326 lines
12 KiB
Python
326 lines
12 KiB
Python
"""v1.1 .docx 字体乱码回归测试。
|
||
|
||
问题:生成的 .docx 在线查看正常,下载保存后字体乱掉。
|
||
原因:StdlibDocxWriter 的 <w:rPrDefault><w:rPr><w:rFonts ...></w:rPr></w:rPrDefault>
|
||
缺 w:eastAsia 属性 + w:hint + w:lang。
|
||
Word 在解析时对没有 eastAsia 属性的 CJK 字符使用系统默认字体,
|
||
当系统字体集不规范或字体缺失时,CJK 字符会被替换为非预期字体(乱码)。
|
||
|
||
修复后检查项:
|
||
1. <w:rPrDefault> 中 <w:rFonts> 必须含 w:ascii/w:hAnsi/w:eastAsia/w:cs 四个属性
|
||
2. 必须有 w:hint="eastAsia" 属性
|
||
3. 必含 <w:lang> 显式声明中文
|
||
4. 用 universal 字体(宋体/Times New Roman),不依赖特定字体存在
|
||
"""
|
||
from __future__ import annotations
|
||
|
||
import os
|
||
import sys
|
||
import re
|
||
import tempfile
|
||
import zipfile
|
||
from pathlib import Path
|
||
|
||
SKILL_DIR = Path(__file__).resolve().parent.parent
|
||
sys.path.insert(0, str(SKILL_DIR / "scripts"))
|
||
|
||
|
||
# ============================================================
|
||
# 测试
|
||
# ============================================================
|
||
|
||
def test_docx_has_eastasia_font():
|
||
"""生成的 docx 必须含 w:eastAsia 属性(核心 bug 修复)。"""
|
||
from report_stdlib_fallbacks import StdlibDocxWriter
|
||
|
||
data = {
|
||
"summary": {"stock_name": "测试", "periodic_count": 1, "broker_count": 1, "announcement_count": 1, "has_buy_rating": True},
|
||
"periodic_reports": [],
|
||
"broker_reports": [],
|
||
"announcements": [],
|
||
}
|
||
|
||
with tempfile.TemporaryDirectory() as tmp:
|
||
output = os.path.join(tmp, "test.docx")
|
||
writer = StdlibDocxWriter()
|
||
writer.export_comprehensive_report(data, "000001", output)
|
||
|
||
# 读取 styles.xml
|
||
with zipfile.ZipFile(output, "r") as zf:
|
||
styles = zf.read("word/styles.xml").decode("utf-8")
|
||
|
||
# 核心检查:rFonts 含 eastAsia
|
||
assert 'w:eastAsia=' in styles, "缺少 w:eastAsia 字体定义(CJK 渲染关键)"
|
||
print(" ✓ 含 w:eastAsia 字体定义")
|
||
|
||
|
||
def test_docx_has_complete_font_quad():
|
||
"""rFonts 必须含 4 个字体属性:ascii/hAnsi/eastAsia/cs。"""
|
||
from report_stdlib_fallbacks import StdlibDocxWriter
|
||
|
||
data = {"summary": {}, "periodic_reports": [], "broker_reports": [], "announcements": []}
|
||
|
||
with tempfile.TemporaryDirectory() as tmp:
|
||
output = os.path.join(tmp, "test.docx")
|
||
StdlibDocxWriter().export_comprehensive_report(data, "T", output)
|
||
|
||
with zipfile.ZipFile(output, "r") as zf:
|
||
styles = zf.read("word/styles.xml").decode("utf-8")
|
||
|
||
# 提取 rFonts 那一行(默认的)
|
||
match = re.search(r'<w:rFonts\s+([^/]*?)/>', styles)
|
||
assert match, "未找到 <w:rFonts>"
|
||
attrs = match.group(1)
|
||
|
||
required = ["w:ascii", "w:hAnsi", "w:eastAsia", "w:cs"]
|
||
for attr in required:
|
||
assert attr in attrs, f"缺少字体属性 {attr}"
|
||
print(f" ✓ 4 个字体属性齐全:{required}")
|
||
|
||
|
||
def test_docx_has_hint_eastasia():
|
||
