def post(self):
subject = self.request.get("subject")
content = self.request.get("content")
verify_subject = len(subject) > 0
verify_content = len(content) > 0
if verify_content and verify_subject:
entry = BlogEntry( subject = subject, content = content )
entry.put()
self.redirect(self.app.config.get('url_base') + "/" + str(entry.key().id()))
else:
self.generate_newpost_page(
subject,
"" if verify_subject else "Missing subject.",
content,
"" if verify_content else "Missing content.")
def transform(collection):
output_collection = list()
for entry in collection:
output_collection.append(BlogEntry(StripPunctuation.eliminate_punctuation(entry.title()), \
entry.date(), entry.url(), StripPunctuation.eliminate_punctuation(entry.text()),
entry.source, entry.crawl_url))
return BlogEntryCollection(output_collection)
def get(self, post_id):
entry = BlogEntry.get_by_id(int(post_id))
if entry:
self.response.headers.add_header("Content-Type", "application/json")
self.response.out.write(entry.to_json())
else:
self.redirect(URL_BASE)
def get(self):
entries = BlogEntry.get_last_10_json()
if entries:
self.response.headers.add_header("Content-Type", "application/json")
self.response.out.write(entries)
else:
self.redirect(URL_BASE)
def get(self):
entries = BlogEntry.get_last_10()
self.render("mainpage.html",
newpost_url=self.app.config.get('url_newpost'),
single_url=self.app.config.get('url_base')+"/",
title="My Blog",
entries=entries)
def test_construct_with_nones(self):
date = datetime.now()
b = BlogEntry(None, date, None, None, None, None)
self.assertEqual(b.title(), None)
self.assertEqual(b.date(), date)
self.assertEqual(b.url(), None)
self.assertEqual(b.text(), None)
self.assertEqual(b.source(), None)
self.assertEqual(b.crawl_url(), None)
def test_construct_with_date(self):
date = datetime.now()
b = BlogEntry("title", date, "url", "raw_text", "source", "crawl_url")
self.assertEqual(b.title(), "title")
self.assertEqual(b.date(), date)
self.assertEqual(b.url(), "url")
self.assertEqual(b.text(), "raw_text")
self.assertEqual(b.source(), "source")
self.assertEqual(b.crawl_url(), "crawl_url")
def test_construct_with_invalid_date(self):
error = False
not_a_valid_date = "not a valid date"
try:
b = BlogEntry(None, not_a_valid_date, None, None, None, None)
except ValueError:
error = True
self.assertTrue(error)
def transform(collection):
output_collection = list()
for entry in collection:
output_collection.append(BlogEntry(TextUtil.to_lower_case(entry.title()), \
entry.date(), entry.url(), TextUtil.to_lower_case(entry.text()),
entry.source, entry.crawl_url))
return BlogEntryCollection(output_collection)
def get(self):
entries, cache_time = BlogEntry.get_last_10_posts()
self.render("mainpage.html",
newpost_url=self.app.config.get('url_newpost'),
single_url=self.app.config.get('url_base')+"/",
title="My Blog",
entries=entries,
cache_time=build_cache_time_str(cache_time) )
def get(self, post_id):
entry = BlogEntry.get_by_id(int(post_id))
if entry:
self.render("singlepost.html",
subject=entry.subject,
content=entry.content,
main_url=self.app.config.get('url_base')
)
else:
self.redirect(self.app.config.get('url_base'))
def get(self, post_id):
post, post_time = BlogEntry.get_post_by_id(int(post_id))
if post:
self.render("singlepost.html",
subject=post.subject,
content=post.content,
main_url=self.app.config.get('url_base'),
cache_time=build_cache_time_str(post_time)
)
else:
self.redirect(self.app.config.get('url_base'))
class TestBlogEntryCollection(unittest.TestCase):
blog_entry_1 = BlogEntry("title", datetime.now(), "url", "raw_text", "source", "crawl_url")
blog_entry_2 = BlogEntry("title", datetime.now(), "url", "raw_text", "source", "crawl_url")
def test_invalid_entries(self):
error = False
try:
BlogEntryCollection(TestBlogEntryCollection.blog_entry_1)
except ValueError:
error = True
self.assertTrue(error)
def test_construct(self):
entries = [TestBlogEntryCollection.blog_entry_1, TestBlogEntryCollection.blog_entry_2]
bc = BlogEntryCollection(entries)
self.assertEquals(2, bc.size())
for entry in bc:
self.assertIn(entry, entries)
for entry in entries:
self.assertIn(entry, bc)
def test_load_from_json(self):
date = datetime.now()
date_string = str(date)
json_entry_string = '\
{ \
"title": ["title"], \
"url": ["url"], \
"timestamp": ["' + date_string + '"], \
"raw_content": ["raw", "text"], \
"source": "source", \
"crawl_url": "crawl_url"\
}'
json_collection_string = '[' + json_entry_string + ', ' + json_entry_string + ']'
json_object = json.loads(json_collection_string)
bc = BlogEntryCollection.from_json_object(json_object)
self.assertEquals(2, bc.size())
def test_transform(self):
date = datetime.now()
b = BlogEntry("title", date, "url", "raw_text", "source", "crawl_url")
collection = BlogEntryCollection([b])
collection1 = DiscardBody.transform(collection)
collection1 = ToLowerCase.transform(collection1)
collection1 = StripPunctuation.transform(collection1)
transformed_entry = collection1[0]
expected_entry = BlogEntry("title", date, "url", "", "source",
"crawl_url")
self.assertEquals(expected_entry.title(), transformed_entry.title())
self.assertEquals(expected_entry.date(), transformed_entry.date())
self.assertEquals(expected_entry.url(), transformed_entry.url())
def test_load_from_json(self):
date = datetime.now()
date_string = str(date)
json_string = '\
{ \
"title": ["title"], \
"url": ["url"], \
"timestamp": ["' + date_string + '"], \
"raw_content": ["raw", "text"], \
"source": "source", \
"crawl_url": "crawl_url"\
}'
json_object = json.loads(json_string)
b = BlogEntry.from_json_object(json_object)
self.assertEqual(b.title(), "title,")
self.assertEqual(b.date(), date)
self.assertEqual(b.url(), "url,")
self.assertEqual(b.text(), "raw,text,")
self.assertEqual(b.source(), "source")
self.assertEqual(b.crawl_url(), "crawl_url")
def transform(collection, foodwords_list, neighboring_words = 2):
collection = ToLowerCase.transform(collection)
collection = StripPunctuation.transform(collection)
output_collection = list()
for entry in collection:
raw_words = TextUtil.make_lowercase_word_vector(entry.text()) #, unique = False, stemming = True
tot_words = len(raw_words)
index = neighboring_words
raw_content_words_list = list()
while index <= (tot_words - neighboring_words):
if TextUtil.list_contains(raw_words[index], foodwords_list):
sub_sentence = raw_words[(index-neighboring_words):(index+neighboring_words+1)]
for word in sub_sentence:
raw_content_words_list.append(word)
index += neighboring_words
else:
index += 1
output_collection.append(BlogEntry(entry.title(), entry.date(), entry.url(), set(raw_content_words_list)))
return output_collection
def transform(collection):
output_collection = list()
for entry in collection:
output_collection.append(BlogEntry(entry.title(), entry.date(), entry.url(), '', entry.source(), entry.crawl_url()))
return BlogEntryCollection(output_collection)