Merge branch 'master' into conflictingFontFix

src/sample.py

@@ -20,6 +20,7 @@ def tounicode(s, encoding='utf-8'):
 
 fontdata = open (sys.argv[1], 'rb').read ()
 text = tounicode(sys.argv[2])
+codepoints = list(map(ord, text))
 # Need to create GLib.Bytes explicitly until this bug is fixed:
 # https://bugzilla.gnome.org/show_bug.cgi?id=729541
 blob = hb.glib_blob_create (GLib.Bytes.new (fontdata))
@@ -39,7 +40,7 @@ class Debugger(object):
 		return True
 debugger = Debugger()
 hb.buffer_set_message_func (buf, debugger.message, 1, 0)
-hb.buffer_add_utf8 (buf, text.encode('utf-8'), 0, -1)
+hb.buffer_add_utf32 (buf, codepoints, 0, len(codepoints))
 hb.buffer_guess_segment_properties (buf)
 
 hb.shape (font, buf, [])

test/shaping/hb_test_tools.py

@@ -7,7 +7,43 @@ from itertools import *
 diff_symbols = "-+=*&^%$#@!~/"
 diff_colors = ['red', 'green', 'blue']
 
-if sys.version_info[0] >= 3:
+try:
+	unichr = unichr
+
+	if sys.maxunicode < 0x10FFFF:
+		# workarounds for Python 2 "narrow" builds with UCS2-only support.
+
+		_narrow_unichr = unichr
+
+		def unichr(i):
+			"""
+			Return the unicode character whose Unicode code is the integer 'i'.
+			The valid range is 0 to 0x10FFFF inclusive.
+
+			>>> _narrow_unichr(0xFFFF + 1)
+			Traceback (most recent call last):
+			  File "<stdin>", line 1, in ?
+			ValueError: unichr() arg not in range(0x10000) (narrow Python build)
+			>>> unichr(0xFFFF + 1) == u'\U00010000'
+			True
+			>>> unichr(1114111) == u'\U0010FFFF'
+			True
+			>>> unichr(0x10FFFF + 1)
+			Traceback (most recent call last):
+			  File "<stdin>", line 1, in ?
+			ValueError: unichr() arg not in range(0x110000)
+			"""
+			try:
+				return _narrow_unichr(i)
+			except ValueError:
+				try:
+					padded_hex_str = hex(i)[2:].zfill(8)
+					escape_str = "\\U" + padded_hex_str
+					return escape_str.decode("unicode-escape")
+				except UnicodeDecodeError:
+					raise ValueError('unichr() arg not in range(0x110000)')
+
+except NameError:
 	unichr = chr
 
 class ColorFormatter: