[U-Boot] [PATCH v3 00/05]efi_loader: EFI_UNICODE_COLLATION_PROTOCOL

[U-Boot] [PATCH v3 00/05]efi_loader: EFI_UNICODE_COLLATION_PROTOCOL

[Heinrich Schuchardt]

This patchset implements the EFI_UNICODE_COLLATION_PROTOCOL and provides
unit tests.

An error in Unicode handling in the FAT file system shall be fixed after
merging Takahiro's FAT patches. Until then some redundancy between old and
new Unicode functions remains.

v3:
	drop accepted patches
	remove codes < 0x80 in capitalization table
	add a test for capitalization of latin O with diaresis
v2:
	use CONFIG_IS_ENABLED(EFI_LOADER) instead of
	defined(CONFIG_EFI_LOADER) && !defined(CONFIG_SPL_BUILD)
	add configuration option EFI_UNICODE_CAPITALIZATION
	use ut_assert*() for testing
	test illegal Unicode sequences
	add more comments

Heinrich Schuchardt (5):
  efi_loader: capitalization table
  lib: charset: upper/lower case conversion
  test: tests for utf_to_lower() utf_to_upper().
  efi_loader: EFI_UNICODE_COLLATION_PROTOCOL
  efi_selftest: EFI_UNICODE_COLLATION_PROTOCOL

 MAINTAINERS                                   |    3 +
 include/capitalization.h                      | 1978 +++++++++++++++++
 include/charset.h                             |   16 +
 include/cp1250.h                              |   40 +
 include/cp437.h                               |   40 +
 include/efi_api.h                             |   21 +
 include/efi_loader.h                          |    5 +
 lib/charset.c                                 |   47 +
 lib/efi_loader/Kconfig                        |   10 +
 lib/efi_loader/Makefile                       |   18 +-
 lib/efi_loader/efi_boottime.c                 |    6 +
 lib/efi_loader/efi_unicode_collation.c        |  329 +++
 lib/efi_selftest/Makefile                     |    1 +
 .../efi_selftest_unicode_collation.c          |  260 +++
 test/unicode_ut.c                             |   34 +
 15 files changed, 2804 insertions(+), 4 deletions(-)
 create mode 100644 include/capitalization.h
 create mode 100644 include/cp1250.h
 create mode 100644 include/cp437.h
 create mode 100644 lib/efi_loader/efi_unicode_collation.c
 create mode 100644 lib/efi_selftest/efi_selftest_unicode_collation.c

-- 
2.18.0

[U-Boot] [PATCH v3 1/5] efi_loader: capitalization table

[Heinrich Schuchardt]

This patch provides a define to initialize a table that maps lower to
capital letters for Unicode code point 0x0000 - 0xffff.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
