Add dashes around html entities in escape_lemma

Part of #39
omwn · Oct 18, 2024 · 23fa3b3 · 23fa3b3
1 parent 0a94425
commit 23fa3b3
Show file tree

Hide file tree

Showing 3 changed files with 11 additions and 1 deletion.
diff --git a/scripts/util.py b/scripts/util.py
@@ -65,7 +65,7 @@ def escape_lemma(lemma: str) -> str:
         elif c in _custom_char_escapes:
             chars.append(_custom_char_escapes[c])
         elif codepoint in codepoint2name:
-            chars.append(codepoint2name[codepoint])
+            chars.append(f"-{codepoint2name[codepoint]}-")
         else:
             esc = f'-{codepoint:04X}-'
             warnings.warn(f'no escape character defined for {c!r}; using {esc}')

diff --git a/tests/__init__.py b/tests/__init__.py
diff --git a/tests/test_util.py b/tests/test_util.py
@@ -0,0 +1,10 @@
+from scripts.util import escape_lemma
+
+def test_escape_lemma():
+    assert escape_lemma("abc") == "abc"
+    assert escape_lemma("a.b.c") == "a.b.c"
+    assert escape_lemma("protégé") == "protégé"
+    assert escape_lemma("a b c") == "a_b_c"
+    assert escape_lemma("a:b:c") == "a-colon-b-colon-c"
+    assert escape_lemma("a-b-c") == "a--b--c"
+    assert escape_lemma("a´b´c") == "a-acute-b-acute-c"