Ejemplo n.º 1
0
def migrate_doc(doc: DocCursor) -> DocCursor:
    """Apply all transforms to a given document. Save it and return."""
    for transform in transforms:
        doc = transform(doc)
    doc.nested_set_renumber(bulk_create=False)
    for node in doc.walk():
        node.save()
    return doc
Ejemplo n.º 2
0
 def get_footnote_node(self, instance: FootnoteCitation):
     footnote_tree = DocCursor(self.cursor_tree,
                               instance.footnote_node.identifier)
     cursor_serializer = type(self.context['parent_serializer'])
     return cursor_serializer(footnote_tree, context={
         'is_root': False
     }).data
Ejemplo n.º 3
0
def test_external_links():
    """The "content" field should contain serialized ExternalLinks."""
    policy = mommy.make(Policy)
    para = DocCursor.new_tree('para', text='Go over there!', policy=policy)
    para.nested_set_renumber()
    para.externallinks.create(start=len('Go over '),
                              end=len('Go over there'),
                              href='http://example.com/aaa')

    result = doc_cursor.DocCursorSerializer(para, context={
        'policy': policy
    }).data
    assert result['content'] == [{
        'content_type': '__text__',
        'inlines': [],
        'text': 'Go over '
    }, {
        'content_type':
        'external_link',
        'inlines': [{
            'content_type': '__text__',
            'inlines': [],
            'text': 'there',
        }],
        'href':
        'http://example.com/aaa',
        'text':
        'there',
    }, {
        'content_type': '__text__',
        'inlines': [],
        'text': '!',
    }]
Ejemplo n.º 4
0
def test_content_no_annotations():
    node = mommy.make(DocNode, text='Some text here')
    cursor = DocCursor.load_from_model(node)
    content = doc_cursor.DocCursorSerializer(cursor).data['content']
    assert len(content) == 1
    assert content[0]['text'] == 'Some text here'
    assert content[0]['content_type'] == '__text__'
Ejemplo n.º 5
0
def test_json_put_works_for_admin_users(admin_client):
    policy = mommy.make(Policy, workflow_phase='published')
    root = DocCursor.new_tree('root', '0', policy=policy)
    root.add_child('sec', text='blah')
    root.nested_set_renumber()

    # Get the original document...
    response = admin_client.get(f"/{policy.pk}")
    assert response.status_code == 200
    result = response.json()

    # Modify it a bit...
    result['children'][0]['title'] = 'boop'
    result['children'][0]['content'][0]['text'] = 'hallo'

    response = admin_client.put(f"/{policy.pk}",
                                data=json.dumps(result),
                                content_type='application/json')
    assert response.status_code == 204

    # Now fetch it again, and make sure our modification stuck.
    response = admin_client.get(f"/{policy.pk}")
    assert response.status_code == 200
    result = response.json()
    assert result['children'][0]['title'] == 'boop'
    assert result['children'][0]['content'][0]['text'] == 'hallo'
Ejemplo n.º 6
0
 def handle(self, *args, **kwargs):
     roots = DocNode.objects.filter(depth=0)
     with tqdm(total=roots.count()) as pbar:
         for root_docnode in roots:
             with transaction.atomic():
                 doc = DocCursor.load_from_model(root_docnode)
                 migrate_doc(doc)
             pbar.update(1)
Ejemplo n.º 7
0
def test_policy_edit_doc_button(admin_client):
    policy = mommy.make(Policy, slug='some-policy')
    root = DocCursor.new_tree('root', policy=policy)
    root.nested_set_renumber()

    result = admin_client.get(f'/admin/reqs/policy/{policy.pk}/change/')
    assert b'_savethendoc' in result.content
    assert b'Save and edit document' in result.content
Ejemplo n.º 8
0
 def to_representation(self, instance: DocCursor) -> List[PrimitiveDict]:
     annotations = nest_annotations(
         instance.annotations(), len(instance.text))
     return NestedAnnotationSerializer(
         annotations,
         context={'cursor': instance,
                  'parent_serializer': self.context['parent_serializer']},
         many=True,
     ).data
Ejemplo n.º 9
0
def test_by_pretty_url(client):
    policy = mommy.make(Policy,
                        omb_policy_id='M-Something-18',
                        workflow_phase='published')
    root = DocCursor.new_tree('root', '0', policy=policy)
    root.nested_set_renumber()

