def test_reject__ng(self):
file = open(os.path.join(os.path.dirname(__file__), "..", "..", "filter_cases_ng.txt"))
for line in file:
url = line.strip()
self.assertEqual(
(url, True),
(url, IronnewsUtility.reject(url)))
file.close()
def test_get_canonical_url(self):
file = open(os.path.join(os.path.dirname(__file__), "..", "..", "canonical_cases.txt"))
for line in file:
value, expected = re.split(r"\t", line.strip())
self.assertEqual(expected, IronnewsUtility.get_canonical_url(value))
file.close()
from google_news import GoogleNews
from ironnews_utility import IronnewsUtility
print "Content-Type: text/plain"
print ""
KeywordManager.initialize()
keyword = KeywordManager.get()
print keyword.encode("utf-8")
articles = GoogleNews.search(keyword, 30)
for article in articles:
url = article["url"]
title = article["title"]
print "---"
print url
print title.encode("utf-8")
if IronnewsUtility.reject(url):
print "reject!"
continue
url2 = IronnewsUtility.get_canonical_url(url)
if url2 != url:
print "canonical! " + url2
ArticleManager.add(url2, title, Article.CATEGORY_RAIL)
KeywordManager.update(keyword)
io.close() print "Content-Type: text/plain" print "" print "feeder" credential = read_credential() articles = get_articles() urls = [article.url for article in articles] random.shuffle(urls) for original_url in urls[0:3]: print "---" print original_url canonical_url = IronnewsUtility.get_canonical_url(original_url) print canonical_url result1 = add_article(credential, canonical_url) print result1 article_id = result1["result"]["1"]["article_id"] result2 = add_tag(credential, article_id, u"googleニュース") print result2 set_registered(original_url)
