Merge pull request #1695 from lutrasecurity/lang

Add language information to PDF
Kozea · Aug 18, 2022 · 67cd41a · 67cd41a
2 parents 9a4c5fc + 42db8bd
commit 67cd41a
Show file tree

Hide file tree

Showing 4 changed files with 12 additions and 2 deletions.
diff --git a/tests/test_api.py b/tests/test_api.py
@@ -998,6 +998,7 @@ def assert_meta(html, **meta):
     meta.setdefault('created', None)
     meta.setdefault('modified', None)
     meta.setdefault('attachments', [])
+    meta.setdefault('lang', None)
     meta.setdefault('custom', {})
     assert vars(FakeHTML(string=html).render().metadata) == meta
 
@@ -1011,6 +1012,7 @@ def test_html_meta_1():
 def test_html_meta_2():
     assert_meta(
         '''
+            <html lang="en"><head>
             <meta name=author content="I Me &amp; Myself">
             <meta name=author content="Smith, John">
             <title>Test document</title>
@@ -1027,6 +1029,7 @@ def test_html_meta_2():
             <meta name=dcterms.modified content=2013>
             <meta name=keywords content="Python; pydyf">
             <meta name=description content="Blah… ">
+            </head></html>
         ''',
         authors=['I Me & Myself', 'Smith, John'],
         title='Test document',
@@ -1035,6 +1038,7 @@ def test_html_meta_2():
         description="Blah… ",
         created='2011-04',
         modified='2013',
+        lang='en',
         custom={'dummy': 'ignored'})
 
 

diff --git a/weasyprint/document.py b/weasyprint/document.py
@@ -104,7 +104,7 @@ class DocumentMetadata:
     """
     def __init__(self, title=None, authors=None, description=None,
                  keywords=None, generator=None, created=None, modified=None,
-                 attachments=None, custom=None):
+                 attachments=None, lang=None, custom=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.
@@ -145,6 +145,9 @@ def __init__(self, title=None, authors=None, description=None,
         #: Extracted from the ``<link rel=attachment>`` elements in HTML
         #: and written to the ``/EmbeddedFiles`` dictionary in PDF.
         self.attachments = attachments or []
+        #: Document language as BCP 47 language tags.
+        #: Extracted from ``<html lang=lang>`` in HTML.
+        self.lang = lang
         #: Custom metadata, as a dict whose keys are the metadata names and
         #: values are the metadata values.
         self.custom = custom or {}

diff --git a/weasyprint/html.py b/weasyprint/html.py
@@ -267,6 +267,7 @@ def get_html_metadata(html):
     modified = None
     attachments = []
     custom = {}
+    lang = html.etree_element.attrib.get('lang', None)
     for element in html.wrapper_element.query_all('title', 'meta', 'link'):
         element = element.etree_element
         if element.tag == 'title' and title is None:
@@ -305,7 +306,7 @@ def get_html_metadata(html):
     return dict(title=title, description=description, generator=generator,
                 keywords=keywords, authors=authors,
                 created=created, modified=modified,
-                attachments=attachments, custom=custom)
+                attachments=attachments, lang=lang, custom=custom)
 
 
 def strip_whitespace(string):

diff --git a/weasyprint/pdf/__init__.py b/weasyprint/pdf/__init__.py
@@ -417,6 +417,8 @@ def generate_pdf(pages, url_fetcher, metadata, fonts, target, zoom,
     if metadata.modified:
         pdf.info['ModDate'] = pydyf.String(
             _w3c_date_to_pdf(metadata.modified, 'modified'))
+    if metadata.lang:
+        pdf.catalog['Lang'] = pydyf.String(metadata.lang)
     if custom_metadata:
         for key, value in metadata.custom.items():
             key = ''.join(char for char in key if char.isalnum())