"""必须有 w:hint="eastAsia" 告诉 Word 优先按东亚字体处理。"""
|
||
from report_stdlib_fallbacks import StdlibDocxWriter
|
||
|
||
data = {"summary": {}, "periodic_reports": [], "broker_reports": [], "announcements": []}
|
||
|
||
with tempfile.TemporaryDirectory() as tmp:
|
||
output = os.path.join(tmp, "test.docx")
|
||
StdlibDocxWriter().export_comprehensive_report(data, "T", output)
|
||
|
||
with zipfile.ZipFile(output, "r") as zf:
|
||
styles = zf.read("word/styles.xml").decode("utf-8")
|
||
|
||
assert 'w:hint="eastAsia"' in styles, "缺少 w:hint='eastAsia'"
|
||
print(" ✓ 含 w:hint='eastAsia'")
|
||
|
||
|
||
def test_docx_has_language_tag():
|
||
"""必须有 <w:lang> 显式声明中文。"""
|
||
from report_stdlib_fallbacks import StdlibDocxWriter
|
||
|
||
data = {"summary": {}, "periodic_reports": [], "broker_reports": [], "announcements": []}
|
||
|
||
with tempfile.TemporaryDirectory() as tmp:
|
||
output = os.path.join(tmp, "test.docx")
|
||
StdlibDocxWriter().export_comprehensive_report(data, "T", output)
|
||
|
||
with zipfile.ZipFile(output, "r") as zf:
|
||
styles = zf.read("word/styles.xml").decode("utf-8")
|
||
|
||
assert "<w:lang" in styles, "缺少 <w:lang>"
|
||
# 必须声明 eastAsia 为中文(zh-CN/zh-TW/zh-HK 任一)
|
||
match = re.search(r'<w:lang[^>]+w:eastAsia="([^"]+)"', styles)
|
||
assert match and "zh" in match.group(1).lower(), f"eastAsia 语言应含 zh,实际: {match.group(1) if match else 'NOT FOUND'}"
|
||
print(f" ✓ <w:lang> 中文标记:{match.group(1)}")
|
||
|
||
|
||
def test_docx_uses_universal_fonts():
|
||
"""字体选择应该是 universal 字体(不依赖特定系统)。"""
|
||
from report_stdlib_fallbacks import StdlibDocxWriter
|
||
|
||
data = {"summary": {}, "periodic_reports": [], "broker_reports": [], "announcements": []}
|
||
|
||
with tempfile.TemporaryDirectory() as tmp:
|
||
output = os.path.join(tmp, "test.docx")
|
||
StdlibDocxWriter().export_comprehensive_report(data, "T", output)
|
||
|
||
with zipfile.ZipFile(output, "r") as zf:
|
||
styles = zf.read("word/styles.xml").decode("utf-8")
|
||
|
||
# 不应依赖 "微软雅黑"(Mac/Linux 没有)
|
||
# 应是 宋体 (SimSun) 或 宋体 unicode 字符串
|
||
# 微软雅黑 is risky on Mac/Linux
|
||
if "微软雅黑" in styles:
|
||
print(" ⚠️ 含微软雅黑,在 Mac/Linux 用户可能无此字体")
|
||
else:
|
||
print(" ✓ 不依赖中文专属字体(微软雅黑)")
|
||
|
||
# 应该含 宋体 或 SimSun
|
||
assert "宋体" in styles or "SimSun" in styles, "eastAsia 应使用 universal 字体(宋体/SimSun)"
|
||
print(" ✓ 使用 universal 字体(宋体/SimSun)")
|
||
|
||
|
||
def test_docx_validation_lifecycle():
|
||
"""完整生成 + 解析 + 检查生命周期。"""
|
||
from report_stdlib_fallbacks import StdlibDocxWriter
|
||
|
||
data = {
|
||
"summary": {"stock_name": "中文测试股", "periodic_count": 1, "broker_count": 1, "announcement_count": 1, "has_buy_rating": True},
|
||
"periodic_reports": [{"title": "测试报告", "publish_date": "2026-07-28", "url": "https://example.com"}],
|
||