    result = json.loads(client.get("/M-Something-18").content.decode("utf-8"))

    assert result == DocCursorSerializer(root, context={'policy': policy}).data
Ejemplo n.º 10
0
def test_404s(client):
    policy = mommy.make(Policy, workflow_phase='published')
    root = DocCursor.new_tree('root', '0', policy=policy)
    root.add_child('sec')
    root.nested_set_renumber()

    assert client.get("/987654321").status_code == 404
    assert client.get(f"/{policy.pk}").status_code == 200
    assert client.get(f"/{policy.pk}/root_0").status_code == 200
    assert client.get(f"/{policy.pk}/root_1").status_code == 404
    assert client.get(f"/{policy.pk}/root_0__sec_1").status_code == 200
    assert client.get(f"/{policy.pk}/root_0__sec_2").status_code == 404
Ejemplo n.º 11
0
 def get_descendant_footnotes(self, instance):
     """Find all footnote nodes that are cited by this node or any of its
     descendants."""
     if not instance.is_root and instance.node_type != 'table':
         return None
     footnotes = []
     for node in instance.cursor.walk():
         for citation in node.footnotecitations.all():
             subtree = DocCursor(instance.cursor.tree,
                                 citation.footnote_node.identifier)
             footnotes.append(self.serialize_doc_cursor(subtree))
     return footnotes
Ejemplo n.º 12
0
def test_nonpublic(client, admin_client):
    policy = mommy.make(Policy,
                        omb_policy_id='M-Something-18',
                        workflow_phase='edit')
    root = DocCursor.new_tree('root', '0', policy=policy)
    root.nested_set_renumber()

    result = client.get('/M-Something-18')
    assert result.status_code == 404

    result = admin_client.get('/M-Something-18')
    assert result.status_code == 200
Ejemplo n.º 13
0
def test_table_of_contents():
    """The table of contents should be generated from titled elements,
    regardless of position in tree."""
    policy = mommy.make(Policy)
    root = DocCursor.new_tree('root', policy=policy)
    root.add_child('container', title='First title')
    root['container_1'].add_child('para')
    root['container_1']['para_1'].add_child('subpar')
    root['container_1']['para_1']['subpar_1'].add_child('item', title='Sub')
    root.add_child('other', title='Second')
    root.nested_set_renumber()

    data = meta.MetaSerializer(
        meta.Meta(root, is_root=True, policy=policy),
        context={
            'parent_serializer': DocCursorSerializer()
        },
    ).data
    assert data['table_of_contents'] == {
        'identifier':
        'root_1',
        'title':
        '',
        'children': [
            {
                'identifier':
                'root_1__container_1',
                'title':
                'First title',
                'children': [{
                    'identifier': ('root_1__container_1__para_1__subpar_1'
                                   '__item_1'),
                    'title':
                    'Sub',
                    'children': [],
                }],
            },
            {
                'identifier': 'root_1__other_1',
                'title': 'Second',
                'children': [],
            },
        ],
    }

    data = meta.MetaSerializer(
        meta.Meta(root['other_1'], is_root=False, policy=policy),
        context={
            'parent_serializer': DocCursorSerializer()
        },
    ).data
    assert 'table_of_contents' not in data
Ejemplo n.º 14
0
def test_ensure_section_has_heading():
    root = DocCursor.new_tree('root')
    sec1 = root.add_child('sec')
    sec1.add_child('sec')
    sec12 = sec1.add_child('sec')
    sec12.add_child('heading', text='Subheading')
    root.add_child('sec')

