def test_basics(self):
d = Document({
"id": "1",
"title": "Annotations: The Missing Manual",
"link": [
{
"href": "https://peerj.com/articles/53/",
"type": "text/html"
},
{
"href": "https://peerj.com/articles/53.pdf",
"type": "application/pdf"
}
],
})
d.save()
d = Document.fetch("1")
assert_equal(d["title"], "Annotations: The Missing Manual")
assert_equal(len(d['link']), 2)
assert_equal(d['link'][0]['href'], "https://peerj.com/articles/53/")
assert_equal(d['link'][0]['type'], "text/html")
assert_equal(d['link'][1]['href'], "https://peerj.com/articles/53.pdf")
assert_equal(d['link'][1]['type'], "application/pdf")
assert d['created']
assert d['updated']
def includeme(config):
app = Flask('annotator') # Create the annotator-store app
app.register_blueprint(store.store) # and register the store api.
# Set up the models
settings = config.get_settings()
if 'es.host' in settings:
app.config['ELASTICSEARCH_HOST'] = settings['es.host']
if 'es.index' in settings:
app.config['ELASTICSEARCH_INDEX'] = settings['es.index']
es.init_app(app)
with app.test_request_context():
Annotation.create_all()
Document.create_all()
# Configure authentication and authorization
app.config['AUTHZ_ON'] = True
app.before_request(before_request)
# Configure the API views -- version 1 is just an annotator.store proxy
api_v1 = wsgiapp2(app)
config.add_view(api_v1, route_name='api')
if not config.registry.queryUtility(interfaces.IStoreClass):
config.registry.registerUtility(Store, interfaces.IStoreClass)
def test_delete(self):
ann = Document(id=1)
ann.save()
newdoc = Document.fetch(1)
newdoc.delete()
nodoc = Document.fetch(1)
assert nodoc == None
def test_search(self):
# Test search retrieve
d = Document({
"id": "1",
"title": "document",
"link": [peerj["html"], peerj["pdf"]]
})
d.save()
res = Document.search(query={'title': 'document'})
assert_equal(len(res), 1)
def test_uris(self):
d = Document({
"id": "1",
"title": "document",
"link": [peerj["html"], peerj["pdf"]]
})
assert_equal(d.uris(), [
"https://peerj.com/articles/53/",
"https://peerj.com/articles/53.pdf"
])
def test_get_all_by_uri(self):
# add two documents and make sure we can search for both
d = Document({
"id":
"1",
"title":
"document1",
"link": [
{
"href": "https://peerj.com/articles/53/",
"type": "text/html"
},
]
})
d.save()
d = Document({
"id":
"2",
"title":
"document2",
"link": [{
"href": "https://peerj.com/articles/53.pdf",
"type": "application/pdf"
}]
})
d.save()
docs = Document.get_all_by_uris([
"https://peerj.com/articles/53/",
"https://peerj.com/articles/53.pdf"
])
assert_equal(len(docs), 2)
def test_get_by_uri(self):
# create 3 documents and make sure get_by_uri works properly
d = Document({
"id": "1",
"title": "document1",
"link": [
{
"href": "https://peerj.com/articles/53/",
"type": "text/html"
},
{
"href": "https://peerj.com/articles/53.pdf",
"type": "application/pdf"
},
],
})
d.save()
d = Document({
"id": "2",
"title": "document2",
"link": [
{
"href": "https://peerj.com/articles/53/",
"type": "text/html"
},
{
"href": "https://peerj.com/articles/53.pdf",
"type": "application/pdf"