---
v3:
	remove codes < 0x80 from tables for code pages 437 and 1250
	sort tables for code pages 437 and 1250
v2:
	add shorter tables for code pages 437 and 1250
---
 MAINTAINERS              |    1 +
 include/capitalization.h | 1978 ++++++++++++++++++++++++++++++++++++++
 2 files changed, 1979 insertions(+)
 create mode 100644 include/capitalization.h

diff --git a/MAINTAINERS b/MAINTAINERS
index 46f826a0fe..8c9cd83347 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -381,6 +381,7 @@ T:	git git://github.com/agraf/u-boot.git
 F:	doc/README.uefi
 F:	doc/README.iscsi
 F:	Documentation/efi.rst
+F:	include/capitalization.h
 F:	include/efi*
 F:	include/pe.h
 F:	include/asm-generic/pe.h
diff --git a/include/capitalization.h b/include/capitalization.h
new file mode 100644
index 0000000000..ade4e8c634
--- /dev/null
+++ b/include/capitalization.h
@@ -0,0 +1,1978 @@
+/* SPDX-License-Identifier: Unicode-DFS-2016 */
+/*
+ * Capitalization tables
+ */
+
+struct capitalization_table {
+	u16 upper;
+	u16 lower;
+};
+
+/*
+ * Correspondence table for small and capital Unicode letters in the range of
+ * 0x0000 - 0xffff based on http://www.unicode.org/Public/UCA/11.0.0/allkeys.txt
+ */
+#define UNICODE_CAPITALIZATION_TABLE { \
+	{ 0x0531, /* ARMENIAN CAPITAL LETTER AYB */ \
+	  0x0561, /* ARMENIAN SMALL LETTER AYB */ }, \
+	{ 0x0532, /* ARMENIAN CAPITAL LETTER BEN */ \
+	  0x0562, /* ARMENIAN SMALL LETTER BEN */ }, \
+	{ 0x053E, /* ARMENIAN CAPITAL LETTER CA */ \
+	  0x056E, /* ARMENIAN SMALL LETTER CA */ }, \
+	{ 0x0549, /* ARMENIAN CAPITAL LETTER CHA */ \
+	  0x0579, /* ARMENIAN SMALL LETTER CHA */ }, \
+	{ 0x0543, /* ARMENIAN CAPITAL LETTER CHEH */ \
+	  0x0573, /* ARMENIAN SMALL LETTER CHEH */ }, \
+	{ 0x0551, /* ARMENIAN CAPITAL LETTER CO */ \
+	  0x0581, /* ARMENIAN SMALL LETTER CO */ }, \
+	{ 0x0534, /* ARMENIAN CAPITAL LETTER DA */ \
+	  0x0564, /* ARMENIAN SMALL LETTER DA */ }, \
+	{ 0x0535, /* ARMENIAN CAPITAL LETTER ECH */ \
+	  0x0565, /* ARMENIAN SMALL LETTER ECH */ }, \
+	{ 0x0537, /* ARMENIAN CAPITAL LETTER EH */ \
+	  0x0567, /* ARMENIAN SMALL LETTER EH */ }, \
+	{ 0x0538, /* ARMENIAN CAPITAL LETTER ET */ \
+	  0x0568, /* ARMENIAN SMALL LETTER ET */ }, \
+	{ 0x0556, /* ARMENIAN CAPITAL LETTER FEH */ \
+	  0x0586, /* ARMENIAN SMALL LETTER FEH */ }, \
+	{ 0x0542, /* ARMENIAN CAPITAL LETTER GHAD */ \
+	  0x0572, /* ARMENIAN SMALL LETTER GHAD */ }, \
+	{ 0x0533, /* ARMENIAN CAPITAL LETTER GIM */ \