    root = migrate_documents.ensure_section_has_heading(root)
    assert root['sec_1']['heading_1'].text == '--Missing Heading--'
    assert root['sec_1']['sec_1']['heading_1'].text == '--Missing Heading--'
    assert root['sec_1']['sec_2']['heading_1'].text == 'Subheading'
    assert root['sec_2']['heading_1'].text == '--Missing Heading--'
Ejemplo n.º 15
0
def ensure_section_has_heading(doc: DocCursor) -> DocCursor:
    """We expect all sections to have a heading. Fill the missing data with
    placeholder text."""
    # materialize so we don't need to worry about a modified iterator
    secs = list(doc.filter(lambda n: n.node_type == 'sec'))
    for sec in secs:
        children = list(sec.children())
        if not children or children[0].node_type != 'heading':
            sec.add_child('heading',
                          insert_pos=0,
                          text='--Missing Heading--',
                          policy_id=doc.policy_id)
    return doc
Ejemplo n.º 16
0
def test_descendant_footnotes_meta(node_type, is_root):
    """Only the root and "table" nodes should get descendant_footnotes."""
    policy = mommy.make(Policy)
    cursor = DocCursor.new_tree(node_type, policy=policy)
    cursor.nested_set_renumber()
    meta_obj = meta.Meta(cursor, is_root, policy)
    result = meta.MetaSerializer(
        meta_obj,
        context={
            'parent_serializer': DocCursorSerializer()
        },
    ).data
    if node_type == 'table' or is_root:
        assert 'descendant_footnotes' in result
    else:
        assert 'descendant_footnotes' not in result
Ejemplo n.º 17
0
def test_descendant_footnotes():
    """We pull out footnotes of all descendants, and only descendants."""
    policy = mommy.make(Policy)
    root = DocCursor.new_tree('root', policy=policy)
    ftnt_a = root.add_child('footnote', 'a')
    root.add_child('para')
    ftnt_b = root['para_1'].add_child('footnote', 'b')
    root.add_child('list')
    root['list_1'].add_child('para')
    root['list_1'].add_child('para')
    root['list_1'].add_child('para')
    ftnt_c = root['list_1']['para_3'].add_child('footnote', 'c')
    root.nested_set_renumber()

    root['para_1'].footnotecitations.create(start=0,
                                            end=1,
                                            footnote_node=ftnt_a.model)
    root['para_1'].footnotecitations.create(start=1,
                                            end=2,
                                            footnote_node=ftnt_b.model)
    root['list_1']['para_2'].footnotecitations.create(
        start=0, end=1, footnote_node=ftnt_c.model)

    def fts(cursor):
        meta_obj = meta.Meta(cursor, is_root=True, policy=policy)
        data = meta.MetaSerializer(
            meta_obj,
            context={
                'parent_serializer': DocCursorSerializer()
            },
        ).data
        return [node['identifier'] for node in data['descendant_footnotes']]

    assert fts(root) == [
        'root_1__footnote_a', 'root_1__para_1__footnote_b',
        'root_1__list_1__para_3__footnote_c'
    ]
    assert fts(root['footnote_a']) == []
    assert fts(root['para_1']) == [
        'root_1__footnote_a', 'root_1__para_1__footnote_b'
    ]
    assert fts(root['list_1']) == ['root_1__list_1__para_3__footnote_c']
    assert fts(
        root['list_1']['para_2']) == ['root_1__list_1__para_3__footnote_c']
    # no citations in para 3
    assert fts(root['list_1']['para_3']) == []
Ejemplo n.º 18
0
def test_content_outside():
    node = mommy.make(DocNode, text='Some text here')
    node.externallinks.create(start=0,
                              end=len('Some '),
                              href='http://example.com/aaa')
    node.externallinks.create(start=len('Some text'),
                              end=len('Some text here'),
                              href='http://example.com/bbb')
    cursor = DocCursor.load_from_model(node)
    content = doc_cursor.DocCursorSerializer(cursor).data['content']

    assert len(content) == 3
    assert [c['text'] for c in content] == ['Some ', 'text', ' here']
    assert [c['content_type'] for c in content] == [
        'external_link',
        '__text__',
        'external_link',
    ]
Ejemplo n.º 19
0
def test_create_document_success(monkeypatch):
    monkeypatch.setattr(forms, 'OMBDocument', Mock())
    monkeypatch.setattr(forms, 'to_db', Mock())
    monkeypatch.setattr(forms, 'migrate_doc', Mock())
    policy = mommy.make(Policy,
                        document_source=SimpleUploadedFile('a.pdf', b'aaa'))
    doc = DocCursor.new_tree('policy', policy=policy)
    doc.add_child('para', text='Content')
    doc.nested_set_renumber()
    forms.to_db.return_value = doc

    forms.create_document(policy)

