def add_bottom_border(par, color='5B3E96', sz=12):
- pPr = par._p.get_or_add_pPr()
- pbdr = OxmlElement('w:pBdr')
- bottom = OxmlElement('w:bottom')
- bottom.set(qn('w:val'), 'single'); bottom.set(qn('w:sz'), str(sz))
- bottom.set(qn('w:space'), '4'); bottom.set(qn('w:color'), color)
- pbdr.append(bottom); pPr.append(pbdr)
- def add_left_bar(par, color='7B5CB8'):
- pPr = par._p.get_or_add_pPr()
- pbdr = OxmlElement('w:pBdr')
- left = OxmlElement('w:left')
- left.set(qn('w:val'), 'single'); left.set(qn('w:sz'), '18')
- left.set(qn('w:space'), '8'); left.set(qn('w:color'), color)
- pbdr.append(left); pPr.append(pbdr)
- def shade_paragraph(par, hexcolor):
- pPr = par._p.get_or_add_pPr()
- shd = OxmlElement('w:shd')
- shd.set(qn('w:val'), 'clear'); shd.set(qn('w:color'), 'auto')
- shd.set(qn('w:fill'), hexcolor)
- pPr.append(shd)
- # ---------------- 行内标记解析:**粗体** ==高亮== ----------------
- TOKEN = re.compile(r'(\*\*.+?\*\*|==.+?==)')
- def add_inline(par, text, size=10.5, base_bold=False, color=DARK):
- """解析 **bold** 与 ==highlight==,支持「**粗体里嵌 ==高亮==」的嵌套"""
- for seg in TOKEN.split(text):
- if not seg:
- continue
- bold = base_bold
- if seg.startswith('**') and seg.endswith('**') and len(seg) > 4:
- seg = seg[2:-2]
- bold = True
- for sub in re.split(r'(==.+?==)', seg):
- if not sub:
- continue
- if sub.startswith('==') and sub.endswith('==') and len(sub) > 4:
- r = par.add_run(sub[2:-2])
- set_run_font(r, size=size, bold=True, color=RGBColor(0x1A, 0x16, 0x20))
- r.font.highlight_color = WD_COLOR_INDEX.YELLOW
- else:
- r = par.add_run(sub)
- set_run_font(r, size=size, bold=bold, color=color)
- # ---------------- Markdown 解析 ----------------
- def parse_table(lines):
- """返回 rows(含表头)"""
- rows = []
- for ln in lines:
- cells = [c.strip() for c in ln.strip().strip('|').split('|')]
- if all(re.fullmatch(r':?-{2,}:?', c) for c in cells if c):
- continue
- rows.append(cells)
- return rows
- def build_docx(md_path, out_docx, doc_title):
- raw = io.open(md_path, encoding='utf-8').read()
- for a, b in MASK:
- raw = raw.replace(a, b)
- # 去掉残留的反引号(docx 里会显示为字面 ` )
- raw = re.sub(r'`([^`\n]{1,80})`', r'\1', raw)
- # 抬头关键信息高亮(原文已被 ** 包裹,需单独处理)
- raw = raw.replace('**42 岁绝经(绝经 10 年,属早绝经)**',
- '==42 岁绝经(绝经 10 年,属早绝经)==')
- for p in KEY_PHRASES:
- if p in raw and ('==' + p + '==') not in raw:
- raw = raw.replace(p, '==' + p + '==', 1)
- # 脱敏声明插入到首个 --- 之前
- raw = raw.replace('\n---\n', '\n> **本文件已脱敏**:受检者姓名与报告编号均已做替换处理,仅保留归因所需的年龄、性别、生理阶段信息。\n\n---\n', 1)
- lines = raw.split('\n')
- doc = Document()
- sec = doc.sections[0]
- sec.page_width = Cm(21.0); sec.page_height = Cm(29.7)
- sec.left_margin = Cm(1.9); sec.right_margin = Cm(1.9)
- sec.top_margin = Cm(1.5); sec.bottom_margin = Cm(1.5)
- st = doc.styles['Normal']
- st.font.name = FONT_EN; st.font.size = Pt(10.5)
- st.element.rPr.rFonts.set(qn('w:eastAsia'), FONT_CN)
- i = 0
- n = len(lines)
