def test_papers_from_embedded_script(mock_get):
mock_get.return_value = MockResponse()
papers = papers_from_embedded_script("http://www.arxiv-sanity.com/")
assert papers is not None
assert len(papers) is 200
assert papers[0]['title'] == 'Efficient Action Detection in Untrimmed Videos via Multi-Task Learning'
def get_similar(self, pid):
""" Get papers similar in content to the named paper. """
logging.info("Getting ArXiv papers similar to paper: {}".format(pid))
similarURL = ASP_BaseURL + str(pid)
similar_papers = papers_from_embedded_script(similarURL)[1:] # TODO is the searched paper always first?
attached_papers = []
for p in similar_papers:
attached_papers.append(paper_snippet(p))
return build_message(text="*Your Library*", markdown=False, parts=attached_papers)
def get_recommended(self, session):
""" Get the papers recommended to the user based on their
saved papers and their search domain. """
# TODO the /recommend endpoint has FILTERS
logging.info("Getting recommended ArXiv papers for user.")
recommendedURL = ASP_BaseURL + "/recommend"
papers = papers_from_embedded_script(recommendedURL, session)
attached_papers = []
for i, p in enumerate(papers):
attached_papers.append(paper_snippet(p, i + 1))
return build_message( text="*Your Recommended*", markdown=False, parts=attached_papers), attached_papers[:5]
def get_paper(self, paper_id):
""" Return specified paper from within the set. """
if not self.previous_attachments:
logging.info("Can't retreive paper. No papers in current context.")
return build_message(text="*Can't retreive paper*"), None
else:
pid = re.findall('[0-9]+\.?[0-9]*', self.previous_attachments[paper_id - 1]['text'])[0]
logging.info("Getting paper: {} from ArXiv.".format(pid))
paperURL = ASP_BaseURL + "/" + str(pid)
paper = papers_from_embedded_script(paperURL)[0] # only get first, rest are related papers
return build_message(text="*Here's your paper*", markdown=False, parts=None), [paper_snippet(paper, 1, include_abstract=True)]
def get_most_recent(self, session):
""" Get most recently published papers within the user's search
domain.
"""
logging.info("Getting most recent ArXiv papers within user's domain.")
papers = papers_from_embedded_script(ASP_BaseURL+"/", session)
attached_papers = []
for i, p in enumerate(papers):
attached_papers.append(paper_snippet(p, i + 1))
return build_message( text="*Your Most Recent*", markdown=False, parts=attached_papers), attached_papers[:5]
def clear_library(self, session):
""" Remove all saved paper's from the users library. """
logging.info("Clearing ArXiv library.")
libraryURL = ASP_BaseURL + "/library"
papers = papers_from_embedded_script(libraryURL)
toggleURL = ASP_BaseURL + "/libtoggle"
for p in papers:
# toggle off each paper from library
r = session.post(toggleURL, data = {'pid':p["pid"]})
if r.status_code != 200:
# TODO log error. pass up to user?
pass
def get_top_recent(self, session):
"""
Get the 'top' recent papers within the user's search domain.
"""
# TODO the /top endpoint has FILTERS
logging.info("Getting top recent ArXiv papers in user's domain.")
topURL = ASP_BaseURL + "/top" # default filters
papers = papers_from_embedded_script(topURL, session=session)
attached_papers = []
for i, p in enumerate(papers):
attached_papers.append(paper_snippet(p, i + 1))
return build_message( text="*Your Library*", markdown=False, parts=attached_papers), attached_papers[:5]
def get_library(self, session):
""" Get all papers saved by the user. Their 'library'. """
logging.info("Getting ArXiv library.")
libraryURL = ASP_BaseURL + "/library"
try:
papers = papers_from_embedded_script(libraryURL, session=session)
except ReadTimeout:
return build_message( text="*Connection Problem. Please try again*", markdown=False, parts=None)
attached_papers = []
for i, p in enumerate(papers):
attached_papers.append(paper_snippet(p, i + 1))
return build_message( text="*Your Library*", markdown=False, parts=attached_papers), attached_papers
def search_arxiv(self, query, num_papers=5):
""" Search arxiv papers by search string. """
logging.info("Searching ArXiv for: {}".format(query))
tokens = query.split(' ')
searchEndpoint = 'search?q=' + '+'.join(tokens)
searchURL = ASP_BaseURL + searchEndpoint
papers = papers_from_embedded_script(searchURL)
bot_uid = self.clients.bot_user_id()
attached_papers = []
# For each paper
for i in range(min(num_papers,len(papers))):
attached_papers.append(paper_snippet(papers[i], i + 1))
attached_papers[-1]['footer'] = "> `<@" + bot_uid + "> show more papers` - to see more papers"
# build the json message object
return build_message( text="*Search Results*", markdown=False, parts=attached_papers), attached_papers