Merge c789bc5 into 52c4908

alexandrainst · Nov 18, 2021 · 7cf4a06 · 7cf4a06
2 parents 52c4908 + c789bc5
commit 7cf4a06
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 2 deletions.
diff --git a/danlp/datasets/unimorph.py b/danlp/datasets/unimorph.py
@@ -27,7 +27,7 @@ def __init__(self, cache_dir=DEFAULT_CACHE_DIR, verbose=False):
         self.database = pd.read_csv(self.file_path, 
                                 sep='\t', 
                                 names=['lemma', 'form', 'feats'], 
-                                encoding='unicode_escape', 
+                                encoding='utf-8',
                                 usecols=[0,1,2],
                                 dtype={'lemma':str, 'form':str, 'feats':str})
 

diff --git a/tests/test_datasets.py b/tests/test_datasets.py
@@ -201,7 +201,7 @@ def test_unimorph(self):
         self.assertEqual(unimorph.get_inflections('svar'), ['svaredes', 'svarede', 'svarer', 'svares', 'svare', 'svar'])
         self.assertEqual(unimorph.get_inflections('trolde', pos='V'), ['troldedes', 'troldede', 'trolder', 'troldes', 'trolde', 'trold'])
         self.assertEqual(unimorph.get_inflections('trolde', pos='N', with_features=True)[0], {'lemma': 'trold', 'form': 'troldene', 'feats': 'N;DEF;NOM;PL', 'pos': 'N'})
-        self.assertEqual(unimorph.get_lemmas('papiret', with_features=True), [{'lemma': 'papir', 'form': 'papiret', 'feats': 'N;DEF;NOM;SG', 'pos': 'N'}])
+        self.assertEqual(unimorph.get_lemmas('nytårsaftens', with_features=True), [{'lemma': 'nytårsaften', 'form': 'nytårsaftens', 'feats': 'N;INDF;GEN;SG', 'pos': 'N'}])
 
 class TestDanedDatasets(unittest.TestCase):
     def test_daned(self):