docx/core-properties-unicode

Metadata

XPath assertions

Render assertions

No render assertions declared.

Generator source

scripts/gen_core_properties_unicode.py

#!/usr/bin/env python3
"""Generate ``fixtures/docx/core-properties-unicode.docx``.

Exercises the non-ASCII round-trip path for ``dc:title`` and ``dc:creator``.
The fixture value combines CJK ideographs, a short Arabic (right-to-left)
fragment, and a multi-codepoint ZWJ family emoji to stress-test any
intermediate normalisation the authoring library might do.
"""

from __future__ import annotations

import datetime as _dt
from pathlib import Path

from docx import Document

_REPO_ROOT = Path(__file__).resolve().parent.parent
_OUT = _REPO_ROOT / "fixtures" / "docx" / "core-properties-unicode.docx"

_PINNED_CREATED = _dt.datetime(2026, 1, 1, 0, 0, 0, tzinfo=_dt.timezone.utc)
_PINNED_MODIFIED = _dt.datetime(2026, 1, 2, 0, 0, 0, tzinfo=_dt.timezone.utc)


def main() -> int:
    doc = Document()
    cp = doc.core_properties
    cp.created = _PINNED_CREATED
    cp.modified = _PINNED_MODIFIED
    cp.title = "测试标题 مرحبا \U0001f468‍\U0001f469‍\U0001f467"
    cp.author = "作者 كاتب \U0001f4dd"
    doc.add_paragraph("Body content; unicode metadata is the feature under test.")
    _OUT.parent.mkdir(parents=True, exist_ok=True)
    doc.save(_OUT, reproducible=True)
    print(_OUT)
    return 0


if __name__ == "__main__":
    raise SystemExit(main())

Manifest (expanded)

{
  "$schema": "../manifest.schema.json",
  "id": "docx/core-properties-unicode",
  "title": "Core properties: non-ASCII round-trip (CJK + RTL + emoji)",
  "format": "docx",
  "category": "metadata",
  "summary": "Proves that ``dc:title`` and ``dc:creator`` round-trip bytes with non-ASCII payloads covering CJK ideographs, right-to-left Arabic, and a multi-codepoint emoji.",
  "spec": {
    "source": "ecma-376-5-part-2",
    "clause": "11.1",
    "notes": "Core-properties carry Unicode by virtue of the XML serialisation. The risk area is not the XML spec but whatever intermediate normalisation, decoding, or quoting the authoring library performs between the Python string and the emitted ``<dc:title>`` / ``<dc:creator>``. The fixture value combines a Chinese title, a short Arabic fragment, and the multi-codepoint family emoji U+1F468 U+200D U+1F469 U+200D U+1F467 to stress-test all three paths at once."
  },
  "fixtures": {
    "machine": "docx/core-properties-unicode"
  },
  "generator": {
    "python": "scripts/gen_core_properties_unicode.py"
  },
  "assertions": [
    {
      "id": "unicode-title-value",
      "part": "docProps/core.xml",
      "namespaces": {
        "cp": "http://schemas.openxmlformats.org/package/2006/metadata/core-properties",
        "dc": "http://purl.org/dc/elements/1.1/"
      },
      "xpath": "//dc:title",
      "must": "equal",
      "value": "测试标题 مرحبا 👨‍👩‍👧",
      "description": "The <dc:title> element must preserve CJK, Arabic and emoji bytes verbatim."
    },
    {
      "id": "unicode-creator-value",
      "part": "docProps/core.xml",
      "namespaces": {
        "cp": "http://schemas.openxmlformats.org/package/2006/metadata/core-properties",
        "dc": "http://purl.org/dc/elements/1.1/"
      },
      "xpath": "//dc:creator",
      "must": "equal",
      "value": "作者 كاتب 📝",
      "description": "The <dc:creator> element must preserve CJK, Arabic and emoji bytes verbatim."
    }
  ]
}