"broker_reports": [{"title": "买入评级", "publish_date": "2026-07-28", "broker_name": "测试券商", "analyst": "分析师", "rating": "买入", "target_price": 100.0}],
|
||
"announcements": [{"title": "测试公告", "publish_date": "2026-07-28", "announcement_type": "分红", "url": ""}],
|
||
}
|
||
|
||
with tempfile.TemporaryDirectory() as tmp:
|
||
output = os.path.join(tmp, "test.docx")
|
||
StdlibDocxWriter().export_comprehensive_report(data, "000001", output)
|
||
|
||
# 用 python-docx 解析验证
|
||
try:
|
||
from docx import Document
|
||
doc = Document(output)
|
||
# 验证文本正确写入
|
||
text_found = False
|
||
for p in doc.paragraphs:
|
||
if "中文测试股" in p.text:
|
||
text_found = True
|
||
break
|
||
for tbl in doc.tables:
|
||
for row in tbl.rows:
|
||
for cell in row.cells:
|
||
if "中文测试股" in cell.text:
|
||
text_found = True
|
||
break
|
||
assert text_found, "中文内容未正确写入"
|
||
print(" ✓ python-docx 可正确解析,中文写入无误")
|
||
except ImportError:
|
||
print(" ⊘ python-docx 未安装,跳过解析验证")
|
||
pass
|
||
|
||
# 文件应 < 50KB(典型报告大小)
|
||
size = os.path.getsize(output)
|
||
assert size < 100_000, f"文件过大: {size} 字节"
|
||
print(f" ✓ 文件大小合理: {size} 字节")
|
||
|
||
|
||
# ============================================================
|
||
# PPTX 字体检查(v1.1 新增)
|
||
# ============================================================
|
||
|
||
def _read_pptx_theme(path: str) -> str:
|
||
"""从 pptx 文件读取 theme1.xml 的内容。"""
|
||
import zipfile
|
||
with zipfile.ZipFile(path, "r") as zf:
|
||
for n in zf.namelist():
|
||
if n.endswith("theme/theme1.xml"):
|
||
return zf.read(n).decode("utf-8")
|
||
return ""
|
||
|
||
|
||
def test_pptx_theme_has_ea_font():
|
||
"""PPTX theme 必须声明东亚字体(v1.1 修复:之前是空字符串)。"""
|
||
from report_stdlib_fallbacks import StdlibPptxWriter
|
||
|
||
data = {"summary": {}, "periodic_reports": [], "broker_reports": [], "announcements": []}
|
||
|
||
with tempfile.TemporaryDirectory() as tmp:
|
||
output = os.path.join(tmp, "test.pptx")
|
||
StdlibPptxWriter().export_comprehensive_report(data, "T", output)
|
||
|
||
theme = _read_pptx_theme(output)
|
||
# 必须含 ea typeface 且非空
|
||
import re
|
||
ea_matches = re.findall(r'<a:ea\s+typeface="([^"]*)"', theme)
|
||
assert ea_matches, "<a:ea typeface=...> 应在 theme 中声明"
|
||
for ea in ea_matches:
|
||
assert ea.strip(), f"<a:ea typeface> 不能为空字符串(v1.1 修复)"
|
||
print(f" ✓ PPTX theme 含东亚字体:{ea_matches}")
|
||
|
||
|
||
def test_pptx_theme_uses_universal_fonts():
|
||
"""PPTX theme 应使用 universal 字体(不依赖 微软雅黑)。"""
|
||
from report_stdlib_fallbacks import StdlibPptxWriter
|
||
|
||
data = {"summary": {}, "periodic_reports": [], "broker_reports": [], "announcements": []}
|
||
|
||
with tempfile.TemporaryDirectory() as tmp:
|
||
output = os.path.join(tmp, "test.pptx")
|
||
StdlibPptxWriter().export_comprehensive_report(data, "T", output)
|
||