    assert forms.OMBDocument.from_file.call_args ==\
        call(policy.document_source.file)
    assert forms.to_db.called
    assert policy.workflow_phase == 'cleanup'
    assert forms.migrate_doc.called
Ejemplo n.º 20
0
 def get_object(self, prefetch_related=True):
     only_published = not self.request.user.is_authenticated
     policy = policy_or_404(self.kwargs['policy_id'], only_published)
     # we'll pass this policy down when we serialize
     self.policy = policy
     query_args = {'policy_id': policy.pk}
     if self.kwargs.get('identifier'):
         query_args['identifier'] = self.kwargs['identifier']
     else:
         query_args['depth'] = 0
     queryset = DocNode.objects
     if prefetch_related:
         queryset = queryset.prefetch_annotations()
     root_doc = get_object_or_404(queryset, **query_args)
     root = DocCursor.load_from_model(root_doc, subtree=False)
     if prefetch_related:
         root.add_models(root_doc.descendants().prefetch_annotations())
     self.check_object_permissions(self.request, root)
     return root
Ejemplo n.º 21
0
def test_correct_data(client):
    policy = mommy.make(Policy, workflow_phase='published')
    root = DocCursor.new_tree('root', '0', policy=policy)
    sec1 = root.add_child('sec')
    root.add_child('sec')
    sec1.add_child('para', 'a')
    root.nested_set_renumber()

    def result(url):
        return json.loads(client.get(url).content.decode('utf-8'))

    def serialize(node):
        return DocCursorSerializer(node, context={'policy': policy}).data

    assert result(f"/{policy.pk}") == serialize(root)
    assert result(f"/{policy.pk}/root_0") == serialize(root)
    assert result(f"/{policy.pk}/root_0__sec_1") == serialize(root['sec_1'])
    assert result(f"/{policy.pk}/root_0__sec_2") == serialize(root['sec_2'])
    assert result(f"/{policy.pk}/root_0__sec_1__para_a") \
        == serialize(root['sec_1']['para_a'])
Ejemplo n.º 22
0
def test_policy_redirect_to_editor(admin_client):
    policy = mommy.make(Policy, slug='some-policy')
    root = DocCursor.new_tree('root', policy=policy)
    root.nested_set_renumber()

    result = admin_client.post(
        f'/admin/reqs/policy/{policy.pk}/change/', {
            'title': 'Some new policy title',
            'omb_policy_id': '',
            'issuance': policy.issuance.isoformat(),
            'sunset': '',
            'public': 'on',
            'workflow_phase': 'cleanup',
            '_savethendoc': 'Save and edit document',
        })

    assert result.status_code == 302
    assert result['Location'] == '/admin/document-editor/some-policy'

    policy.refresh_from_db()
    assert policy.title == 'Some new policy title'
Ejemplo n.º 23
0
def ensure_listitem_in_list(doc: DocCursor) -> DocCursor:
    """We expect all listitems to have a parent list. However, the parser has
    emitted lists followed by bare listitems. We'll place the listitem in the
    list that imemdiately precedes it."""
    for li in doc.filter(lambda n: n.node_type == 'listitem'):
        parent = li.parent()
        prev_sibling = li.left_sibling()

        if not parent:
            logger.warning('Root of %s is an li.',
                           doc.policy.title_with_number)
            continue
        if parent.node_type == 'list':  # all is well
            continue

        if prev_sibling and prev_sibling.node_type == 'list':
            li.append_to(prev_sibling)
            return ensure_listitem_in_list(doc)
        # else: create new list to wrap this one
        logger.warning('Could not fix li in %s', doc.policy.title_with_number)

    return doc  # no changes needed
Ejemplo n.º 24
0
def test_create_requirements():
    policy = mommy.make(Policy)
    mommy.make(Requirement, policy=policy, _quantity=5)
    doc = DocCursor.new_tree('policy', policy=policy)
    doc.add_child('para', text='First paragraph')
    doc.add_child('sec')
    doc['sec_1'].add_child('heading', text='A section')
    doc['sec_1'].add_child('para', text='Second paragraph')
    doc['sec_1'].add_child('para', text='Final paragraph')
    doc.nested_set_renumber()
    # These will be deleted
    assert Requirement.objects.filter(policy=policy).count() == 5

    forms.create_requirements(policy)

