fix get_doc_char_span and covering span (#78)

PKSHATechnology-Research · Aug 15, 2020 · 2ac9db9 · 2ac9db9
1 parent 2890576
commit 2ac9db9
Show file tree

Hide file tree

Showing 2 changed files with 15 additions and 11 deletions.
diff --git a/camphr/utils.py b/camphr/utils.py
@@ -54,11 +54,11 @@ def token_from_char_pos(doc: Doc, i: int) -> Token:
     return doc[bisect.bisect(token_idxs, i) - 1]
 
 
-def _get_covering_span(doc: Doc, i: int, j: int) -> Span:
+def _get_covering_span(doc: Doc, i: int, j: int, **kwargs) -> Span:
     token_idxs = [t.idx for t in doc]
     i = bisect.bisect(token_idxs, i) - 1
     j = bisect.bisect_left(token_idxs, j)
-    return doc[i:j]
+    return Span(doc, i, j, **kwargs)
 
 
 def destruct_token(doc: Doc, *char_pos: int) -> Doc:
@@ -84,7 +84,7 @@ def get_doc_char_span(
     """
     span = doc.char_span(i, j, **kwargs)
     if not span and covering:
-        span = _get_covering_span(doc, i, j)
+        span = _get_covering_span(doc, i, j, **kwargs)
     if not span and destructive:
         destruct_token(doc, i, j)
         span = doc.char_span(i, j, **kwargs)

diff --git a/tests/test_utils.py b/tests/test_utils.py
@@ -42,16 +42,20 @@ def test_split_keepsep(text, sep, expected):
 
 
 @pytest.mark.parametrize(
-    "tokens,i,j,destructive,covering,expected",
+    "tokens,i,j,destructive,covering,expected,label",
     [
-        (["Foo", "bar", "baz"], 0, 2, True, False, "Fo"),
-        (["Foo", "bar", "baz"], 0, 5, True, False, "Foo b"),
-        (["Foo", "bar", "baz"], 0, 5, False, True, "Foo bar"),
-        (["Foo", "bar", "baz"], 0, 3, False, True, "Foo"),
-        (["Foo", "bar", "baz"], 1, 5, False, True, "Foo bar"),
+        (["Foo", "bar", "baz"], 0, 2, True, False, "Fo", None),
+        (["Foo", "bar", "baz"], 0, 5, True, False, "Foo b", None),
+        (["Foo", "bar", "baz"], 0, 5, False, True, "Foo bar", None),
+        (["Foo", "bar", "baz"], 0, 3, False, True, "Foo", None),
+        (["Foo", "bar", "baz"], 1, 5, False, True, "Foo bar", "LABEL"),
     ],
 )
-def test_get_doc_char_span(vocab, tokens, i, j, destructive, covering, expected):
+def test_get_doc_char_span(vocab, tokens, i, j, destructive, covering, expected, label):
     doc = Doc(vocab, tokens)
-    span = get_doc_char_span(doc, i, j, destructive=destructive, covering=covering)
+    span = get_doc_char_span(
+        doc, i, j, destructive=destructive, covering=covering, label=label or ""
+    )
     assert span.text == expected
+    if label is not None:
+        assert span.label_ == label