misc: only let rst consider arabic sequences (#29183)

tests/test_misc.py

@@ -418,3 +418,30 @@ def test_normalize_geolocation():
     assert normalize_geolocation({'lat': 0.0, 'lon': -360.0}) == {'lat': 0.0, 'lon': 0.0}
     assert normalize_geolocation({'lat': 0.0, 'lon': 400.0}) == {'lat': 0.0, 'lon': 40.0}
     assert normalize_geolocation({'lat': 0.0, 'lon': -400.0}) == {'lat': 0.0, 'lon': -40.0}
+
+@pytest.mark.skipif('docutils is None')
+def test_email_with_enumeration(emails):
+    pub = create_temporary_pub()
+    pub.cfg['emails'] = {'footer': 'Footer\nText'}
+    mail_body = '''
+A. FooAlpha1
+B. FooAlpha2
+
+1. Num1
+2. Num2
+
+M. Francis Kuntz
+
+'''
+    send_email('test', mail_body=mail_body, email_rcpt='test@localhost')
+    assert emails.count() == 1
+    assert emails.emails['test']['msg'].is_multipart()
+    assert emails.emails['test']['msg'].get_content_subtype() == 'alternative'
+    assert emails.emails['test']['msg'].get_payload()[0].get_content_type() == 'text/plain'
+    assert emails.emails['test']['msg'].get_payload()[1].get_content_type() == 'text/html'
+    text = emails.emails['test']['msg'].get_payload()[0].get_payload()
+    html = emails.emails['test']['msg'].get_payload()[1].get_payload()
+    assert html.count('<ol') == 1
+    assert not '<ul' in html
+    assert 'arabic simple' in html
+    assert 'M. Francis Kuntz' in html

wcs/qommon/emails.py

@@ -36,6 +36,8 @@ try:
     import docutils
     import docutils.core
     import docutils.io
+    import docutils.parsers.rst
+    import docutils.parsers.rst.states
 except ImportError:
     docutils = None
 
@@ -53,6 +55,22 @@ from template import Template
 Charset.add_charset('utf-8', Charset.QP, Charset.QP, 'utf-8')
 
 
+if docutils:
+    # custom parser to only allow arabic sequences, this prevents the rst
+    # parser to consider M. as starting a (upper alpha / roman) sequence.
+    class Body(docutils.parsers.rst.states.Body):
+        def is_enumerated_list_item(self, ordinal, sequence, format):
+            if format == 'period' and sequence != 'arabic':
+                return False
+            return docutils.parsers.rst.states.Body.is_enumerated_list_item(
+                    self, ordinal, sequence, format)
+
+    class CustomRstParser(docutils.parsers.rst.Parser):
+        def __init__(self, *args, **kwargs):
+            docutils.parsers.rst.Parser.__init__(self, *args, **kwargs)
+            self.state_classes = tuple([Body] + list(self.state_classes[1:]))
+
+
 def custom_template_email(key, mail_body_data, email_rcpt, **kwargs):
     if not EmailsDirectory.is_enabled(key):
         return
@@ -146,7 +164,7 @@ def email(subject, mail_body, email_rcpt, replyto = None, bcc = None,
                     destination = None,
                     destination_path = None,
                     reader = None, reader_name = 'standalone',
-                    parser = None, parser_name = 'restructuredtext',
+                    parser=CustomRstParser(), parser_name=None,
                     writer = None, writer_name = 'html',
                     settings = None, settings_spec = None,
                     settings_overrides = {'input_encoding': encoding,