def retrieve_element(self, element, config):
if self.is_aggregate():
og_folder = Path(config["source"])
return Etype.Any(og_folder.name,
paths=[x[1] for x in self.results])
else:
return Etype.Any(element.id, paths=[element.path])
def post_analyse(self, elements):
for el in elements:
el_json = el.paths[0]
with open(el_json) as f:
tweets = json.load(f)
initial_tweet = tweets[0]
self.logger(f"Adding tweet {initial_tweet['id']} to graph...")
self.add_to_graph(initial_tweet)
for tweet in tweets[1:]:
self.logger(f"Adding reply {tweet['id']} to graph...")
self.add_to_graph(tweet, inreplyto=initial_tweet)
xlsx_path = TMP / "final.xlsx"
self.graph.to_xlsx(xlsx_path)
return Etype.Any("FINAL", xlsx_path)
def post_analyse(self, _):
# TODO: a kind of hack... should maybe make available as a func, i.e. `self.get_analysed()`
analysed_els = self.disk.read_elements([self.dest_q])
for el in analysed_els:
el_json = el.paths[0]
with open(el_json) as f:
tweets = json.load(f)
initial_tweet = tweets[0]
self.logger(f"Adding tweet {initial_tweet['id']} to graph...")
self.add_to_graph(initial_tweet)
for tweet in tweets[1:]:
self.logger(f"Adding reply {tweet['id']} to graph...")
self.add_to_graph(tweet, inreplyto=initial_tweet)
xlsx_path = TMP / "final.xlsx"
self.graph.to_xlsx(xlsx_path)
return Etype.Any("FINAL", xlsx_path)
def test_Any(base):
e = Etype.Any(base.id, [base.txt1])
assert len(e.paths) == 1
e = Etype.Any(base.id, [base.txt1, base.md1, base.im3])
assert len(e.paths) == 3