+	  0x0563, /* ARMENIAN SMALL LETTER GIM */ }, \
+	{ 0x0540, /* ARMENIAN CAPITAL LETTER HO */ \
+	  0x0570, /* ARMENIAN SMALL LETTER HO */ }, \
+	{ 0x053B, /* ARMENIAN CAPITAL LETTER INI */ \
+	  0x056B, /* ARMENIAN SMALL LETTER INI */ }, \
+	{ 0x0541, /* ARMENIAN CAPITAL LETTER JA */ \
+	  0x0571, /* ARMENIAN SMALL LETTER JA */ }, \
+	{ 0x054B, /* ARMENIAN CAPITAL LETTER JHEH */ \
+	  0x057B, /* ARMENIAN SMALL LETTER JHEH */ }, \
+	{ 0x0554, /* ARMENIAN CAPITAL LETTER KEH */ \
+	  0x0584, /* ARMENIAN SMALL LETTER KEH */ }, \
+	{ 0x053F, /* ARMENIAN CAPITAL LETTER KEN */ \
+	  0x056F, /* ARMENIAN SMALL LETTER KEN */ }, \
+	{ 0x053C, /* ARMENIAN CAPITAL LETTER LIWN */ \
+	  0x056C, /* ARMENIAN SMALL LETTER LIWN */ }, \
+	{ 0x0544, /* ARMENIAN CAPITAL LETTER MEN */ \
+	  0x0574, /* ARMENIAN SMALL LETTER MEN */ }, \
+	{ 0x0546, /* ARMENIAN CAPITAL LETTER NOW */ \
+	  0x0576, /* ARMENIAN SMALL LETTER NOW */ }, \
+	{ 0x0555, /* ARMENIAN CAPITAL LETTER OH */ \
+	  0x0585, /* ARMENIAN SMALL LETTER OH */ }, \
+	{ 0x054A, /* ARMENIAN CAPITAL LETTER PEH */ \
+	  0x057A, /* ARMENIAN SMALL LETTER PEH */ }, \
+	{ 0x0553, /* ARMENIAN CAPITAL LETTER PIWR */ \
+	  0x0583, /* ARMENIAN SMALL LETTER PIWR */ }, \
+	{ 0x054C, /* ARMENIAN CAPITAL LETTER RA */ \
+	  0x057C, /* ARMENIAN SMALL LETTER RA */ }, \
+	{ 0x0550, /* ARMENIAN CAPITAL LETTER REH */ \
+	  0x0580, /* ARMENIAN SMALL LETTER REH */ }, \
+	{ 0x054D, /* ARMENIAN CAPITAL LETTER SEH */ \
+	  0x057D, /* ARMENIAN SMALL LETTER SEH */ }, \
+	{ 0x0547, /* ARMENIAN CAPITAL LETTER SHA */ \
+	  0x0577, /* ARMENIAN SMALL LETTER SHA */ }, \
+	{ 0x054F, /* ARMENIAN CAPITAL LETTER TIWN */ \
+	  0x057F, /* ARMENIAN SMALL LETTER TIWN */ }, \
+	{ 0x0539, /* ARMENIAN CAPITAL LETTER TO */ \
+	  0x0569, /* ARMENIAN SMALL LETTER TO */ }, \
+	{ 0x054E, /* ARMENIAN CAPITAL LETTER VEW */ \
+	  0x057E, /* ARMENIAN SMALL LETTER VEW */ }, \
+	{ 0x0548, /* ARMENIAN CAPITAL LETTER VO */ \
+	  0x0578, /* ARMENIAN SMALL LETTER VO */ }, \
+	{ 0x053D, /* ARMENIAN CAPITAL LETTER XEH */ \
+	  0x056D, /* ARMENIAN SMALL LETTER XEH */ }, \
+	{ 0x0545, /* ARMENIAN CAPITAL LETTER YI */ \
+	  0x0575, /* ARMENIAN SMALL LETTER YI */ }, \
+	{ 0x0552, /* ARMENIAN CAPITAL LETTER YIWN */ \
+	  0x0582, /* ARMENIAN SMALL LETTER YIWN */ }, \
+	{ 0x0536, /* ARMENIAN CAPITAL LETTER ZA */ \
+	  0x0566, /* ARMENIAN SMALL LETTER ZA */ }, \
+	{ 0x053A, /* ARMENIAN CAPITAL LETTER ZHE */ \
+	  0x056A, /* ARMENIAN SMALL LETTER ZHE */ }, \
