custom metadata

2024-12-31 15:14:39 +01:00
parent 9ec1ca465c
commit b6d403315a
4 changed files with 113 additions and 18 deletions
@@ -1140,6 +1140,7 @@ def assert_meta(html, **meta):
    meta.setdefault('attachments', [])
    meta.setdefault('lang', None)
    meta.setdefault('custom', {})
+    meta.setdefault('rdf_metadata_generator', None)
    assert vars(FakeHTML(string=html).render().metadata) == meta


@@ -716,3 +716,35 @@ def test_bleed(style, media, bleed, trim):
    assert f'/MediaBox {str(media).replace(",", "")}'.encode() in pdf
    assert f'/BleedBox {str(bleed).replace(",", "")}'.encode() in pdf
    assert f'/TrimBox {str(trim).replace(",", "")}'.encode() in pdf
+
+
+@assert_no_logs
+def test_default_rdf_metadata() -> None:
+    pdf_document = FakeHTML(string='<body>test</body>').render()
+
+    pdf_document.metadata.title = None
+
+    pdf_bytes = pdf_document.write_pdf(
+        pdf_variant='pdf/a-3b',
+        pdf_identifier=b'example-bytes',
+        uncompressed_pdf=True,
+    )
+    assert b'<rdf:RDF xmlns:pdf="http://ns.adobe.com/pdf/1.3/"' in pdf_bytes
+
+
+@assert_no_logs
+def test_custom_rdf_metadata() -> None:
+    def rdf_metadata_generator(*args, **kwargs) -> bytes:
+        return b'TEST_METADATA'
+
+    pdf_document = FakeHTML(string='<body>test</body>').render()
+
+    pdf_document.metadata.title = None
+    pdf_document.metadata.rdf_metadata_generator = rdf_metadata_generator
+
+    pdf_bytes = pdf_document.write_pdf(
+        pdf_variant='pdf/a-3b',
+        pdf_identifier=b'example-bytes',
+        uncompressed_pdf=True,
+    )
+    assert b'TEST_METADATA' in pdf_bytes
@@ -4,8 +4,9 @@ import functools
 import io
 from hashlib import md5
 from pathlib import Path
+from typing import Callable, Optional

-from . import CSS, DEFAULT_OPTIONS
+from . import CSS, DEFAULT_OPTIONS, Attachment
 from .anchors import gather_anchors, make_page_bookmark_tree
 from .css import get_all_computed_styles
 from .css.counters import CounterStyle
@@ -105,12 +106,33 @@ class DocumentMetadata:
    """Meta-information belonging to a whole :class:`Document`.

    New attributes may be added in future versions of WeasyPrint.
-
    """
+    title: Optional[str]
+    authors: list[str]
+    description: Optional[str]
+    keywords: list[str]
+    generator: Optional[str]
+    created: Optional[str]
+    modified: Optional[str]
+    attachments: list[Attachment]
+    lang: Optional[str]
+    custom: dict
+    rdf_metadata_generator: Optional[Callable] = None

-    def __init__(self, title=None, authors=None, description=None,
-                 keywords=None, generator=None, created=None, modified=None,
-                 attachments=None, lang=None, custom=None):
+    def __init__(
+        self,
+        title: Optional[str] = None,
+        authors: Optional[list[str]] = None,
+        description: Optional[str] = None,
+        keywords: Optional[list[str]] = None,
+        generator: Optional[str] = None,
+        created: Optional[str] = None,
+        modified: Optional[str] = None,
+        attachments: Optional[list[Attachment]] = None,
+        lang: Optional[str] = None,
+        custom: Optional[dict] = None,
+        rdf_metadata_generator: Optional[Callable] = None,
+    ):
        #: The title of the document, as a string or :obj:`None`.
        #: Extracted from the ``<title>`` element in HTML
        #: and written to the ``/Title`` info field in PDF.
@@ -156,6 +178,9 @@ class DocumentMetadata:
        #: Custom metadata, as a dict whose keys are the metadata names and
        #: values are the metadata values.
        self.custom = custom or {}
+        #: Custom RDF metadata generator, which will replace the default generator.
+        #: The function should return bytes containing an RDF XML.
+        self.rdf_metadata_generator = rdf_metadata_generator


 class DiskCache:
@@ -1,13 +1,20 @@
-"""PDF metadata stream generation."""
+"""
+PDF metadata stream generation.
+"""

+from typing import TYPE_CHECKING
 from xml.etree.ElementTree import Element, SubElement, register_namespace, tostring

 import pydyf

-from .. import __version__
+from weasyprint import __version__
+
+if TYPE_CHECKING:
+    from weasyprint.document import DocumentMetadata
+

 # XML namespaces used for metadata
-NS = {
+NS: dict[str, str] = {
    'rdf': 'http://www.w3.org/1999/02/22-rdf-syntax-ns#',
    'dc': 'http://purl.org/dc/elements/1.1/',
    'xmp': 'http://ns.adobe.com/xap/1.0/',
@@ -19,13 +26,50 @@ for key, value in NS.items():
    register_namespace(key, value)


-def add_metadata(pdf, metadata, variant, version, conformance, compress):
+def add_metadata(
+    pdf: pydyf.PDF,
+    metadata: 'DocumentMetadata',
+    variant: str,
+    version: str,
+    conformance: str,
+    compress: bool,
+) -> None:
    """Add PDF stream of metadata.

    Described in ISO-32000-1:2008, 14.3.2.

    """
-    # Add metadata
+    # Add metadata. If `DocumentMetadata` has a generator, we will use it,
+    # otherwise we will use the default generator.
+    if metadata.rdf_metadata_generator is None:
+        xml_data = generate_rdf_metadata(metadata, variant, version, conformance)
+    else:
+        xml_data = metadata.rdf_metadata_generator(
+            metadata=metadata,
+            variant=variant,
+            version=version,
+            conformance=conformance,
+        )
+
+    header = b'<?xpacket begin="" id="W5M0MpCehiHzreSzNTczkc9d"?>'
+    footer = b'<?xpacket end="r"?>'
+    stream_content = b'\n'.join((header, xml_data, footer))
+    extra = {'Type': '/Metadata', 'Subtype': '/XML'}
+    metadata = pydyf.Stream([stream_content], extra, compress)
+    pdf.add_object(metadata)
+    pdf.catalog['Metadata'] = metadata.reference
+
+
+def generate_rdf_metadata(
+    metadata: 'DocumentMetadata',
+    variant: str,
+    version: str,
+    conformance: str,
+) -> bytes:
+    """Generates RDF metadata. Might be replaced by
+    DocumentMetadata.rdf_matadata_generator().
+
+    """
    namespace = f'pdf{variant}id'
    rdf = Element(f'{{{NS["rdf"]}}}RDF')

@@ -82,11 +126,4 @@ def add_metadata(pdf, metadata, variant, version, conformance, compress):
        element.attrib[f'{{{NS["rdf"]}}}about'] = ''
        element = SubElement(element, f'{{{NS["xmp"]}}}ModifyDate')
        element.text = metadata.modified
-    xml = tostring(rdf, encoding='utf-8')
-    header = b'<?xpacket begin="" id="W5M0MpCehiHzreSzNTczkc9d"?>'
-    footer = b'<?xpacket end="r"?>'
-    stream_content = b'\n'.join((header, xml, footer))
-    extra = {'Type': '/Metadata', 'Subtype': '/XML'}
-    metadata = pydyf.Stream([stream_content], extra, compress)
-    pdf.add_object(metadata)
-    pdf.catalog['Metadata'] = metadata.reference
+    return tostring(rdf, encoding='utf-8')