theme = _read_pptx_theme(output)
|
||
|
||
if "微软雅黑" in theme:
|
||
print(" ⚠️ PPTX theme 含微软雅黑")
|
||
else:
|
||
print(" ✓ PPTX theme 不依赖 微软雅黑")
|
||
|
||
|
||
# ============================================================
|
||
# XLSX 字体检查(v1.1 新增)
|
||
# ============================================================
|
||
|
||
def _read_xlsx_styles(path: str) -> str:
|
||
"""从 xlsx 文件读取 styles.xml 内容。"""
|
||
import zipfile
|
||
with zipfile.ZipFile(path, "r") as zf:
|
||
return zf.read("xl/styles.xml").decode("utf-8")
|
||
|
||
|
||
def test_xlsx_uses_universal_font():
|
||
"""XLSX style 必须使用 universal 字体(v1.1 修复:之前用 微软雅黑)。"""
|
||
from report_stdlib_fallbacks import StdlibXlsxWriter
|
||
|
||
data = {"summary": {}, "periodic_reports": [], "broker_reports": [], "announcements": []}
|
||
|
||
with tempfile.TemporaryDirectory() as tmp:
|
||
output = os.path.join(tmp, "test.xlsx")
|
||
StdlibXlsxWriter().export_comprehensive_report(data, "T", output)
|
||
|
||
styles = _read_xlsx_styles(output)
|
||
# 不应该用 微软雅黑(Mac/Linux 没有)
|
||
assert "微软雅黑" not in styles, "XLSX 不应依赖 微软雅黑"
|
||
# 应该用 宋体 或其他 universal 字体
|
||
assert "宋体" in styles or "SimSun" in styles, "XLSX 应使用 universal 字体"
|
||
print(f" ✓ XLSX 使用 universal 字体(宋体/SimSun)")
|
||
|
||
|
||
def test_xlsx_openpyxl_path_uses_universal_font():
|
||
"""openpyxl 路径下生成的 xlsx 也用 universal 字体。"""
|
||
try:
|
||
from openpyxl import load_workbook
|
||
except ImportError:
|
||
print(" ⊘ openpyxl 未安装,跳过")
|
||
return
|
||
|
||
# 写一个使用 宋体 字体的工作簿,验证能 round-trip
|
||
from openpyxl import Workbook
|
||
from openpyxl.styles import Font
|
||
|
||
wb = Workbook()
|
||
ws = wb.active
|
||
ws['A1'] = "中文测试"
|
||
ws['A1'].font = Font(name='宋体')
|
||
with tempfile.NamedTemporaryFile(suffix=".xlsx", delete=False) as f:
|
||
path = f.name
|
||
try:
|
||
wb.save(path)
|
||
|
||
# 重新加载看是否能识别中文字体
|
||
wb2 = load_workbook(path)
|
||
font = wb2.active['A1'].font
|
||
assert font.name == "宋体", f"字体应为宋体,实际: {font.name}"
|
||
print(f" ✓ openpyxl 路径字体声明一致:{font.name}")
|
||
finally:
|
||
os.remove(path)
|
||
|
||
|
||
# ============================================================
|
||
# 运行所有
|
||
# ============================================================
|
||
|
||
def main():
|
||
tests = [
|
||
(name, obj)
|
||
for name, obj in globals().items()
|
||
if name.startswith("test_") and callable(obj)
|
||
]
|
||
passed = 0
|
||
failed = 0
|
||
for name, fn in tests:
|
||
try:
|
||
fn()
|
||
passed += 1
|
||
print(f" ✓ {name}")
|
||
except Exception as e:
|
||
print(f" ✗ {name}: {e}")
|
||
import traceback
|
||
traceback.print_exc()
|
||
failed += 1
|
||
|
||
print(f"\n{'='*60}\n字体修复测试: {passed} 通过, {failed} 失败\n{'='*60}")
|
||
return 0 if failed == 0 else 1
|
||
|
||
|
||
if __name__ == "__main__":
|
||
sys.exit(main())
|