def test_detect_multimatch_for_query_and_single_match_for_content(self):
CONTENT_TO_MATCH = "oui"
FAILING_MATCH = "non"
TARGET_MATCH = '<div class="cible" id="bob">{content}</div>'.format(
content=CONTENT_TO_MATCH)
TARGET_NO_MATCH = '<div class="cible" id="bob">{content}</div>'.format(
content=FAILING_MATCH)
PAGE = """<!DOCTYPE html>
<html lang="fr">
<head>
<meta charset="utf-8">
<title>test page</title>
</head>
<body>
{target_1}
{target_2}
</body>
</html>""".format(target_1=TARGET_MATCH, target_2=TARGET_NO_MATCH)
QUERY = ".cible#bob"
res = mincer.utils.extract_content_from_html(QUERY, CONTENT_TO_MATCH,
PAGE)
assert is_div(res)
assert CONTENT_TO_MATCH in res
def test_return_a_no_result_partial_if_no_result_are_found(
self, client, tmp_db, fake_serv, fake_prov):
QUERY = "search without result"
URL = self._build_url_from_query(QUERY)
response = client.get(URL)
# We have an answer...
assert response.status_code == OK
# Any web page can use this content
assert response.headers["Access-Control-Allow-Origin"] == "*"
# ...it's an HTML document...
assert response.mimetype == "text/html"
# Let's convert it for easy inspection
data = response.get_data(as_text=True)
# ...containing a <div> with correct class and id
assert is_div(data,
cls_name=HtmlClasses.NO_RESULT,
id_name=fake_prov.slug)
# And we have the provider info in it
assert has_div_with_class(data, cls_name=HtmlClasses.PROVIDER)
prov_data = all_div_content(data, query=HtmlClasses.provider_query())
assert is_substring_in(fake_prov.name, prov_data)
REMOTE_URL = fake_prov.remote_url.format(param=quote_plus(QUERY))
assert is_substring_in(REMOTE_URL, prov_data)
def test_return_a_no_result_partial_if_no_result_are_found(
self, client, tmp_db, koha_search_prov):
# This search returns absolutly no result
SEARCH_QUERY = 'zxkml'
URL = self._build_url(SEARCH_QUERY)
response = client.get(URL)
# We have an answer...
assert response.status_code == OK
# Any web page can use this content
assert response.headers["Access-Control-Allow-Origin"] == "*"
# ...it's an HTML document...
assert response.mimetype == "text/html"
# Let's convert it for easy inspection
data = response.get_data(as_text=True)
# ...containing only a <div>
assert is_div(data, cls_name=HtmlClasses.NO_RESULT)
# And we have the provider info in it
assert has_div_with_class(data, cls_name=HtmlClasses.PROVIDER)
prov_data = all_div_content(data, query=HtmlClasses.provider_query())
assert is_substring_in(koha_search_prov.name, prov_data)
REMOTE_URL = koha_search_prov.remote_url.format(
param=quote_plus(SEARCH_QUERY))
assert is_substring_in(dominescape(REMOTE_URL), prov_data)
def test_search_works_with_unicode_query(self, client, tmp_db,
koha_search_prov):
# This search returns only a few results (in japanese)
SEARCH_QUERY = '龍 車 日' # dragon car day
url = self._build_url(SEARCH_QUERY)
response = client.get(url)
# We have an answer...
assert response.status_code == OK
# Any web page can use this content
assert response.headers["Access-Control-Allow-Origin"] == "*"
# ...it's an HTML document...
assert response.mimetype == "text/html"
# Let's convert it for easy inspection
data = response.get_data(as_text=True)
# ...containing only a <div>
assert is_div(data, cls_name=HtmlClasses.RESULT)
# And we have the correct books in it
results = all_div_content(data, query=HtmlClasses.result_item_query())
assert is_substring_in("新疆史志", results)
assert is_substring_in("永井龍男集", results)
def test_return_result_partial_if_result_are_found(self, client, tmp_db,
koha_booklist_prov):
# We are using the ID of of an existing list
LIST_ID = "9896"
URL = self._build_url(LIST_ID)
response = client.get(URL)