},
],
})
d.save()
d = Document({
"id": "3",
"title": "document3",
"link": [
{
"href": "http://nature.com/123/",
"type": "text/html"
}
],
})
d.save()
doc = Document.get_by_uri("https://peerj.com/articles/53/")
assert doc
assert_equal(doc['title'], "document1")
def test_basics(self):
# Creating a single document and verifies the saved attributes
d = Document({
"id": "1",
"title": "Annotations: The Missing Manual",
"link": [peerj["html"], peerj["pdf"]]
})
d.save()
d = Document.fetch("1")
assert_equal(d["title"], "Annotations: The Missing Manual")
assert_equal(len(d['link']), 2)
assert_equal(d['link'][0]['href'], "https://peerj.com/articles/53/")
assert_equal(d['link'][0]['type'], "text/html")
assert_equal(d['link'][1]['href'], "https://peerj.com/articles/53.pdf")
assert_equal(d['link'][1]['type'], "application/pdf")
assert d['created']
assert d['updated']
def test_search(self):
d = Document({
"id":
"1",
"title":
"document",
"link": [{
"href": "https://peerj.com/articles/53/",
"type": "text/html"
}, {
"href": "https://peerj.com/articles/53.pdf",
"type": "application/pdf"
}],
})
d.save()
res = Document.search(query={'title': 'document'})
assert_equal(len(res), 1)
def test_get_all_by_uri(self):
# add two documents and make sure we can search for both
d = Document({
"id": "1",
"title": "document1",
"link": [
{
"href": "https://peerj.com/articles/53/",
"type": "text/html"
},
]
})
d.save()
d = Document({
"id": "2",
"title": "document2",
"link": [
{
"href": "https://peerj.com/articles/53.pdf",
"type": "application/pdf"
}
]
})
d.save()
docs = Document.get_all_by_uris(["https://peerj.com/articles/53/", "https://peerj.com/articles/53.pdf"])
assert_equal(len(docs), 2)
def test_search(self):
d = Document({
"id": "1",
"title": "document",
"link": [
{
"href": "https://peerj.com/articles/53/",
"type": "text/html"
},
{
"href": "https://peerj.com/articles/53.pdf",
"type": "application/pdf"
}
],
})
d.save()
res = Document.search(title='document')
assert_equal(len(res), 1)
def test_uris(self):
d = Document({
"id":
"1",
"title":
"document",
"link": [{
"href": "https://peerj.com/articles/53/",
"type": "text/html"
}, {
"href": "https://peerj.com/articles/53.pdf",
"type": "application/pdf"
}],
})
assert_equal(d.uris(), [
"https://peerj.com/articles/53/",
"https://peerj.com/articles/53.pdf"
])
def test_uris(self):
d = Document({
"id": "1",
"title": "document",
"link": [
{
"href": "https://peerj.com/articles/53/",
"type": "text/html"
},
{
"href": "https://peerj.com/articles/53.pdf",
"type": "application/pdf"
}
],
})
assert_equal(d.uris(), [
"https://peerj.com/articles/53/",
"https://peerj.com/articles/53.pdf"
])
def test_merge_links(self):
d = Document({
"id": "1",
"title": "document",
"link": [
{
"href": "https://peerj.com/articles/53/",
"type": "text/html"
},
{
"href": "https://peerj.com/articles/53.pdf",
"type": "application/pdf"
}
],
})
d.save()
d = Document.fetch(1)
assert d
assert_equal(len(d['link']), 2)
d.merge_links([
{
"href": "https://peerj.com/articles/53/",
"type": "text/html"
},
{
"href": "http://peerj.com/articles/53.doc",
