[glyphdata] Improve ligature splitting

Handle names like “moMa_underscore-thai” where it should be split into
“moMa-thai” and plain “underscore” not “underscore-thai”. I updated the
test expectation because Glyphs is just bing silly building production
names inconsistently.

Lib/glyphsLib/glyphdata.py

@@ -295,16 +295,29 @@ def _translate_category(glyph_name, unicode_category):
 
 
 def _split_ligature_glyph_name(name, data):
+    # Split name to ligature parts
     parts = name.split("_")
-    if len(parts) > 1 and "-" in parts[-1]:
-        script = parts[-1].rsplit("-", 1)[-1]
-        for i, part in enumerate(parts):
-            new = f"{part}-{script}"
-            if "-" in part:
-                continue
-            if _lookup_attributes(part, data) and not _lookup_attributes(new, data):
-                continue
-            parts[i] = new
+
+    # If the last part has a script suffix, strip it and re-split the name.
+    if "-" in parts[-1]:
+        base, script = name.rsplit("-", 1)
+        parts = base.split("_")
+
+        # If there is more than one part, try adding the script suffix to each
+        # part, if this results in a known glyph name, use it as the part name.
+        if len(parts) > 1:
+            for i, part in enumerate(parts):
+                new = f"{part}-{script}"
+                # If the part already has a script suffix, keep it unchanged.
+                if "-" in part:
+                    continue
+                # If the non suffixed name exists and the suffixed name does
+                # not exist, keep the part name unchanged.
+                if _lookup_attributes(part, data) and not _lookup_attributes(new, data):
+                    continue
+                parts[i] = new
+    else:
+        parts = name.split("_")
     return parts
 
 

tests/glyphdata_test.py

@@ -315,14 +315,14 @@ def test_glyphdata_no_duplicates(self):
     "ma_uMatra-tamil": "uni0BAE0BC1",
     "ma_uuMatra-tamil": "uni0BAE0BC2",
     "mo-khmer.below.ro": "uni17D21798.ro",
-    "moMa_underscore-thai": "uni0E21_uni005F",
+    "moMa_underscore-thai": "uni0E21005F",  # uni0E21_uni005F
     "na_iMatra-tamil": "uni0BA80BBF",
     "na_uMatra-tamil": "uni0BA80BC1",
     "na_uuMatra-tamil": "uni0BA80BC2",
     "ng_ya-deva": "uni0919094D092F",
     "nga_uMatra-tamil": "uni0B990BC1",
     "nga_uuMatra-tamil": "uni0B990BC2",
-    "ngoNgu_underscore-thai": "uni0E07_uni005F",
+    "ngoNgu_underscore-thai": "uni0E07005F",  #  uni0E07_uni005F
     "niggahita_maiCatawa-lao": "uni0ECD0ECB",
     "niggahita_maiCatawa-lao.right": "uni0ECD0ECB.right",
     "niggahita_maiEk-lao": "uni0ECD0EC8",
@@ -345,7 +345,7 @@ def test_glyphdata_no_duplicates(self):
     "nnna_uuMatra-tamil": "uni0BA90BC2",
     "nno-khmer.below.narrow1": "uni17D2178E.narrow1",
     "nno-khmer.below.narrow2": "uni17D2178E.narrow2",
-    "noNu_underscore-thai": "uni0E19_uni005F",
+    "noNu_underscore-thai": "uni0E19005F",  # uni0E19_uni005F
     "nya_iMatra-tamil": "uni0B9E0BBF",
     "nya_uMatra-tamil": "uni0B9E0BC1",
     "nya_uuMatra-tamil": "uni0B9E0BC2",
@@ -397,7 +397,7 @@ def test_glyphdata_no_duplicates(self):
     "va_uuMatra-tamil": "uni0BB50BC2",
     "ya_uMatra-tamil": "uni0BAF0BC1",
     "ya_uuMatra-tamil": "uni0BAF0BC2",
-    "yoYing_underscore-thai": "uni0E0D_uni005F",
+    "yoYing_underscore-thai": "uni0E0D005F",  # uni0E0D_uni005F
 }