docx/content-type-registration

Metadata

XPath assertions

Render assertions

No render assertions declared.

Generator source

scripts/gen_content_type_registration.py

#!/usr/bin/env python3
"""Generate ``fixtures/docx/content-type-registration.docx``.

A minimal empty document. The manifest assertions target the structural
OPC plumbing (``[Content_Types].xml`` and ``_rels/.rels``) rather than any
specific WordprocessingML content, so the body can be empty.

Manifest: ``features/docx/content-type-registration.json``
"""

from __future__ import annotations

from pathlib import Path

from docx import Document

_REPO_ROOT = Path(__file__).resolve().parent.parent
_OUT = _REPO_ROOT / "fixtures" / "docx" / "content-type-registration.docx"


def main() -> int:
    doc = Document()
    # one paragraph so document.xml has at least one w:p — some validators
    # flag an empty body, and this keeps the fixture opaque to unrelated
    # future tightening of the corpus' sanity checks.
    doc.add_paragraph("Content-type registration fixture.")
    _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/content-type-registration",
  "title": "Content-type registration (Default + Override)",
  "format": "docx",
  "category": "opc-packaging",
  "summary": "[Content_Types].xml must register the required Default entries for rels and xml extensions, plus an Override binding the main document part to the WordprocessingML document content type.",
  "spec": {
    "source": "ecma-376-5-part-2",
    "clause": "10.1",
    "element": "ct:Types",
    "notes": "ECMA-376 Part 2 (OPC) §10.1 — every part's content type is discovered by matching the part URI against Override entries first, then falling back to Default entries keyed on the part's URI extension. Every well-formed OOXML package therefore needs Default entries for at least the `rels` and `xml` extensions (because _rels/.rels and metadata XML parts rely on them), plus one Override per non-default content type. This manifest asserts the Defaults are present and that /word/document.xml is registered as the WordprocessingML main document content type — a structural OPC invariant python-docx must always satisfy, independent of any document content."
  },
  "fixtures": {
    "machine": "docx/content-type-registration"
  },
  "generator": {
    "python": "scripts/gen_content_type_registration.py"
  },
  "assertions": [
    {
      "id": "default-rels-registered",
      "part": "[Content_Types].xml",
      "namespaces": {
        "ct": "http://schemas.openxmlformats.org/package/2006/content-types"
      },
      "xpath": "//ct:Default[@Extension='rels']/@ContentType",
      "must": "equal",
      "value": "application/vnd.openxmlformats-package.relationships+xml",
      "description": "The `rels` extension Default must map to the OPC relationships content type."
    },
    {
      "id": "default-xml-registered",
      "part": "[Content_Types].xml",
      "namespaces": {
        "ct": "http://schemas.openxmlformats.org/package/2006/content-types"
      },
      "xpath": "//ct:Default[@Extension='xml']/@ContentType",
      "must": "equal",
      "value": "application/xml",
      "description": "The `xml` extension Default must map to application/xml."
    },
    {
      "id": "override-document-main",
      "part": "[Content_Types].xml",
      "namespaces": {
        "ct": "http://schemas.openxmlformats.org/package/2006/content-types"
      },
      "xpath": "//ct:Override[@PartName='/word/document.xml']/@ContentType",
      "must": "equal",
      "value": "application/vnd.openxmlformats-officedocument.wordprocessingml.document.main+xml",
      "description": "The main document part must be explicitly bound to the WordprocessingML document content type via an Override."
    },
    {
      "id": "override-styles-present",
      "part": "[Content_Types].xml",
      "namespaces": {
        "ct": "http://schemas.openxmlformats.org/package/2006/content-types"
      },
      "xpath": "//ct:Override[@PartName='/word/styles.xml']/@ContentType",
      "must": "equal",
      "value": "application/vnd.openxmlformats-officedocument.wordprocessingml.styles+xml",
      "description": "The styles part must be registered via Override (its URI shares the .xml Default extension but the part needs the WML-styles content type, so Override resolution takes precedence per OPC §10.1.2.4)."
    },
    {
      "id": "override-core-properties-present",
      "part": "[Content_Types].xml",
      "namespaces": {
        "ct": "http://schemas.openxmlformats.org/package/2006/content-types"
      },
      "xpath": "//ct:Override[@PartName='/docProps/core.xml']/@ContentType",
      "must": "equal",
      "value": "application/vnd.openxmlformats-package.core-properties+xml",
      "description": "The OPC core-properties part must be registered via Override."
    },
    {
      "id": "no-duplicate-overrides",
      "part": "[Content_Types].xml",
      "namespaces": {
        "ct": "http://schemas.openxmlformats.org/package/2006/content-types"
      },
      "xpath": "//ct:Override[@PartName = following-sibling::ct:Override/@PartName]",
      "must": "absent",
      "description": "OPC §10.1.2.2 forbids two Override entries for the same PartName — catches a family of serialiser-duplication bugs."
    },
    {
      "id": "primary-document-rel-exists",
      "part": "_rels/.rels",
      "namespaces": {
        "r": "http://schemas.openxmlformats.org/package/2006/relationships"
      },
      "xpath": "//r:Relationship[@Type='http://schemas.openxmlformats.org/officeDocument/2006/relationships/officeDocument']/@Target",
      "must": "match",
      "value": "^/?word/document\\.xml$",
      "description": "The package's /_rels/.rels must advertise exactly one officeDocument relationship whose Target is word/document.xml (leading slash optional per OPC §8.3)."
    }
  ]
}