def _CheckRanges(self):
"""Verify that in each range tuple the source and target ranges
have the same length."""
if self._uni_to_number_ranges:
for range in self._uni_to_number_ranges:
assert (range[1] - range[0]) == (range[3] - range[2])
if self._uni_to_old_number_ranges:
for range in self._uni_to_old_number_ranges:
assert (range[1] - range[0]) == (range[3] - range[2])
if self._uni_to_shift_jis_ranges:
for range in self._uni_to_shift_jis_ranges:
# Shift the Shift-JIS codes down to JIS X 0208 and compute the
# linear differences.
shift_jis_start = row_cell.FromShiftJis((range[2] >> 8) - 0x10,
range[2] & 0xff)
shift_jis_end = row_cell.FromShiftJis((range[3] >> 8) - 0x10,
range[3] & 0xff)
assert (range[1] - range[0]) == (shift_jis_end - shift_jis_start)
if self._uni_to_jis_ranges:
for range in self._uni_to_jis_ranges:
jis_start = row_cell.From2022((range[2] >> 8), range[2] & 0xff)
jis_end = row_cell.From2022((range[3] >> 8), range[3] & 0xff)
assert (range[1] - range[0]) == (jis_end - jis_start)
def _JisFromUnicode(ranges, uni):
"""Map a Unicode code point to a JIS X 0208 (ISO-2022-JP) code.
In a range of JIS codes, only valid codes according to the encoding
scheme are counted. For example, after 757E follows 7621.
Args:
ranges: A list of ranges. See _RangeFromUnicode().
uni: A Unicode code point (a hex digit string).
Returns:
The JIS code (integer) corresponding to the
Unicode code point, according to the ranges;
or None if none of the ranges contains the code point.
"""
uni = int(uni, 16)
range = _RangeFromUnicode(ranges, uni)
offset = uni - range[0]
rc = row_cell.From2022(range[2] >> 8, range[2] & 0xff) + offset
(b1, b2) = rc.To2022()
return (b1 << 8) | b2
def testFrom2022(self):
rc = row_cell.From2022(0x21, 0x7e)
self.assertEqual(rc, row_cell.RowCell(1, 94))
self.assertRaises(ValueError, row_cell.From2022, 0, 94)
self.assertRaises(ValueError, row_cell.From2022, 1, 95)