vent_agent_3/utils/pdf_builder.py

import re
from reportlab.lib.pagesizes import A4
from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer, Table, TableStyle
from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle
from reportlab.lib import colors
from reportlab.pdfbase import pdfmetrics
from reportlab.pdfbase.cidfonts import UnicodeCIDFont
from config.settings import SERVER_HOST, REPORT_OUTPUT_DIR

# 注册中文字体
pdfmetrics.registerFont(UnicodeCIDFont('STSong-Light'))
FONT_CN = 'STSong-Light'

TABLE_BASE_STYLE = [
    ('BACKGROUND', (0, 0), (-1, 0), colors.lightgrey),
    ('ALIGN', (0, 0), (-1, -1), 'LEFT'),
    ('VALIGN', (0, 0), (-1, -1), 'MIDDLE'),
    ('TOPPADDING', (0, 0), (-1, -1), 6),
    ('BOTTOMPADDING', (0, 0), (-1, -1), 6),
    ('LEFTPADDING', (0, 0), (-1, -1), 4),
    ('RIGHTPADDING', (0, 0), (-1, -1), 4),
    ('GRID', (0, 0), (-1, -1), 1, colors.black),
]


def convert_bold(text: str) -> str:
    return re.sub(r'\*\*(.*?)\*\*', r'<b>\1</b>', text)


def markdown_to_pdf(md_content: str, save_path: str) -> None:
    """Markdown转PDF，支持四级标题、加粗、表格"""
    doc = SimpleDocTemplate(
        save_path,
        pagesize=A4,
        rightMargin=30, leftMargin=30, topMargin=30, bottomMargin=30
    )
    story = []
    styles = getSampleStyleSheet()

    # 自定义样式
    style_h1 = ParagraphStyle('CustomH1', parent=styles['Heading1'], fontSize=16, spaceAfter=12, fontName=FONT_CN)
    style_h2 = ParagraphStyle('CustomH2', parent=styles['Heading2'], fontSize=14, spaceAfter=10, fontName=FONT_CN)
    style_h3 = ParagraphStyle('CustomH3', parent=styles['Heading3'], fontSize=12, spaceAfter=8, fontName=FONT_CN)
    style_h4 = ParagraphStyle('CustomH4', parent=styles['Heading3'], fontSize=11, spaceAfter=6, fontName=FONT_CN)
    style_normal = ParagraphStyle('CustomNormal', parent=styles['Normal'], fontSize=11, leading=18, fontName=FONT_CN)
    table_text_style = ParagraphStyle('TableText', parent=styles['Normal'], fontSize=10, fontName=FONT_CN, leading=14)

    lines = md_content.splitlines()
    table_rows = []
    in_table = False

    for raw_line in lines:
        line = raw_line.strip()
        if not line:
            continue
        # 四级标题
        if line.startswith("#### "):
            txt = convert_bold(line.lstrip("#### ").strip())
            story.append(Paragraph(txt, style_h4))
            story.append(Spacer(1, 4))
            continue
        elif line.startswith("### "):
            txt = convert_bold(line.lstrip("### ").strip())
            story.append(Paragraph(txt, style_h3))
            story.append(Spacer(1, 4))
            continue
        elif line.startswith("## "):
            txt = convert_bold(line.lstrip("## ").strip())
            story.append(Paragraph(txt, style_h2))
            story.append(Spacer(1, 5))
            continue
        elif line.startswith("# "):
            txt = convert_bold(line.lstrip("# ").strip())
            story.append(Paragraph(txt, style_h1))
            story.append(Spacer(1, 6))
            continue
        # 表格分隔符
        if all(c in "-| " for c in line):
            continue
        # 表格行
        if line.startswith("|") and line.endswith("|"):
            in_table = True
            cells = [cell.strip() for cell in line.split("|")[1:-1]]
            paras = [Paragraph(convert_bold(c), table_text_style) for c in cells]
            table_rows.append(paras)
            continue
        # 结束表格渲染
        if in_table:
            in_table = False
            if table_rows:
                col_cnt = len(table_rows[0])
                col_w = [530 / col_cnt] * col_cnt
                tbl = Table(table_rows, colWidths=col_w, repeatRows=1)
                tbl.setStyle(TableStyle(TABLE_BASE_STYLE))
                story.append(tbl)
                story.append(Spacer(1, 10))
            table_rows = []
        # 普通正文
        fmt_line = convert_bold(line)
        story.append(Paragraph(fmt_line, style_normal))
        story.append(Spacer(1, 3))
    # 剩余表格
    if in_table and table_rows:
        col_cnt = len(table_rows[0])
        col_w = [530 / col_cnt] * col_cnt
        tbl = Table(table_rows, colWidths=col_w, repeatRows=1)
        tbl.setStyle(TableStyle(TABLE_BASE_STYLE))
        story.append(tbl)
        story.append(Spacer(1, 10))
    doc.build(story)


def get_pdf_download_url(pdf_filename: str) -> str:
    return f"{SERVER_HOST}/static/report_pdf/{pdf_filename}"