"type": "application/vnd.ms-word.document"
}
])
d.save()
assert_equal(len(d['link']), 3)
d = Document.fetch(1)
assert d
assert_equal(len(d['link']), 3)
doc = Document.get_by_uri("https://peerj.com/articles/53/")
assert doc
assert_equal(len(doc['link']), 3)
def test_deficient_links(self):
# Test that bad links are not saved
d = Document({
"id": "1",
"title": "Chaos monkey: The messed up links",
"link": [{
"href": "http://cuckoo.baboon/"
}, {
# I'm an empty link entry
}, {
"type": "text/html"
}, {
"href": "http://cuckoo.baboon/",
"type": "text/html"
}]
})
d.save()
d = Document.fetch("1")
assert_equal(len(d['link']), 2)
assert_equal(d['link'][0]['href'], "http://cuckoo.baboon/")
assert_equal(d['link'][1]['href'], "http://cuckoo.baboon/")
assert_equal(d['link'][1]['type'], "text/html")
def test_delete(self):
ann = Document(id=1)
ann.save()
newdoc = Document.fetch(1)
newdoc.delete()
nodoc = Document.fetch(1)
assert nodoc == None
def test_basics(self):
d = Document({
"id":
"1",
"title":
"Annotations: The Missing Manual",
"link": [{
"href": "https://peerj.com/articles/53/",
"type": "text/html"
}, {
"href": "https://peerj.com/articles/53.pdf",
"type": "application/pdf"
}],
})
d.save()
d = Document.fetch("1")
assert_equal(d["title"], "Annotations: The Missing Manual")
assert_equal(len(d['link']), 2)
assert_equal(d['link'][0]['href'], "https://peerj.com/articles/53/")
assert_equal(d['link'][0]['type'], "text/html")
assert_equal(d['link'][1]['href'], "https://peerj.com/articles/53.pdf")
assert_equal(d['link'][1]['type'], "application/pdf")
assert d['created']
assert d['updated']
def test_delete(self):
# Test deleting a document
ann = Document(id=1)
ann.save()
newdoc = Document.fetch(1)
newdoc.delete()
nodoc = Document.fetch(1)
assert nodoc is None
def drop_indices(app):
from .model import Annotation
from annotator.document import Document
with app.test_request_context():
Annotation.drop_all()
Document.drop_all()
def test_merge_links(self):
d = Document({
"id":
"1",
"title":
"document",
"link": [{
"href": "https://peerj.com/articles/53/",
"type": "text/html"
}, {
"href": "https://peerj.com/articles/53.pdf",
"type": "application/pdf"
}],
})
d.save()
d = Document.fetch(1)
assert d
assert_equal(len(d['link']), 2)
d.merge_links([{
"href": "https://peerj.com/articles/53/",
"type": "text/html"
}, {
"href": "http://peerj.com/articles/53.doc",
"type": "application/vnd.ms-word.document"
}])
d.save()
assert_equal(len(d['link']), 3)
d = Document.fetch(1)
assert d
assert_equal(len(d['link']), 3)
doc = Document.get_by_uri("https://peerj.com/articles/53/")
assert doc
assert_equal(len(doc['link']), 3)
def test_get_by_uri(self):
# create 3 documents and make sure get_by_uri works properly
d = Document({
"id":
"1",
"title":
"document1",
"link": [
{
"href": "https://peerj.com/articles/53/",
"type": "text/html"
},
{
"href": "https://peerj.com/articles/53.pdf",
"type": "application/pdf"
},
],
})
d.save()
d = Document({
"id":
"2",
"title":
"document2",
"link": [
{
"href": "https://peerj.com/articles/53/",
"type": "text/html"