+	{ 0x24B6, /* CIRCLED LATIN CAPITAL LETTER A */ \
+	  0x24D0, /* CIRCLED LATIN SMALL LETTER A */ }, \
+	{ 0x24B7, /* CIRCLED LATIN CAPITAL LETTER B */ \
+	  0x24D1, /* CIRCLED LATIN SMALL LETTER B */ }, \
+	{ 0x24B8, /* CIRCLED LATIN CAPITAL LETTER C */ \
+	  0x24D2, /* CIRCLED LATIN SMALL LETTER C */ }, \
+	{ 0x24B9, /* CIRCLED LATIN CAPITAL LETTER D */ \
+	  0x24D3, /* CIRCLED LATIN SMALL LETTER D */ }, \
+	{ 0x24BA, /* CIRCLED LATIN CAPITAL LETTER E */ \
+	  0x24D4, /* CIRCLED LATIN SMALL LETTER E */ }, \
+	{ 0x24BB, /* CIRCLED LATIN CAPITAL LETTER F */ \
+	  0x24D5, /* CIRCLED LATIN SMALL LETTER F */ }, \
+	{ 0x24BC, /* CIRCLED LATIN CAPITAL LETTER G */ \
+	  0x24D6, /* CIRCLED LATIN SMALL LETTER G */ }, \
+	{ 0x24BD, /* CIRCLED LATIN CAPITAL LETTER H */ \
+	  0x24D7, /* CIRCLED LATIN SMALL LETTER H */ }, \
+	{ 0x24BE, /* CIRCLED LATIN CAPITAL LETTER I */ \
+	  0x24D8, /* CIRCLED LATIN SMALL LETTER I */ }, \
+	{ 0x24BF, /* CIRCLED LATIN CAPITAL LETTER J */ \
+	  0x24D9, /* CIRCLED LATIN SMALL LETTER J */ }, \
+	{ 0x24C0, /* CIRCLED LATIN CAPITAL LETTER K */ \
+	  0x24DA, /* CIRCLED LATIN SMALL LETTER K */ }, \
+	{ 0x24C1, /* CIRCLED LATIN CAPITAL LETTER L */ \
+	  0x24DB, /* CIRCLED LATIN SMALL LETTER L */ }, \
+	{ 0x24C2, /* CIRCLED LATIN CAPITAL LETTER M */ \
+	  0x24DC, /* CIRCLED LATIN SMALL LETTER M */ }, \
+	{ 0x24C3, /* CIRCLED LATIN CAPITAL LETTER N */ \
+	  0x24DD, /* CIRCLED LATIN SMALL LETTER N */ }, \
+	{ 0x24C4, /* CIRCLED LATIN CAPITAL LETTER O */ \
+	  0x24DE, /* CIRCLED LATIN SMALL LETTER O */ }, \
+	{ 0x24C5, /* CIRCLED LATIN CAPITAL LETTER P */ \
+	  0x24DF, /* CIRCLED LATIN SMALL LETTER P */ }, \
+	{ 0x24C6, /* CIRCLED LATIN CAPITAL LETTER Q */ \
+	  0x24E0, /* CIRCLED LATIN SMALL LETTER Q */ }, \
+	{ 0x24C7, /* CIRCLED LATIN CAPITAL LETTER R */ \
+	  0x24E1, /* CIRCLED LATIN SMALL LETTER R */ }, \
+	{ 0x24C8, /* CIRCLED LATIN CAPITAL LETTER S */ \
+	  0x24E2, /* CIRCLED LATIN SMALL LETTER S */ }, \
+	{ 0x24C9, /* CIRCLED LATIN CAPITAL LETTER T */ \
+	  0x24E3, /* CIRCLED LATIN SMALL LETTER T */ }, \
+	{ 0x24CA, /* CIRCLED LATIN CAPITAL LETTER U */ \
+	  0x24E4, /* CIRCLED LATIN SMALL LETTER U */ }, \
+	{ 0x24CB, /* CIRCLED LATIN CAPITAL LETTER V */ \
+	  0x24E5, /* CIRCLED LATIN SMALL LETTER V */ }, \
+	{ 0x24CC, /* CIRCLED LATIN CAPITAL LETTER W */ \
+	  0x24E6, /* CIRCLED LATIN SMALL LETTER W */ }, \
+	{ 0x24CD, /* CIRCLED LATIN CAPITAL LETTER X */ \