    assert Requirement.objects.filter(policy=policy).count() == 4
    assert set(Requirement.objects.values_list('req_text', flat=True)) == {
        'First paragraph',
        'A section',
        'Second paragraph',
        'Final paragraph',
    }
Ejemplo n.º 25
0
def test_ensure_listitem_in_list():
    root = DocCursor.new_tree('root')
    list_el = root.add_child('list')
    list_el.add_child('listitem', '1')
    root.add_child('listitem', '2')
    root.add_child('listitem', '3')
    assert [n.identifier for n in root.walk()] == [
        'root_1',
        'root_1__list_1',
        'root_1__list_1__listitem_1',
        'root_1__listitem_2',
        'root_1__listitem_3',
    ]

    root = migrate_documents.ensure_listitem_in_list(root)
    assert [n.identifier for n in root.walk()] == [
        'root_1',
        'root_1__list_1',
        'root_1__list_1__listitem_1',
        'root_1__list_1__listitem_2',
        'root_1__list_1__listitem_3',
    ]
Ejemplo n.º 26
0
def test_content_middle_annotation():
    cursor = DocCursor.new_tree('policy',
                                policy=mommy.make(Policy),
                                text='Some text here')
    footnote_cursor = cursor.add_child('child')
    cursor.nested_set_renumber()
    cursor.footnotecitations.create(start=len('Some '),
                                    end=len('Some text'),
                                    footnote_node=footnote_cursor.model)
    content = doc_cursor.DocCursorSerializer(cursor).data['content']

    assert len(content) == 3
    assert [c['text'] for c in content] == ['Some ', 'text', ' here']
    assert [c['content_type'] for c in content] == [
        '__text__',
        'footnote_citation',
        '__text__',
    ]
    assert content[1]['footnote_node'] == doc_cursor.DocCursorSerializer(
        footnote_cursor, context={
            'is_root': False
        }).data
Ejemplo n.º 27
0
def test_akn_put_works_for_admin_users(admin_client):
    policy = mommy.make(Policy, workflow_phase='published')
    root = DocCursor.new_tree('root', '0', policy=policy)
    root.add_child('sec', text='blah')
    root.nested_set_renumber()

    # Get the original document...
    response = admin_client.get(f"/{policy.pk}?format=akn")
    assert response.status_code == 200
    assert response['content-type'] == 'application/akn+xml; charset=utf-8'

    # Modify it a bit...
    xml = response.content.replace(b'blah', b'hallo')

    response = admin_client.put(f"/{policy.pk}",
                                data=xml,
                                content_type='application/akn+xml')
    assert response.status_code == 204

    # Now fetch it again, and make sure our modification stuck.
    response = admin_client.get(f"/{policy.pk}")
    assert response.status_code == 200
    result = response.json()
    assert result['children'][0]['content'][0]['text'] == 'hallo'
Ejemplo n.º 28
0
 def get_table_of_contents(self, instance):
     if instance.is_root:
         only_titled = DocNode.objects.exclude(title='')
         titled_cursor = DocCursor.load_from_model(instance.cursor.model,
                                                   queryset=only_titled)
         return TableOfContentsSerializer(titled_cursor).data
Ejemplo n.º 29
0
def test_end_to_end():
    """Create a tree, then serialize it. Trivially modify the serialized
    value and deserialize it."""
    policy = mommy.make(
        Policy,
        issuance=date(2001, 2, 3),
        omb_policy_id='M-18-18',
        title='Some Title',
        uri='http://example.com/thing.pdf',
    )
    root = DocCursor.new_tree('root', '0', policy=policy, title='Policy A')
    root.add_child('sec', text='Section 1', title='First Section')
    sec2 = root.add_child('sec', title='Section 2')
    pa = sec2.add_child('para', 'a', marker='(a)')
    pa.add_child('para', '1', text='Paragraph (a)(1)', marker='(1)')
    sec2.add_child('para', 'b', marker='b.')
    root.nested_set_renumber()

