Turn on UCONFIG_NO_NON_HTML5_CONVERTER to save 100kB

source/data/mappings/convrtrs.txt

 # ******************************************************************************
 # *
 # *   Copyright (C) 1995-2013, International Business Machines
 # *   Corporation and others.  All Rights Reserved.
 # *
 # ******************************************************************************
 
 # If this converter alias table looks very confusing, a much easier to
 # understand view can be found at this demo:
 # http://demo.icu-project.org/icu-bin/convexp
@@ skipping 499 matching lines @@
                         cp923 { JAVA }
                         923 { JAVA }
                         windows-28605 { WINDOWS* }
 
 # CJK encodings
 
 # Chrome: Instead of ibm-943_P15A-2003, we use what's specified in the WHATWG
 # encoding standard (HTML5) for Shift_JIS. Keep all the aliases (even though
 not all of them not required by the encoding spec) for now.
 
-shift_jis-html5 { UTR22* }
+shift_jis-html
                         ibm-943 # Leave untagged because this isn't the default
                         Shift_JIS { IANA* MIME* WINDOWS JAVA }
                         MS_Kanji { IANA WINDOWS JAVA }
                         csShiftJIS { IANA WINDOWS JAVA }
                         windows-31j { IANA JAVA } # A further extension of Shift_JIS to include NEC special characters (Row 13)
                         csWindows31J { IANA WINDOWS JAVA } # A further extension of Shift_JIS to include NEC special characters (Row 13)
                         x-sjis { WINDOWS JAVA }
                         x-ms-cp932 { WINDOWS }
                         cp932 { WINDOWS }
                         windows-932 { WINDOWS* }
                         cp943c { JAVA* }    # This is slightly different, but the backslash mapping is the same.
                         IBM-943C #{ AIX* } # Add this tag once AIX aliases becomes available
                         ms932
                         pck     # Probably SOLARIS
                         sjis    # This might be for ibm-1351
                         ibm-943_VSUB_VPUA
                         x-MS932_0213 { JAVA }
                         x-JISAutoDetect { JAVA }
 
 # Chrome: Instead of ibm-33722_P*, we use what's specified in the WHATWG
 # encoding standard (HTML5). All the
 # 3-byte seqeunces in the normative EUC-JP are now decode-only.
-euc-jp-html5 { UTR22* }
+euc-jp-html
                         EUC-JP { MIME* IANA JAVA* WINDOWS*}
                         Extended_UNIX_Code_Packed_Format_for_Japanese { IANA* JAVA WINDOWS }
                         csEUCPkdFmtJapanese { IANA JAVA WINDOWS }
                         windows-51932 { WINDOWS }
                         X-EUC-JP { MIME JAVA WINDOWS }   # Japan EUC. x-euc-jp is a MIME name
                         eucjis {JAVA}
                         ujis # Linux sometimes uses this name. This is an unfortunate generic and rarely used name. Its use is discouraged.
 
 
 windows-950-2000 { UTR22* }
                         Big5 { IANA* MIME* JAVA* WINDOWS }
                         csBig5 { IANA WINDOWS }
                         windows-950 { WINDOWS* }
                         x-windows-950 { JAVA }
                         x-big5
                         ms950
+# Chrome: HTML5 has big5-hkscs as an alias for big5
+#         TODO(jshin): Decide if Chrome should follow spec. crbug.com/277040
 ibm-1375_P100-2007 { UTR22* }   # Big5-HKSCS-2004 with Unicode 3.1 mappings. This uses supplementary characters.
                         ibm-1375 { IBM* }
                         Big5-HKSCS { IANA* JAVA* }
                         big5hk { JAVA }
                         HKSCS-BIG5  # From http://www.openi18n.org/localenameguide/
 
-# Chrome: HTML5 has big5-hkscs as an alias for big5
-#         TODO(jshin): Decide if Chrome should follow spec. crbug.com/277040
 ibm-5471_P100-2006 { UTR22* }   # Big5-HKSCS-2001 with Unicode 3.0 mappings. This uses many PUA characters.
                         ibm-5471 { IBM* }
                         Big5-HKSCS
                         MS950_HKSCS { JAVA* }
                         hkbig5 # from HP-UX 11i, which can't handle supplementary characters.
                         big5-hkscs:unicode3.0
                         x-MS950-HKSCS { JAVA }
                         # windows-950 # Windows-950 can be w/ or w/o HKSCS extensions. By default it's not.
                         # windows-950_hkscs
 # GBK
@@ skipping 49 matching lines @@
                         TIS-620 { IANA* WINDOWS MIME* }
                         windows-874 { JAVA* WINDOWS* MIME }
                         MS874 { JAVA }
                         x-windows-874 { JAVA }
                         iso-8859-11 { IANA WINDOWS MIME } # iso-8859-11 is similar to TIS-620. ibm-13162 is a closer match.
 
