Sifter updates for 17.0

From Ken on feb18:

The changes to the source code are minor -- just the usual updates for
new repertoire and date, plus some tweaks for the split for Tangut
implicit weighting.

I had to explicitly touch cttfooter.txt for the Tangut changes and for
the addition of Extension J, as well as the range for Extension C.

From Ken on feb19:

These were to fix the output for the CTT to include a time stamp on the
generation and to include the CTT table name *inside* the generated
file. The output now is just named "ctt.txt".

From Ken on apr24:

I've made the small updates to the sifter code and to the
CTT template text for the changes in CJK ranges.

Author: markusicu

c/uca/sifter/cttfooter.txt

@@ -13,40 +13,50 @@
 % Weights for unified Han characters follow the Unified Repertoire and
 %   Ordering, which is a language-neutral, traditional radical-stroke order.
 
-%   The original URO and Extensions A through I, plus the 12 unified Han characters
+%   The original URO and Extensions A through J, plus the 12 unified Han characters
 %   in the CJK compatibility area are weighted implicitly as defined here.
 
 %   WEIGHT_BASE = 0xFB40 for original URO and 12 unified Han from CJK compat area.
 %      cp >= 0x04E00 && cp <= 0x09FFF % URO
 %   WEIGHT_BASE = 0xFB80 for Extension A through Extension I Han characters.
 %      cp >= 0x03400 && cp <= 0x04DBF % Ext. A
 %      cp >= 0x20000 && cp <= 0x2A6DF % Ext. B
-%      cp >= 0x2A700 && cp <= 0x2B739 % Ext. C
+%      cp >= 0x2A700 && cp <= 0x2B73F % Ext. C
 %      cp >= 0x2B740 && cp <= 0x2B81D % Ext. D
-%      cp >= 0x2B820 && cp <= 0x2CEA1 % Ext. E
+%      cp >= 0x2B820 && cp <= 0x2CEAD % Ext. E
 %      cp >= 0x2CEB0 && cp <= 0x2EBE0 % Ext. F
 %      cp >= 0x2EBF0 && cp <= 0x2EE5D % Ext. I
 %      cp >= 0x30000 && cp <= 0x3134A % Ext. G
 %      cp >= 0x31350 && cp <= 0x323AF % Ext. H
+%      cp >= 0x323B0 && cp <= 0x33479 % Ext. J
 %   For a given Han character at code point cp:
 %   base1 = WEIGHT_BASE + ( cp >> 15 )
 %   base2 = ( cp & 0x7FFF ) | 0x8000
 %   Then weight the character as: <U{cp}> "<R{base1}><T{base2}>";<BASE>;<MIN>;<SFFFF>
 
-% Tangut ideographic and component characters are weighted implicitly as defined here.
+% Tangut ideographic characters are weighted implicitly as defined here.
 
 %   WEIGHT_BASE = 0xFB00
-%      cp >= 0x17000 && cp <= 0x187F7 % Tangut ideographs
+%      cp >= 0x17000 && cp <= 0x187FF % Tangut ideographs
+%      cp >= 0x18D00 && cp <= 0x18D1E % Tangut ideograph supplement
+%   For a given Tangut character at code point cp:
+%   base1 = WEIGHT_BASE
+%   base2 = ( cp - 0x17000 ) | 0x8000
+%   Then weight the character as: <U{cp}> "<R{base1}><T{base2}>";<BASE>;<MIN>;<SFFFF>
+
+% Tangut component characters are weighted implicitly as defined here.
+
+%   WEIGHT_BASE = 0xFB01
 %      cp >= 0x18800 && cp <= 0x18AFF % Tangut components
-%      cp >= 0x18D00 && cp <= 0x18D08 % Tangut ideograph supplement
+%      cp >= 0x18D80 && cp <= 0x18DFF % Tangut component supplement
 %   For a given Tangut character at code point cp:
 %   base1 = WEIGHT_BASE
 %   base2 = ( cp - 0x17000 ) | 0x8000
 %   Then weight the character as: <U{cp}> "<R{base1}><T{base2}>";<BASE>;<MIN>;<SFFFF>
 
 % Nushu ideographic characters are weighted implicitly as defined here.
 
-%   WEIGHT_BASE = 0xFB01
+%   WEIGHT_BASE = 0xFB02
 %      cp >= 0x1B170 && cp <= 0x1B2FB % Nushu
 %   For a given Nushu character at code point cp:
 %   base1 = WEIGHT_BASE
@@ -55,7 +65,7 @@
 
 % Khitan Small Script ideographic characters are weighted implicitly as defined here.
 
-%   WEIGHT_BASE = 0xFB02
+%   WEIGHT_BASE = 0xFB03
 %      cp >= 0x18B00 && cp <= 0x18CD5 % Khitan Small Script
 %   For a given Khitan Small Script character at code point cp:
 %   base1 = WEIGHT_BASE