tiku/tiku-server/app/services/docx_builder.py

"""docx 合并 + 导出排版模块"""

import re
import shutil
import tempfile
from datetime import datetime
from pathlib import Path

from docx import Document
from docx.enum.text import WD_ALIGN_PARAGRAPH
from docx.oxml import OxmlElement
from docx.oxml.ns import qn
from docx.shared import Pt, Cm


TYPE_ORDER = ["单选题", "多选题", "填空题", "解答题"]
TYPE_HEADERS = {
    "单选题": "一、单选题",
    "多选题": "二、多选题",
    "填空题": "三、填空题",
    "解答题": "四、解答题",
}


def build_export_docx(
    problem_ids: list[str],
    problems_dir: str | Path,
    title: str = "试卷",
    include_solution: bool = True,
) -> Path:
    """从题目 ID 列表合并生成导出 docx

    Args:
        problem_ids: 题目 ID 列表（已按题型分组、按难度排序）
        problems_dir: 单题 docx 存储目录
        title: 试卷标题
        include_solution: True = 教师版, False = 学生版

    Returns: 生成的 docx 文件路径
    """
    problems_dir = Path(problems_dir)
    doc = Document()

    # 设置默认样式
    style = doc.styles["Normal"]
    font = style.font
    font.name = "宋体"
    font.size = Pt(10.5)
    style.element.rPr.rFonts.set(qn('w:eastAsia'), '宋体')

    # 标题
    title_para = doc.add_paragraph()
    title_para.alignment = WD_ALIGN_PARAGRAPH.CENTER
    title_run = title_para.add_run(title)
    title_run.bold = True
    title_run.font.size = Pt(14)
    title_run.font.name = "宋体"
    title_run._element.rPr.rFonts.set(qn('w:eastAsia'), '宋体')

    subtitle = "教师版" if include_solution else "学生版"
    sub_para = doc.add_paragraph()
    sub_para.alignment = WD_ALIGN_PARAGRAPH.CENTER
    sub_run = sub_para.add_run(subtitle)
    sub_run.font.size = Pt(10.5)

    doc.add_paragraph()  # 空行

    # 从数据库获取题目并按题型分组
    from app.database import get_connection
    conn = get_connection(problems_dir.parent / "bank.db")
    try:
        problems = []
        for pid in problem_ids:
            row = conn.execute("SELECT * FROM problems WHERE id = ?", (pid,)).fetchone()
            if row:
                problems.append(dict(row))

        # 按题型分组
        groups: dict[str, list] = {}
        for p in problems:
            qt = p.get("question_type", "")
            if qt not in groups:
                groups[qt] = []
            groups[qt].append(p)

        # 按题型排序
        sorted_groups = [(qt, groups[qt]) for qt in TYPE_ORDER if qt in groups]
        # 加上未知题型
        for qt, items in groups.items():
            if qt not in TYPE_ORDER:
                sorted_groups.append((qt, items))

        question_num = 1
        for qt, items in sorted_groups:
            # 题型标题
            header_text = TYPE_HEADERS.get(qt, qt)
            header_para = doc.add_paragraph()
            header_run = header_para.add_run(header_text)
            header_run.bold = True
            header_run.font.size = Pt(12)
            header_run.font.name = "宋体"
            header_run._element.rPr.rFonts.set(qn('w:eastAsia'), '宋体')

            for item in items:
                pid = item["id"]
                docx_path = problems_dir / f"{pid}.docx"
                if not docx_path.exists():
                    continue

                # 复制该题的内容
                _copy_question(doc, docx_path, question_num, include_solution)
                question_num += 1

    finally:
        conn.close()

    # 添加 Word 自动编号
    _apply_auto_numbering(doc)

    # 保存
    papers_dir = problems_dir.parent / "papers"
    papers_dir.mkdir(parents=True, exist_ok=True)
    ts = datetime.now().strftime("%Y%m%d_%H%M%S")
    suffix = "教师版" if include_solution else "学生版"
    safe_title = title.replace("/", "_").replace(" ", "_")
    output_path = papers_dir / f"{safe_title}_{ts}_{suffix}.docx"
    doc.save(str(output_path))

    return output_path


def _copy_question(doc: Document, source_path: Path, num: int, include_solution: bool):
    """复制题目的段落到目标 docx"""
    source = Document(str(source_path))
    in_answer = False

    for para in source.paragraphs:
        text = para.text.strip()
        if not text:
            # 复制空段落（保留格式间距）
            p = doc.add_paragraph()
            continue

        # 检测答案区
        if text.startswith("【答案】"):
            in_answer = True
            if not include_solution:
                break

        if in_answer and not include_solution:
            break

        # 复制段落
        p = doc.add_paragraph()
        for run in para.runs:
            r = p.add_run(run.text)
            r.bold = run.bold
            r.italic = run.italic
            r.font.name = "宋体"
            r._element.rPr.rFonts.set(qn('w:eastAsia'), '宋体')
            r.font.size = Pt(10.5)

        # 保留段落对齐
        if para.alignment:
            p.alignment = para.alignment


def _apply_auto_numbering(doc: Document):
    """对题号段落应用 Word 自动编号"""
    # 识别题型标题后的题号段落
    numbering_id = 1
    in_section = False

    for para in doc.paragraphs:
        text = para.text.strip()

        # 检测题型标题
        is_header = False
        for qt, header in TYPE_HEADERS.items():
            if header in text:
                is_header = True
                in_section = True
                numbering_id += 1
                break

        if is_header:
            continue

        # 对题号段落加编号
        if in_section and re.match(r'\d+[．.]', text):
            _add_numbering(para, numbering_id)


def _add_numbering(para, num_id: int):
    """给段落添加 Word 自动编号"""
    pPr = para._element.get_or_add_pPr()
    numPr = OxmlElement('w:numPr')

    ilvl = OxmlElement('w:ilvl')
    ilvl.set(qn('w:val'), '0')
    numPr.append(ilvl)

    numId = OxmlElement('w:numId')
    numId.set(qn('w:val'), str(num_id))
    numPr.append(numId)

    pPr.insert(0, numPr)