    result = doc_cursor.DocCursorSerializer(root, context={
        'policy': policy
    }).data
    assert result == {
        'identifier': 'root_0',
        'node_type': 'root',
        'type_emblem': '0',
        'text': '',
        'title': 'Policy A',
        'marker': '',
        'depth': 0,
        'content': [],
        'meta': {
            'descendant_footnotes': [],
            'policy': {     # Note this field does not appear on children
                'issuance': '2001-02-03',
                'omb_policy_id': 'M-18-18',
                'original_url': 'http://example.com/thing.pdf',
                'title': 'Some Title',
                'title_with_number': 'M-18-18: Some Title',
            },
            'table_of_contents': {
                'identifier': 'root_0',
                'title': 'Policy A',
                'children': [
                    {
                        'children': [],
                        'identifier': 'root_0__sec_1',
                        'title': 'First Section',
                    }, {
                        'children': [],
                        'identifier': 'root_0__sec_2',
                        'title': 'Section 2',
                    },
                ],
            },
        },
        'children': [
            {
                'identifier': 'root_0__sec_1',
                'node_type': 'sec',
                'type_emblem': '1',
                'text': 'Section 1',
                'title': 'First Section',
                'marker': '',
                'depth': 1,
                'meta': {},
                'content': [{
                    'content_type': '__text__',
                    'inlines': [],
                    'text': 'Section 1',
                }],
                'children': [],
            },
            {
                'identifier': 'root_0__sec_2',
                'node_type': 'sec',
                'type_emblem': '2',
                'text': '',
                'title': 'Section 2',
                'marker': '',
                'depth': 1,
                'meta': {},
                'content': [],
                'children': [
                    {
                        'identifier': 'root_0__sec_2__para_a',
                        'node_type': 'para',
                        'type_emblem': 'a',
                        'text': '',
                        'title': '',
                        'marker': '(a)',
                        'depth': 2,
                        'meta': {},
                        'content': [],
                        'children': [
                            {
                                'identifier': 'root_0__sec_2__para_a__para_1',
                                'node_type': 'para',
                                'type_emblem': '1',
                                'text': 'Paragraph (a)(1)',
                                'title': '',
                                'marker': '(1)',
                                'depth': 3,
                                'meta': {},
                                'content': [{
                                    'content_type': '__text__',
                                    'inlines': [],
                                    'text': 'Paragraph (a)(1)',
                                }],
                                'children': [],
                            },
                        ],
                    },
                    {
                        'identifier': 'root_0__sec_2__para_b',
                        'node_type': 'para',
                        'type_emblem': 'b',
                        'text': '',
                        'title': '',
                        'marker': 'b.',
                        'depth': 2,
                        'meta': {},
                        'content': [],
                        'children': [],
                    },
                ],
            },
        ],
    }

    result['title'] = 'MODIFIED Policy A'

    val = doc_cursor.DocCursorSerializer().to_internal_value(result)

    new_root = doc_cursor.DocCursorSerializer().update(root, val)

    assert new_root.title == 'MODIFIED Policy A'
    assert new_root.policy.pk == policy.pk
Ejemplo n.º 30
0
def test_footnote_citations():
    """The "content" field should contain serialized FootnoteCitations."""
    policy = mommy.make(Policy)
    para = DocCursor.new_tree('para',
                              text='Some1 message2 here',
                              policy=policy)
    footnote1 = para.add_child('footnote').model
    footnote2 = para.add_child('footnote').model
    para.nested_set_renumber()
    para.footnotecitations.create(start=len('Some'),
                                  end=len('Some1'),
                                  footnote_node=footnote1)
    para.footnotecitations.create(start=len('Some1 message'),
                                  end=len('Some1 message2'),
                                  footnote_node=footnote2)

    result = doc_cursor.DocCursorSerializer(para, context={
        'policy': policy
    }).data
    assert result['content'] == [{
        'content_type': '__text__',
        'inlines': [],
        'text': 'Some',
    }, {
        'content_type':
        'footnote_citation',
        'inlines': [{
            'content_type': '__text__',
            'inlines': [],
            'text': '1',
        }],
        'text':
        '1',
        'footnote_node':
        doc_cursor.DocCursorSerializer(
            para['footnote_1'],
            context={
                'policy': policy,
                'is_root': False
            },
        ).data,
    }, {
        'content_type': '__text__',
        'inlines': [],
        'text': ' message',
    }, {
        'content_type':
        'footnote_citation',
        'inlines': [{
            'content_type': '__text__',
            'inlines': [],
            'text': '2',
        }],
        'text':
        '2',
        'footnote_node':
        doc_cursor.DocCursorSerializer(
            para['footnote_2'],
            context={
                'policy': policy,
                'is_root': False
            },
        ).data,
    }, {
        'content_type': '__text__',
        'inlines': [],
        'text': ' here',
    }]