},
{
"href": "https://peerj.com/articles/53.pdf",
"type": "application/pdf"
},
],
})
d.save()
d = Document({
"id":
"3",
"title":
"document3",
"link": [{
"href": "http://nature.com/123/",
"type": "text/html"
}],
})
d.save()
doc = Document.get_by_uri("https://peerj.com/articles/53/")
assert doc
assert_equal(doc['title'], "document1")
def test_save_merge_documents(self):
d1 = Document({
"id": "1",
"title": "document1",
"link": [peerj["html"], peerj["pdf"]]
})
d1.save()
d2 = Document({
"id": "2",
"title": "document2",
"link": [peerj["doc"], peerj["docx"]]
})
d2.save()
# They are not merged yet
d1 = Document.fetch(1)
d2 = Document.fetch(2)
assert d1
assert d2
d3 = Document({
"id": "3",
"title": "document3",
"link": [peerj["doc"], peerj["docx"]]
})
d3.save()
# d2 is merged into d3
d2 = Document.fetch(2)
d3 = Document.fetch(3)
assert d2 is None
assert d3
d4 = Document({
"id": "4",
"title": "document4",
"link": [
{
"href": "https://totallydifferenturl.com",
"type": "text/html"
}
]
})
# A new document is created for d4
# It is not merged
d4.save()
d4 = Document.fetch(4)
assert d4
d5 = Document({
"id": "5",
"title": "document5",
"link": [peerj["pdf"], peerj["doc"]]
})
d5.save()
# The documents have been merged into d5
d1 = Document.fetch(1)
d2 = Document.fetch(2)
d3 = Document.fetch(3)
d4 = Document.fetch(4)
d5 = Document.fetch(5)
assert d1 is None
assert d2 is None
assert d3 is None
assert d4
assert d5
def test_new(self):
d = Document()
assert_equal('{}', repr(d))
def test_save(self):
d1 = Document({
"id": "1",
"title": "document1",
"link": [peerj["html"], peerj["pdf"]]
})
d1.save()
d2 = Document({
"id": "2",
"title": "document2",
"link": [peerj["pdf"], peerj["doc"]]
})
d2.save()
d3 = Document({
"id": "3",
"title": "document3",
"link": [peerj["doc"], peerj["docx"]]
})
d3.save()
d4 = Document({
"id": "4",
"title": "document4",
"link": [peerj["docx"]]
})
d4.save()
d1 = Document.fetch(1)
d2 = Document.fetch(2)
d3 = Document.fetch(3)
d4 = Document.fetch(4)
assert d1 is None
assert d2 is None
assert d3 is None
assert d4
def test_merge_links(self):
d = Document({
"id": "1",
"title": "document",
"link": [peerj["html"], peerj["pdf"]]
})
d.save()
d = Document.fetch(1)
assert d
assert_equal(len(d['link']), 2)
d.merge_links([peerj["html"], peerj["doc"]])
d.save()
assert_equal(len(d['link']), 3)
d = Document.fetch(1)
assert d
assert_equal(len(d['link']), 3)
doc = Document.get_by_uri("https://peerj.com/articles/53/")
assert doc
assert_equal(len(doc['link']), 3)
def includeme(config):
"""Include the annotator-store API backend via http or route embedding.
Example INI file:
.. code-block:: ini
[app:h]
api.key: 00000000-0000-0000-0000-000000000000
api.endpoint: https://example.com/api
or use a relative path for the endpoint to embed the annotation store
directly in the application.
.. code-block:: ini
[app:h]
api.endpoint: /api
The default is to embed the store as a route bound to "/api".
"""
app = flask.Flask('annotator') # Create the annotator-store app
app.register_blueprint(store.store) # and register the store api.