 # Platform codepages
 # Chrome: only keep ibm-878 for KOI8-R, ibm-1168 for KOI8-RU and ibm-866
 ibm-878_P100-1996 { UTR22* }    ibm-878 { IBM* } KOI8-R { IANA* MIME* WINDOWS JAVA* } koi8 { WINDOWS JAVA } csKOI8R { IANA WINDOWS JAVA } windows-20866 { WINDOWS* } cp878   # Russian internet
 # Chrome: Use the table from the WHATWG encoding standard (HTML5).
-ibm-866_html5-2012 { UTR22* }    ibm-866 { IBM* } IBM866 { IANA* MIME* JAVA } cp866 { IANA MIME WINDOWS JAVA* } 866 { IANA JAVA } csIBM866 { IANA JAVA } # PC Russian (w/o euro update)
+ibm866-html   ibm-866 { IBM* } IBM866 { IANA* MIME* JAVA } cp866 { IANA MIME WINDOWS JAVA* } 866 { IANA JAVA } csIBM866 { IANA JAVA } # PC Russian (w/o euro update)
 ibm-1168_P100-2002 { UTR22* }   ibm-1168 { IBM* } KOI8-U { IANA* WINDOWS } windows-21866 { WINDOWS* } # Ukrainian KOI8. koi8-ru != KOI8-U and Microsoft is wrong for aliasing them as the same.
 
 # The cp aliases in this section aren't really windows aliases, but it was used by ICU for Windows.
 # cp is usually used to denote IBM in Java, and that is why we don't do that anymore.
 # The windows-* aliases mean windows codepages.
 ibm-5346_P100-1998 { UTR22* }   ibm-5346 { IBM* } windows-1250 { IANA* JAVA* WINDOWS* } cp1250 { WINDOWS JAVA } # Windows Latin2 (w/ euro update)
 ibm-5347_P100-1998 { UTR22* }   ibm-5347 { IBM* } windows-1251 { IANA* JAVA* WINDOWS* } cp1251 { WINDOWS JAVA } ANSI1251 # Windows Cyrillic (w/ euro update). ANSI1251 is from Solaris
 ibm-5348_P100-1997 { UTR22* }   ibm-5348 { IBM* } windows-1252 { IANA* JAVA* WINDOWS* } cp1252 { JAVA }         # Windows Latin1 (w/ euro update)
 ibm-5349_P100-1998 { UTR22* }   ibm-5349 { IBM* } windows-1253 { IANA* JAVA* WINDOWS* } cp1253 { JAVA }         # Windows Greek (w/ euro update)
 
@@ skipping 17 matching lines @@
 macos-7_3-10.2 { UTR22* }       x-mac-cyrillic { MIME* WINDOWS } windows-10007 { WINDOWS* } mac-cyrillic maccy x-MacCyrillic { JAVA } x-MacUkraine { JAVA* } # Apple Cyrillic
 
 # Partially algorithmic converters
 
 # [U_ENABLE_GENERIC_ISO_2022]
 # The _generic_ ISO-2022 converter is disabled starting 2003-dec-03 (ICU 2.8).
 # For details see the icu mailing list from 2003-dec-01 and the ucnv2022.c file.
 # Language-specific variants of ISO-2022 continue to be available as listed below.
 # ISO_2022                         ISO-2022
 
-# Chrome: The encoding standard only supports ISO-2022-JP and HZ-GB. 
-# Keep ISO-2022-{KR,CN,CN-Ext} until we're sure what to do about
-# replacement encodings. See crbug.com/277037
-# TODO(jshin): Remove them when the bug is resolved.
+# Chrome: The encoding standard only supports ISO-2022-JP.
+# Remove ISO-2022-{KR,CN,CN-Ext} and HZ-GB from the alias table.
+# See crbug.com/277037 and https://www.w3.org/Bugs/Public/show_bug.cgi?id=25339
+# about HZ-GB.
 ISO_2022,locale=ja,version=0    ISO-2022-JP { IANA* MIME* JAVA* } csISO2022JP { IANA JAVA } x-windows-iso2022jp { JAVA } x-windows-50220 { JAVA }
-ISO_2022,locale=ko,version=0    ISO-2022-KR { IANA* MIME* JAVA* } csISO2022KR { IANA JAVA } # This uses ibm-949
-ISO_2022,locale=zh,version=0    ISO-2022-CN { IANA* JAVA* } csISO2022CN { JAVA } x-ISO-2022-CN-GB { JAVA }
-ISO_2022,locale=zh,version=1    ISO-2022-CN-EXT { IANA* }
-HZ                              HZ-GB-2312 { IANA* }
 
 # Chrome: HTML5 does not need ISCII.
 #         Remove all Lotus entries as well.
 
 # EBCDIC codepages according to the CDRA
 # Chrome: Removed all EBCDIC code pages.
 
 # These are not installed by default. They are rarely used.
 # Many of them can be added through the online ICU Data Library Customization tool
 # Chrome: Removed all these entries except for ISO-8859-16 required by HTML5.
 
 iso-8859_16-2001 { UTR22* }     ISO-8859-16 { IANA* } iso-ir-226 { IANA } ISO_8859-16:2001 { IANA } latin10 { IANA } l10 { IANA }