- while i < n:
- ln = lines[i]
- s = ln.strip()
- # 空行
- if not s:
- i += 1; continue
- # 表格
- if s.startswith('|') and i + 1 < n and lines[i+1].strip().startswith('|'):
- block = []
- while i < n and lines[i].strip().startswith('|'):
- block.append(lines[i]); i += 1
- rows = [r for r in parse_table(block) if any(c for c in r)]
- if not rows:
- continue
- ncol = max(len(r) for r in rows)
- # --- 自适应列宽:按"该列最长内容"分配,幂压缩避免长文本列吃掉全部宽度 ---
- def wlen(s):
- return sum(2 if ord(ch) > 127 else 1 for ch in s)
- mxs = []
- for ci in range(ncol):
- mx = 1
- for row in rows:
- cell_txt = row[ci] if ci < len(row) else ''
- mx = max(mx, wlen(cell_txt))
- mxs.append(mx)
- weights = [m ** 0.7 for m in mxs]
- tot = sum(weights)
- ratios = [max(0.085, min(0.42, x / tot)) for x in weights]
- ssum = sum(ratios)
- ratios = [r / ssum for r in ratios]
- CONTENT_W = 17.2 # cm,页面宽 21 - 左右边距各 1.9
- col_w = [Cm(CONTENT_W * r) for r in ratios]
- tbl = doc.add_table(rows=0, cols=ncol)
- tbl.style = 'Table Grid'
- tbl.alignment = WD_TABLE_ALIGNMENT.CENTER
- tbl.autofit = False
- for ri, row in enumerate(rows):
- cells = tbl.add_row().cells
- for ci in range(ncol):
- txt = row[ci] if ci < len(row) else ''
- cell = cells[ci]
- cell.width = col_w[ci]
- cell.text = ''
- p = cell.paragraphs[0]
- p.paragraph_format.space_before = Pt(1.2)
- p.paragraph_format.space_after = Pt(1.2)
- p.paragraph_format.line_spacing = 1.18
- if ri == 0:
- shade_cell(cell, '5B3E96')
- add_inline(p, txt, size=8.8, base_bold=True,
- color=RGBColor(0xFF, 0xFF, 0xFF))
- else:
- if ri % 2 == 0:
- shade_cell(cell, 'FBF9FE')
- add_inline(p, txt, size=8.8)
- # 表头行跨页重复 + 行不跨页断裂
- try:
- trPr = tbl.rows[0]._tr.get_or_add_trPr()
- th = OxmlElement('w:tblHeader'); th.set(qn('w:val'), 'true')
- trPr.append(th)
- for r_ in tbl.rows:
- trPr_ = r_._tr.get_or_add_trPr()
- cs = OxmlElement('w:cantSplit'); cs.set(qn('w:val'), 'true')
- trPr_.append(cs)
- except Exception:
- pass
- sp = doc.add_paragraph()
- sp.paragraph_format.space_after = Pt(2)
- continue
- # 标题
- if s.startswith('#### '):
- p = doc.add_paragraph(); p.paragraph_format.space_before = Pt(8)
- p.paragraph_format.space_after = Pt(3)
- p.paragraph_format.keep_with_next = True
- add_inline(p, s[5:], size=11, base_bold=True, color=RGBColor(0x4A,0x3F,0x63))
- i += 1; continue
- if s.startswith('### '):
- p = doc.add_paragraph(); p.paragraph_format.space_before = Pt(11)
- p.paragraph_format.space_after = Pt(4)
- p.paragraph_format.keep_with_next = True
- add_inline(p, s[4:], size=11.8, base_bold=True, color=RGBColor(0x3F,0x35,0x55))
- add_bottom_border(p, color='CFC2E8', sz=6)
- i += 1; continue
- if s.startswith('## '):
- p = doc.add_paragraph(); p.paragraph_format.space_before = Pt(15)
- p.paragraph_format.space_after = Pt(6)
- p.paragraph_format.left_indent = Cm(0.18)
- p.paragraph_format.keep_with_next = True
- add_inline(p, s[3:], size=13.4, base_bold=True, color=PURPLE)