settings = config.get_settings()
if 'es.host' in settings:
app.config['ELASTICSEARCH_HOST'] = settings['es.host']
if 'es.index' in settings:
app.config['ELASTICSEARCH_INDEX'] = settings['es.index']
es.init_app(app)
try:
with app.test_request_context():
Annotation.create_all()
Document.create_all()
except socket.error:
raise Exception(
"Can not access ElasticSearch at %s! Are you sure it's running?" %
(app.config["ELASTICSEARCH_HOST"],)
)
# Configure authentication and authorization
app.config['AUTHZ_ON'] = True
app.before_request(before_request)
app.after_request(after_request)
# Configure the API routes
api_config = {'static': True}
api_endpoint = config.registry.settings.get('api.endpoint', None)
api_url = config.registry.settings.get('api.url', api_endpoint)
if api_endpoint is not None:
api_path = api_endpoint.rstrip('/')
api_pattern = '/'.join([api_path, '*subpath'])
# Configure the API views -- version 1 is just an annotator.store proxy
api_v1 = wsgiapp2(app)
config.add_route('api_real', api_pattern)
config.add_view(api_v1, route_name='api_real')
config.add_view(api_v1, name='api_virtual')
if api_url is not None:
api_url = api_url.strip('/')
if urlparse.urlparse(api_url).scheme:
def set_app_url(request, elements, kw):
kw.setdefault('_app_url', api_url)
return (elements, kw)
api_config['pregenerator'] = set_app_url
config.add_route('api', '/*subpath', **api_config)
else:
config.add_route('api', api_url + '/*subpath', **api_config)
if not config.registry.queryUtility(interfaces.IStoreClass):
config.registry.registerUtility(Store, interfaces.IStoreClass)
def test_get_by_uri_not_found(self):
assert Document.get_by_uri("bogus") is None
def test_get_by_uri(self):
# Make sure that only the document with the given uri is retrieved
d = Document({
"id": "1",
"title": "document1",
"link": [peerj["html"], peerj["pdf"]]
})
d.save()
d = Document({
"id": "2",
"title": "document2",
"link": [
{
"href": "http://nature.com/123/",
"type": "text/html"
}
],
})
d.save()
d = Document({
"id": "3",
"title": "document3",
"link": [peerj["doc"]]
})
d.save()
doc = Document.get_by_uri("https://peerj.com/articles/53/")
assert doc
assert_equal(doc['title'], "document1")
def includeme(config):
"""Include the annotator-store API backend via http or route embedding.
Example INI file:
.. code-block:: ini
[app:h]
api.key: 00000000-0000-0000-0000-000000000000
api.endpoint: https://example.com/api
or use a relative path for the endpoint to embed the annotation store
directly in the application.
.. code-block:: ini
[app:h]
api.endpoint: /api
The default is to embed the store as a route bound to "/api".
"""
app = flask.Flask('annotator') # Create the annotator-store app
app.register_blueprint(store.store) # and register the store api.
settings = config.get_settings()
if 'es.host' in settings:
app.config['ELASTICSEARCH_HOST'] = settings['es.host']
if 'es.index' in settings:
app.config['ELASTICSEARCH_INDEX'] = settings['es.index']
es.init_app(app)
try:
with app.test_request_context():
Annotation.create_all()
Document.create_all()
except socket.error:
raise Exception(
"Can not access ElasticSearch at %s! Are you sure it's running?" %
(app.config["ELASTICSEARCH_HOST"], ))
except:
with app.test_request_context():
Annotation.update_settings()
Annotation.create_all()
Document.create_all()
# Configure authentication and authorization
app.config['AUTHZ_ON'] = True
app.before_request(before_request)
app.after_request(after_request)
# Configure the API routes
api_config = {'static': True}
api_endpoint = config.registry.settings.get('api.endpoint', None)
api_url = config.registry.settings.get('api.url', api_endpoint)
if api_endpoint is not None:
api_path = api_endpoint.rstrip('/')
api_pattern = '/'.join([api_path, '*subpath'])
# Configure the API views -- version 1 is just an annotator.store proxy
api_v1 = wsgiapp2(app)
config.add_route('api_real', api_pattern)
config.add_view(api_v1, route_name='api_real')
config.add_view(api_v1, name='api_virtual')
if api_url is not None:
api_url = api_url.strip('/')
if urlparse.urlparse(api_url).scheme:
def set_app_url(request, elements, kw):
kw.setdefault('_app_url', api_url)
return (elements, kw)
api_config['pregenerator'] = set_app_url
config.add_route('api', '/*subpath', **api_config)
else:
config.add_route('api', api_url + '/*subpath', **api_config)
if not config.registry.queryUtility(interfaces.IStoreClass):
config.registry.registerUtility(Store, interfaces.IStoreClass)
def drop_indices(app):
from .model import Annotation
from annotator.document import Document
with app.test_request_context():
Annotation.drop_all()
Document.drop_all()