+	  0x24E7, /* CIRCLED LATIN SMALL LETTER X */ }, \
+	{ 0x24CE, /* CIRCLED LATIN CAPITAL LETTER Y */ \
+	  0x24E8, /* CIRCLED LATIN SMALL LETTER Y */ }, \
+	{ 0x24CF, /* CIRCLED LATIN CAPITAL LETTER Z */ \
+	  0x24E9, /* CIRCLED LATIN SMALL LETTER Z */ }, \
+	{ 0x2CC8, /* COPTIC CAPITAL LETTER AKHMIMIC KHEI */ \
+	  0x2CC9, /* COPTIC SMALL LETTER AKHMIMIC KHEI */ }, \
+	{ 0x2C80, /* COPTIC CAPITAL LETTER ALFA */ \
+	  0x2C81, /* COPTIC SMALL LETTER ALFA */ }, \
+	{ 0x2CF2, /* COPTIC CAPITAL LETTER BOHAIRIC KHEI */ \
+	  0x2CF3, /* COPTIC SMALL LETTER BOHAIRIC KHEI */ }, \
+	{ 0x2CC2, /* COPTIC CAPITAL LETTER CROSSED SHEI */ \
+	  0x2CC3, /* COPTIC SMALL LETTER CROSSED SHEI */ }, \
+	{ 0x2CB6, /* COPTIC CAPITAL LETTER CRYPTOGRAMMIC EIE */ \
+	  0x2CB7, /* COPTIC SMALL LETTER CRYPTOGRAMMIC EIE */ }, \
+	{ 0x2CED, /* COPTIC CAPITAL LETTER CRYPTOGRAMMIC GANGIA */ \
+	  0x2CEE, /* COPTIC SMALL LETTER CRYPTOGRAMMIC GANGIA */ }, \
+	{ 0x2CBC, /* COPTIC CAPITAL LETTER CRYPTOGRAMMIC NI */ \
+	  0x2CBD, /* COPTIC SMALL LETTER CRYPTOGRAMMIC NI */ }, \
+	{ 0x2CEB, /* COPTIC CAPITAL LETTER CRYPTOGRAMMIC SHEI */ \
+	  0x2CEC, /* COPTIC SMALL LETTER CRYPTOGRAMMIC SHEI */ }, \
+	{ 0x2C86, /* COPTIC CAPITAL LETTER DALDA */ \
+	  0x2C87, /* COPTIC SMALL LETTER DALDA */ }, \
+	{ 0x03EE, /* COPTIC CAPITAL LETTER DEI */ \
+	  0x03EF, /* COPTIC SMALL LETTER DEI */ }, \
+	{ 0x2CB2, /* COPTIC CAPITAL LETTER DIALECT-P ALEF */ \
+	  0x2CB3, /* COPTIC SMALL LETTER DIALECT-P ALEF */ }, \
+	{ 0x2CCA, /* COPTIC CAPITAL LETTER DIALECT-P HORI */ \
+	  0x2CCB, /* COPTIC SMALL LETTER DIALECT-P HORI */ }, \
+	{ 0x2CB8, /* COPTIC CAPITAL LETTER DIALECT-P KAPA */ \
+	  0x2CB9, /* COPTIC SMALL LETTER DIALECT-P KAPA */ }, \
+	{ 0x2CBA, /* COPTIC CAPITAL LETTER DIALECT-P NI */ \
+	  0x2CBB, /* COPTIC SMALL LETTER DIALECT-P NI */ }, \
+	{ 0x2C88, /* COPTIC CAPITAL LETTER EIE */ \
+	  0x2C89, /* COPTIC SMALL LETTER EIE */ }, \
+	{ 0x03E4, /* COPTIC CAPITAL LETTER FEI */ \
+	  0x03E5, /* COPTIC SMALL LETTER FEI */ }, \
+	{ 0x2CAA, /* COPTIC CAPITAL LETTER FI */ \
+	  0x2CAB, /* COPTIC SMALL LETTER FI */ }, \
+	{ 0x2C84, /* COPTIC CAPITAL LETTER GAMMA */ \
+	  0x2C85, /* COPTIC SMALL LETTER GAMMA */ }, \
+	{ 0x03EA, /* COPTIC CAPITAL LETTER GANGIA */ \
+	  0x03EB, /* COPTIC SMALL LETTER GANGIA */ }, \
+	{ 0x2C8E, /* COPTIC CAPITAL LETTER HATE */ \
+	  0x2C8F, /* COPTIC SMALL LETTER HATE */ }, \
+	{ 0x03E8, /* COPTIC CAPITAL LETTER HORI */ \
+	  0x03E9, /* COPTIC SMALL LETTER HORI */ }, \