- add_left_bar(p); shade_paragraph(p, 'F7F4FC')
- i += 1; continue
- if s.startswith('# '):
- p = doc.add_paragraph(); p.paragraph_format.space_before = Pt(2)
- p.paragraph_format.space_after = Pt(8)
- add_inline(p, s[2:], size=18, base_bold=True, color=PURPLE)
- add_bottom_border(p, sz=16)
- i += 1; continue
- # 分隔线
- if re.fullmatch(r'-{3,}', s):
- p = doc.add_paragraph(); p.paragraph_format.space_before = Pt(2)
- p.paragraph_format.space_after = Pt(2)
- add_bottom_border(p, color='E2DCEE', sz=6)
- i += 1; continue
- # 引用块
- if s.startswith('>'):
- block = []
- while i < n and lines[i].strip().startswith('>'):
- block.append(lines[i].strip()[1:].strip()); i += 1
- for b in block:
- if not b:
- continue
- p = doc.add_paragraph()
- p.paragraph_format.left_indent = Cm(0.35)
- p.paragraph_format.space_before = Pt(3)
- p.paragraph_format.space_after = Pt(3)
- p.paragraph_format.line_spacing = 1.6
- txt = re.sub(r'^[-*]\s+', '· ', b)
- add_inline(p, txt, size=9.9, color=RGBColor(0x3C,0x34,0x50))
- add_left_bar(p, color='A98FE0'); shade_paragraph(p, 'FAF7FE')
- continue
- # 列表
- m = re.match(r'^(\s*)([-*]|\d+\.)\s+(.*)$', ln)
- if m:
- indent = len(m.group(1)) // 2
- marker = m.group(2)
- body = m.group(3)
- p = doc.add_paragraph()
- p.paragraph_format.left_indent = Cm(0.55 + indent * 0.45)
- p.paragraph_format.first_line_indent = Cm(-0.42)
- p.paragraph_format.space_before = Pt(2)
- p.paragraph_format.space_after = Pt(2)
- prefix = '· ' if marker in ('-', '*') else marker + ' '
- add_inline(p, prefix + body, size=10.4)
- i += 1; continue
- # 普通段落
- p = doc.add_paragraph()
- p.paragraph_format.space_before = Pt(3)
- p.paragraph_format.space_after = Pt(3)
- p.paragraph_format.line_spacing = 1.72
- p.alignment = WD_ALIGN_PARAGRAPH.JUSTIFY
- add_inline(p, s, size=10.5)
- i += 1
- doc.save(out_docx)
- return out_docx
- # ---------------- Word COM → PDF ----------------
- PS = r'''
- param([string]$DocxPath, [string]$PdfPath)
- $ErrorActionPreference = 'Stop'
- $word = New-Object -ComObject Word.Application
- $word.Visible = $false
- $word.DisplayAlerts = 0
- try {
- $doc = $word.Documents.Open($DocxPath, $false, $true)
- $doc.SaveAs([ref]$PdfPath, [ref]17)
- $doc.Close([ref]0)
- Write-Output "PDF_OK"
- } finally {
- $word.Quit()
- [System.Runtime.InteropServices.Marshal]::ReleaseComObject($word) | Out-Null
- }
- '''
- if __name__ == '__main__':
- ps_path = os.path.join(OUT, '_topdf.ps1')
- io.open(ps_path, 'w', encoding='utf-8-sig').write(PS)
- jobs = [
- ('560822024-张学红-归因分析-简版.md', 'L女士-归因分析-简版', 'L 女士 · 肠道菌群归因分析(简版)'),
- ('560822024-张学红-归因分析-完整版.md', 'L女士-归因分析-完整版', 'L 女士 · 肠道菌群归因分析(完整版)'),
- ]
- for src, stem, title in jobs:
- docx_path = os.path.join(OUT, stem + '.docx')
- build_docx(os.path.join(BASE, src), docx_path, title)
- print('docx:', os.path.basename(docx_path), f'{os.path.getsize(docx_path)/1024:.0f} KB')
|