Merge pull request #1150 from anthrotype/ot-tags-from-script

unicodedata: add ot_tags_from_script function

Lib/fontTools/unicodedata/OTTags.py

@@ -0,0 +1,37 @@
+# Data updated to OpenType 1.8.2 as of January 2018.
+
+# Complete list of OpenType script tags at:
+# https://www.microsoft.com/typography/otspec/scripttags.htm
+
+# Most of the script tags are the same as the ISO 15924 tag but lowercased,
+# so we only have to handle the exceptional cases:
+# - KATAKANA and HIRAGANA both map to 'kana';
+# - spaces at the end are preserved, unlike ISO 15924;
+# - we map special script codes for Inherited, Common and Unknown to DFLT.
+
+DEFAULT_SCRIPT = "DFLT"
+
+SCRIPT_EXCEPTIONS = {
+    "Hira": "kana",
+    "Hrkt": "kana",
+    "Laoo": "lao ",
+    "Yiii": "yi  ",
+    "Nkoo": "nko ",
+    "Vaii": "vai ",
+    "Zinh": DEFAULT_SCRIPT,
+    "Zyyy": DEFAULT_SCRIPT,
+    "Zzzz": DEFAULT_SCRIPT,
+}
+
+NEW_SCRIPT_TAGS = {
+    "Beng": ("bng2",),
+    "Deva": ("dev2",),
+    "Gujr": ("gjr2",),
+    "Guru": ("gur2",),
+    "Knda": ("knd2",),
+    "Mlym": ("mlm2",),
+    "Orya": ("ory2",),
+    "Taml": ("tml2",),
+    "Telu": ("tel2",),
+    "Mymr": ("mym2",),
+}

Lib/fontTools/unicodedata/__init__.py

@@ -13,7 +13,7 @@ except ImportError:  # pragma: no cover
     # fall back to built-in unicodedata (possibly outdated)
     from unicodedata import *
 
-from . import Blocks, Scripts, ScriptExtensions
+from . import Blocks, Scripts, ScriptExtensions, OTTags
 
 
 __all__ = [tostr(s) for s in (
@@ -38,6 +38,7 @@ __all__ = [tostr(s) for s in (
     "script_extension",
     "script_name",
     "script_code",
+    "ot_tags_from_script",
 )]
 
 
@@ -147,3 +148,24 @@ def block(char):
     code = byteord(char)
     i = bisect_right(Blocks.RANGES, code)
     return Blocks.VALUES[i-1]
+
+
+def ot_tags_from_script(script_code):
+    """ Return a list of OpenType script tags associated with a given
+    Unicode script code.
+    Return ['DFLT'] script tag for invalid/unknown script codes.
+    """
+    if script_code not in Scripts.NAMES:
+        return [OTTags.DEFAULT_SCRIPT]
+
+    script_tags = [
+        OTTags.SCRIPT_EXCEPTIONS.get(
+            script_code,
+            script_code[0].lower() + script_code[1:]
+        )
+    ]
+    if script_code in OTTags.NEW_SCRIPT_TAGS:
+        script_tags.extend(OTTags.NEW_SCRIPT_TAGS[script_code])
+        script_tags.reverse()  # last in, first out
+
+    return script_tags

Tests/unicodedata_test.py

@@ -203,6 +203,21 @@ def test_block():
     assert unicodedata.block("\u1c90") == "No_Block"
 
 
+def test_ot_tags_from_script():
+    # simple
+    assert unicodedata.ot_tags_from_script("Latn") == ["latn"]
+    # script mapped to multiple new and old script tags
+    assert unicodedata.ot_tags_from_script("Deva") == ["dev2", "deva"]
+    # exceptions
+    assert unicodedata.ot_tags_from_script("Hira") == ["kana"]
+    # special script codes map to DFLT
+    assert unicodedata.ot_tags_from_script("Zinh") == ["DFLT"]
+    assert unicodedata.ot_tags_from_script("Zyyy") == ["DFLT"]
+    assert unicodedata.ot_tags_from_script("Zzzz") == ["DFLT"]
+    # this is invalid or unknown
+    assert unicodedata.ot_tags_from_script("Aaaa") == ["DFLT"]
+
+
 if __name__ == "__main__":
     import sys
     sys.exit(pytest.main(sys.argv))