def test_store_stanford(self):
# get tokenvalues etc from stanford test case
aa = amcattest.create_test_analysed_article()
as1, as2 = [
amcattest.create_test_analysis_sentence(analysed_article=aa)
for _i in range(2)
]
from amcat.nlp import stanford
tokens, triples, corefsets = stanford.interpret_xml(
[as1.id, as2.id], stanford.TestStanford._get_test_xml())
store_analysis(aa, tokens, triples, corefsets)
self.assertEqual({str(t.word.lemma)
for t in as1.tokens.all()}, {"Mary", "meet", "John"})
self.assertEqual(
{(str(t.parent.word), str(t.child.word), str(t.relation))
for t in as2.triples}, {('likes', 'She', 'nsubj'),
('likes', 'him', 'dobj')})
self.assertEqual(
{
frozenset(str(t.word.lemma) for t in c.tokens.all())
for c in aa.coreferencesets.all()
}, {frozenset(["Mary", "she"]),
frozenset(["John", "he"])})
def _get_test_script(self, words=None, codes={}, window_size=5):
aa = amcattest.create_test_analysed_article()
tokens = self._get_test_tokens(aa, words)
aset = amcattest.create_test_set(articles=[aa.article])
lexicon_lang = Language.objects.get(pk=2)
cb = self._get_test_codebook(lexicon_lang, codes)
return WindowedSNAScript(articleset=aset.id, plugin=aa.plugin.id,
codebook=cb.id, lexicon_language=lexicon_lang.id,
window_size=window_size)
def _get_test_script(self, words=None, codes={}, window_size=5):
aa = amcattest.create_test_analysed_article()
tokens = self._get_test_tokens(aa, words)
aset = amcattest.create_test_set(articles=[aa.article])
lexicon_lang = Language.objects.get(pk=2)
cb = self._get_test_codebook(lexicon_lang, codes)
return WindowedSNAScript(articleset=aset.id,
plugin=aa.plugin.id,
codebook=cb.id,
lexicon_language=lexicon_lang.id,
window_size=window_size)
def test_store_stanford(self):
# get tokenvalues etc from stanford test case
aa = amcattest.create_test_analysed_article()
as1, as2 = [amcattest.create_test_analysis_sentence(analysed_article=aa) for _i in range(2)]
from amcat.nlp import stanford
tokens, triples, corefsets = stanford.interpret_xml([as1.id, as2.id], stanford.TestStanford._get_test_xml())
store_analysis(aa, tokens, triples, corefsets)
self.assertEqual({str(t.word.lemma) for t in as1.tokens.all()}, {"Mary", "meet", "John"})
self.assertEqual({(str(t.parent.word), str(t.child.word), str(t.relation)) for t in as2.triples},
{('likes', 'She', 'nsubj'), ('likes', 'him', 'dobj')})
self.assertEqual({frozenset(str(t.word.lemma) for t in c.tokens.all()) for c in aa.coreferencesets.all()},
{frozenset(["Mary", "she"]), frozenset(["John", "he"])})