def test_it_merges_documents_when_duplicates_found(req):
docuri_1 = models.DocumentURI(claimant='http://example.org/',
uri='http://example.org/',
type='self-claim')
docuri_2 = models.DocumentURI(claimant='https://example.net/',
uri='https://example.org/',
type='rel-canonical')
req.db.add_all([
models.Document(document_uris=[docuri_1]),
models.Document(document_uris=[docuri_2]),
])
req.db.flush()
normalize_uris.normalize_document_uris(req)
assert req.db.query(models.Document).count() == 1
class TestDocumentSearchIndexPresenter:
@pytest.mark.parametrize(
"document,expected",
[
(models.Document(title="Foo"), {
"title": ["Foo"]
}),
(models.Document(title=""), {}),
(models.Document(title=None), {}),
(models.Document(web_uri="http://foo.org"), {
"web_uri": "http://foo.org"
}),
(models.Document(web_uri=""), {}),
(models.Document(web_uri=None), {}),
(
models.Document(title="Foo", web_uri="http://foo.org"),
{
"title": ["Foo"],
"web_uri": "http://foo.org"
},
),
(None, {}),
],
)
def test_asdict(self, document, expected):
assert expected == DocumentSearchIndexPresenter(document).asdict()
class TestDocumentSearchIndexPresenter:
@pytest.mark.parametrize(
"document,expected",
[
(models.Document(title="Foo"), {
"title": ["Foo"]
}),
(models.Document(title=""), {}),
(models.Document(title=None), {}),
# *Searching* for an annotation by `annotation.document` (e.g. by
# document `title`` or `web_uri`) isn't enabled. But you can
# retrieve an annotation by ID, or by searching on other field(s),
# and then access its `document`. Bouncer
# (https://github.com/hypothesis/bouncer) accesses h's
# Elasticsearch index directly and uses this `document` field.
(models.Document(web_uri="http://foo.org"), {
"web_uri": "http://foo.org"
}),
(models.Document(web_uri=""), {}),
(models.Document(web_uri=None), {}),
(
models.Document(title="Foo", web_uri="http://foo.org"),
{
"title": ["Foo"],
"web_uri": "http://foo.org"
},
),
(None, {}),
],
)
def test_asdict(self, document, expected):
assert expected == DocumentSearchIndexPresenter(document).asdict()
def test_it_deletes_duplicate_document_meta_objects(req):
docmeta_1 = models.DocumentMeta(_claimant="http://example.org/",
type="title",
value=["Test Title"])
docmeta_1._claimant_normalized = "http://example.org"
docmeta_2 = models.DocumentMeta(_claimant="https://example.org/",
type="title",
value=["Test Title"])
docmeta_2._claimant_normalized = "https://example.org"
req.db.add_all(
[models.Document(meta=[docmeta_1]),
models.Document(meta=[docmeta_2])])
req.db.flush()
normalize_uris.normalize_document_meta(req)
assert req.db.query(models.DocumentMeta).count() == 1
def _annotation(**kwargs):
args = {
"userid": "acct:[email protected]",
"target_selectors": [],
"created": datetime.datetime.utcnow(),
"updated": datetime.datetime.utcnow(),
"document": models.Document(),
}
args.update(**kwargs)
return models.Annotation(**args)
def _annotation(**kwargs):
args = {
'userid': 'acct:[email protected]',
'target_selectors': [],
'created': datetime.datetime.utcnow(),
'updated': datetime.datetime.utcnow(),
'document': models.Document(),
}
args.update(**kwargs)
return models.Annotation(**args)
def document(session, uri):
"""Add a new Document for the given uri to the db and return it."""
document_ = models.Document()
session.add(document_)