+	{ 0x2C92, /* COPTIC CAPITAL LETTER IAUDA */ \
+	  0x2C93, /* COPTIC SMALL LETTER IAUDA */ }, \
+	{ 0x2C94, /* COPTIC CAPITAL LETTER KAPA */ \
+	  0x2C95, /* COPTIC SMALL LETTER KAPA */ }, \
+	{ 0x03E6, /* COPTIC CAPITAL LETTER KHEI */ \
+	  0x03E7, /* COPTIC SMALL LETTER KHEI */ }, \
+	{ 0x2CAC, /* COPTIC CAPITAL LETTER KHI */ \
+	  0x2CAD, /* COPTIC SMALL LETTER KHI */ }, \
+	{ 0x2C9C, /* COPTIC CAPITAL LETTER KSI */ \
+	  0x2C9D, /* COPTIC SMALL LETTER KSI */ }, \
+	{ 0x2C96, /* COPTIC CAPITAL LETTER LAULA */ \
+	  0x2C97, /* COPTIC SMALL LETTER LAULA */ }, \
+	{ 0x2CD0, /* COPTIC CAPITAL LETTER L-SHAPED HA */ \
+	  0x2CD1, /* COPTIC SMALL LETTER L-SHAPED HA */ }, \
+	{ 0x2C98, /* COPTIC CAPITAL LETTER MI */ \
+	  0x2C99, /* COPTIC SMALL LETTER MI */ }, \
+	{ 0x2C9A, /* COPTIC CAPITAL LETTER NI */ \
+	  0x2C9B, /* COPTIC SMALL LETTER NI */ }, \
+	{ 0x2C9E, /* COPTIC CAPITAL LETTER O */ \
+	  0x2C9F, /* COPTIC SMALL LETTER O */ }, \
+	{ 0x2CB4, /* COPTIC CAPITAL LETTER OLD COPTIC AIN */ \
+	  0x2CB5, /* COPTIC SMALL LETTER OLD COPTIC AIN */ }, \
+	{ 0x2CD8, /* COPTIC CAPITAL LETTER OLD COPTIC DJA */ \
+	  0x2CD9, /* COPTIC SMALL LETTER OLD COPTIC DJA */ }, \
+	{ 0x2CC6, /* COPTIC CAPITAL LETTER OLD COPTIC ESH */ \
+	  0x2CC7, /* COPTIC SMALL LETTER OLD COPTIC ESH */ }, \
+	{ 0x2CD6, /* COPTIC CAPITAL LETTER OLD COPTIC GANGIA */ \
+	  0x2CD7, /* COPTIC SMALL LETTER OLD COPTIC GANGIA */ }, \
+	{ 0x2CCE, /* COPTIC CAPITAL LETTER OLD COPTIC HA */ \
+	  0x2CCF, /* COPTIC SMALL LETTER OLD COPTIC HA */ }, \
+	{ 0x2CD4, /* COPTIC CAPITAL LETTER OLD COPTIC HAT */ \
+	  0x2CD5, /* COPTIC SMALL LETTER OLD COPTIC HAT */ }, \
+	{ 0x2CD2, /* COPTIC CAPITAL LETTER OLD COPTIC HEI */ \
+	  0x2CD3, /* COPTIC SMALL LETTER OLD COPTIC HEI */ }, \
+	{ 0x2CCC, /* COPTIC CAPITAL LETTER OLD COPTIC HORI */ \
+	  0x2CCD, /* COPTIC SMALL LETTER OLD COPTIC HORI */ }, \
+	{ 0x2CBE, /* COPTIC CAPITAL LETTER OLD COPTIC OOU */ \
+	  0x2CBF, /* COPTIC SMALL LETTER OLD COPTIC OOU */ }, \
+	{ 0x2CC4, /* COPTIC CAPITAL LETTER OLD COPTIC SHEI */ \
+	  0x2CC5, /* COPTIC SMALL LETTER OLD COPTIC SHEI */ }, \
+	{ 0x2CDA, /* COPTIC CAPITAL LETTER OLD COPTIC SHIMA */ \
+	  0x2CDB, /* COPTIC SMALL LETTER OLD COPTIC SHIMA */ }, \
+	{ 0x2CDE, /* COPTIC CAPITAL LETTER OLD NUBIAN NGI */ \
+	  0x2CDF, /* COPTIC SMALL LETTER OLD NUBIAN NGI */ }, \
+	{ 0x2CE0, /* COPTIC CAPITAL LETTER OLD NUBIAN NYI */ \
+	  0x2CE1, /* COPTIC SMALL LETTER OLD NUBIAN NYI */ }, \