# We have an answer...
assert response.status_code == OK
# Any web page can use this content
assert response.headers["Access-Control-Allow-Origin"] == "*"
# ...it's an HTML document
assert response.mimetype == "text/html"
# Let's convert it for easy inspection
data = response.get_data(as_text=True)
# ...containing only a <div>
assert is_div(data, cls_name=HtmlClasses.RESULT)
# And we have the provider info in it
assert has_div_with_class(data, cls_name=HtmlClasses.PROVIDER)
prov_data = all_div_content(data, query=HtmlClasses.provider_query())
assert is_substring_in(koha_booklist_prov.name, prov_data)
REMOTE_URL = koha_booklist_prov.remote_url.format(
param=quote_plus(LIST_ID))
assert is_substring_in(dominescape(REMOTE_URL), prov_data)
# And we have the correct books in it
results = all_div_content(data, query=HtmlClasses.result_item_query())
assert is_substring_in("Africa in Russia, Russia in Africa", results)
assert is_substring_in("Cahiers d'études africaines", results)
assert is_substring_in("Étudier à l'Est", results)
assert is_substring_in("Forced labour in colonial Africa", results)
assert is_substring_in("Le gel", results)
assert is_substring_in(
"Revue européenne des migrations internationales", results)
assert is_substring_in("The Cold War in the Third World", results)
def test_detect_matching_content(self):
CONTENT_TO_MATCH = "ca va matcher"
TARGET_NODE = '<div class="cible" id="bob">{content}</div>'.format(
content=CONTENT_TO_MATCH)
PAGE = """<!DOCTYPE html>
<html lang="fr">
<head>
<meta charset="utf-8">
<title>test page</title>
</head>
<body>
{node}
</body>
</html>""".format(node=TARGET_NODE)
QUERY = ".cible#bob"
res = mincer.utils.extract_content_from_html(QUERY, CONTENT_TO_MATCH,
PAGE)
assert is_div(res)
assert CONTENT_TO_MATCH in res
def test_search_works(self, client, tmp_db, koha_search_prov):
# This search returns only a few results
SEARCH_QUERY = 'afrique voiture'
URL = self._build_url(SEARCH_QUERY)
response = client.get(URL)
# We have an answer...
assert response.status_code == OK
# Any web page can use this content
assert response.headers["Access-Control-Allow-Origin"] == "*"
# ...it's an HTML document...
assert response.mimetype == "text/html"
# Let's convert it for easy inspection
data = response.get_data(as_text=True)
# TODO add this new assert to the other query related tests
# ...containing a <div> with correct class and id
assert is_div(data, cls_name=HtmlClasses.RESULT, id_name="koha-search")
# And we have the provider info in it
assert has_div_with_class(data, cls_name=HtmlClasses.PROVIDER)
prov_data = all_div_content(data, query=HtmlClasses.provider_query())
assert is_substring_in(koha_search_prov.name, prov_data)
REMOTE_URL = koha_search_prov.remote_url.format(
param=quote_plus(SEARCH_QUERY))
assert is_substring_in(dominescape(REMOTE_URL), prov_data)
# And we have the correct books in it
results = all_div_content(data, query=HtmlClasses.result_item_query())
assert is_substring_in(
"Différenciation régionale et régionalisation en Afrique francophone et à Madagascar",
results)
assert is_substring_in("Qui se nourrit de la famine en Afrique ?",
results)
assert is_substring_in("Les jachères en Afrique tropicale", results)
def test_search_works_with_unicode_query(self, client, tmp_db, fake_serv,
fake_prov):
# A query with some japanese
QUERY = "search with unicode 龍 車 日" # dragon car day
URL = self._build_url_from_query(QUERY)
response = client.get(URL)
# We have an answer...
assert response.status_code == OK
# Any web page can use this content
assert response.headers["Access-Control-Allow-Origin"] == "*"
# ...it's an HTML document...
assert response.mimetype == "text/html"
# Let's convert it for easy inspection
data = response.get_data(as_text=True)
# ...containing a <div> with correct class and id
assert is_div(data,
cls_name=HtmlClasses.RESULT,
id_name=fake_prov.slug)
# And we have the provider info in it
assert has_div_with_class(data, cls_name=HtmlClasses.PROVIDER)
prov_data = all_div_content(data, query=HtmlClasses.provider_query())
assert is_substring_in(fake_prov.name, prov_data)
REMOTE_URL = fake_prov.remote_url.format(param=quote_plus(QUERY))
assert is_substring_in(REMOTE_URL, prov_data)
# And we have the correct books in it
results = all_div_content(data,
query=".{surrounding} .{item}".format(
surrounding=HtmlClasses.RESULT,
item=HtmlClasses.RESULT_ITEM))
assert is_substring_in("Result with japanese 新疆史志", results)
assert is_substring_in("Result with japanese 永井龍男集", results)