# Flush the session so that document.id gets generated.
session.flush()
session.add(models.DocumentURI(
claimant=uri, document_id=document_.id, uri=uri))
return document_
class TestDocumentSearchIndexPresenter(object):
@pytest.mark.parametrize('document,expected', [
(models.Document(title='Foo'), {'title': ['Foo']}),
(models.Document(title=''), {}),
(models.Document(title=None), {}),
(models.Document(web_uri='http://foo.org'), {'web_uri': 'http://foo.org'}),
(models.Document(web_uri=''), {}),
(models.Document(web_uri=None), {}),
(models.Document(title='Foo', web_uri='http://foo.org'), {'title': ['Foo'], 'web_uri': 'http://foo.org'}),
(None, {})
])
def test_asdict(self, document, expected):
assert expected == DocumentSearchIndexPresenter(document).asdict()
def test_asdict(self, db_session):
document = models.Document(
title='Foo',
document_uris=[
models.DocumentURI(uri='http://foo.com',
claimant='http://foo.com'),
models.DocumentURI(uri='http://foo.org',
claimant='http://foo.com',
type='rel-canonical')
])
db_session.add(document)
db_session.flush()
presenter = DocumentJSONPresenter(document)
expected = {'title': ['Foo']}
assert expected == presenter.asdict()
def test_it_normalizes_document_meta_claimant(req):
docmeta_1 = models.DocumentMeta(_claimant='http://example.org/',
_claimant_normalized='http://example.org',
type='title',
value=['Test Title'])
docmeta_2 = models.DocumentMeta(_claimant='http://example.net/',
_claimant_normalized='http://example.net',
type='title',
value=['Test Title'])
req.db.add(models.Document(meta=[docmeta_1, docmeta_2]))
req.db.flush()
normalize_uris.normalize_document_meta(req)
assert docmeta_1.claimant_normalized == 'httpx://example.org'
assert docmeta_2.claimant_normalized == 'httpx://example.net'
def test_it_deletes_duplicate_document_uri_objects(req):
docuri_1 = models.DocumentURI(_claimant='http://example.org/',
_claimant_normalized='http://example.org',
_uri='http://example.org/',
_uri_normalized='http://example.org',
type='self-claim')
docuri_2 = models.DocumentURI(_claimant='https://example.org/',
_claimant_normalized='https://example.org',
_uri='https://example.org/',
_uri_normalized='https://example.org',
type='self-claim')
req.db.add(models.Document(document_uris=[docuri_1, docuri_2]))
req.db.flush()
normalize_uris.normalize_document_uris(req)
assert req.db.query(models.DocumentURI).count() == 1
def test_asdict_does_not_render_other_meta_than_title(self, db_session):
document = models.Document(
title='Foo',
meta=[
models.DocumentMeta(type='title',
value=['Foo'],
claimant='http://foo.com'),
models.DocumentMeta(type='twitter.url',
value=['http://foo.com'],
claimant='http://foo.com'),
models.DocumentMeta(type='facebook.title',
value=['FB Title'],
claimant='http://foo.com')
])
db_session.add(document)
db_session.flush()
presenter = DocumentJSONPresenter(document)
assert {'title': ['Foo']} == presenter.asdict()
def test_asdict(self, db_session):
document = models.Document(
title="Foo",
document_uris=[
models.DocumentURI(uri="http://foo.com",
claimant="http://foo.com"),
models.DocumentURI(
uri="http://foo.org",
claimant="http://foo.com",
type="rel-canonical",
),
],
)
db_session.add(document)
db_session.flush()
presenter = DocumentJSONPresenter(document)
expected = {"title": ["Foo"]}
assert expected == presenter.asdict()
def test_it_normalizes_document_uris_claimant(req):
docuri_1 = models.DocumentURI(_claimant='http://example.org/',
_claimant_normalized='http://example.org',
_uri='http://example.org/',
_uri_normalized='http://example.org',
type='self-claim')
docuri_2 = models.DocumentURI(_claimant='http://example.org/',
_claimant_normalized='http://example.org',
_uri='https://example.org/',
_uri_normalized='https://example.org',
type='rel-canonical')
req.db.add(models.Document(document_uris=[docuri_1, docuri_2]))
req.db.flush()
normalize_uris.normalize_document_uris(req)
assert docuri_1.claimant_normalized == 'httpx://example.org'
assert docuri_2.claimant_normalized == 'httpx://example.org'
def test_it_normalizes_document_meta_claimant(req):
docmeta_1 = models.DocumentMeta(
_claimant="http://example.org/",
_claimant_normalized="http://example.org",
type="title",
value=["Test Title"],
)
docmeta_2 = models.DocumentMeta(
_claimant="http://example.net/",
_claimant_normalized="http://example.net",
type="title",
value=["Test Title"],
)
req.db.add(models.Document(meta=[docmeta_1, docmeta_2]))
req.db.flush()
normalize_uris.normalize_document_meta(req)
assert docmeta_1.claimant_normalized == "httpx://example.org"
assert docmeta_2.claimant_normalized == "httpx://example.net"
def test_asdict_does_not_render_other_meta_than_title(self, db_session):
document = models.Document(
title="Foo",
meta=[
models.DocumentMeta(type="title",
value=["Foo"],
claimant="http://foo.com"),
models.DocumentMeta(
type="twitter.url",
value=["http://foo.com"],
claimant="http://foo.com",
),
models.DocumentMeta(type="facebook.title",
value=["FB Title"],
claimant="http://foo.com"),
],
)
db_session.add(document)
db_session.flush()
presenter = DocumentJSONPresenter(document)
assert {"title": ["Foo"]} == presenter.asdict()
def test_it_normalizes_document_uris_claimant(req):
docuri_1 = models.DocumentURI(
_claimant="http://example.org/",
_claimant_normalized="http://example.org",
_uri="http://example.org/",
_uri_normalized="http://example.org",
type="self-claim",
)
docuri_2 = models.DocumentURI(
_claimant="http://example.org/",
_claimant_normalized="http://example.org",
_uri="https://example.org/",
_uri_normalized="https://example.org",
type="rel-canonical",
)
req.db.add(models.Document(document_uris=[docuri_1, docuri_2]))
req.db.flush()
normalize_uris.normalize_document_uris(req)
assert docuri_1.claimant_normalized == "httpx://example.org"
assert docuri_2.claimant_normalized == "httpx://example.org"