+	{ 0x2CDC, /* COPTIC CAPITAL LETTER OLD NUBIAN SHIMA */ \
+	  0x2CDD, /* COPTIC SMALL LETTER OLD NUBIAN SHIMA */ }, \
+	{ 0x2CE2, /* COPTIC CAPITAL LETTER OLD NUBIAN WAU */ \
+	  0x2CE3, /* COPTIC SMALL LETTER OLD NUBIAN WAU */ }, \
+	{ 0x2CB0, /* COPTIC CAPITAL LETTER OOU */ \
+	  0x2CB1, /* COPTIC SMALL LETTER OOU */ }, \
+	{ 0x2CA0, /* COPTIC CAPITAL LETTER PI */ \
+	  0x2CA1, /* COPTIC SMALL LETTER PI */ }, \
+	{ 0x2CAE, /* COPTIC CAPITAL LETTER PSI */ \
+	  0x2CAF, /* COPTIC SMALL LETTER PSI */ }, \
+	{ 0x2CA2, /* COPTIC CAPITAL LETTER RO */ \
+	  0x2CA3, /* COPTIC SMALL LETTER RO */ }, \
+	{ 0x2CC0, /* COPTIC CAPITAL LETTER SAMPI */ \
+	  0x2CC1, /* COPTIC SMALL LETTER SAMPI */ }, \
+	{ 0x03E2, /* COPTIC CAPITAL LETTER SHEI */ \
+	  0x03E3, /* COPTIC SMALL LETTER SHEI */ }, \
+	{ 0x03EC, /* COPTIC CAPITAL LETTER SHIMA */ \
+	  0x03ED, /* COPTIC SMALL LETTER SHIMA */ }, \
+	{ 0x2CA4, /* COPTIC CAPITAL LETTER SIMA */ \
+	  0x2CA5, /* COPTIC SMALL LETTER SIMA */ }, \
+	{ 0x2C8A, /* COPTIC CAPITAL LETTER SOU */ \
+	  0x2C8B, /* COPTIC SMALL LETTER SOU */ }, \
+	{ 0x2CA6, /* COPTIC CAPITAL LETTER TAU */ \
+	  0x2CA7, /* COPTIC SMALL LETTER TAU */ }, \
+	{ 0x2C90, /* COPTIC CAPITAL LETTER THETHE */ \
+	  0x2C91, /* COPTIC SMALL LETTER THETHE */ }, \
+	{ 0x2CA8, /* COPTIC CAPITAL LETTER UA */ \
+	  0x2CA9, /* COPTIC SMALL LETTER UA */ }, \
+	{ 0x2C82, /* COPTIC CAPITAL LETTER VIDA */ \
+	  0x2C83, /* COPTIC SMALL LETTER VIDA */ }, \
+	{ 0x2C8C, /* COPTIC CAPITAL LETTER ZATA */ \
+	  0x2C8D, /* COPTIC SMALL LETTER ZATA */ }, \
+	{ 0x0410, /* CYRILLIC CAPITAL LETTER A */ \
+	  0x0430, /* CYRILLIC SMALL LETTER A */ }, \
+	{ 0x04D0, /* CYRILLIC CAPITAL LETTER A WITH BREVE */ \
+	  0x04D1, /* CYRILLIC SMALL LETTER A WITH BREVE */ }, \
+	{ 0x04D2, /* CYRILLIC CAPITAL LETTER A WITH DIAERESIS */ \
+	  0x04D3, /* CYRILLIC SMALL LETTER A WITH DIAERESIS */ }, \
+	{ 0x04BC, /* CYRILLIC CAPITAL LETTER ABKHASIAN CHE */ \
+	  0x04BD, /* CYRILLIC SMALL LETTER ABKHASIAN CHE */ }, \
+	{ 0x04BE, /* CYRILLIC CAPITAL LETTER ABKHASIAN CHE WITH DESCENDER */ \
+	  0x04BF, /* CYRILLIC SMALL LETTER ABKHASIAN CHE WITH DESCENDER */ }, \
+	{ 0x04E0, /* CYRILLIC CAPITAL LETTER ABKHASIAN DZE */ \
+	  0x04E1, /* CYRILLIC SMALL LETTER ABKHASIAN DZE */ }, \
+	{ 0x04A8, /* CYRILLIC CAPITAL LETTER ABKHASIAN HA */ \
+	  0x04A9, /* CYRILLIC SMALL LETTER ABKHASIAN HA */ }, \
+	{ 0x051E, /* CYRILLIC CAPITAL LETTER ALEUT KA */ \
+	  0x051F, /* CYRILLIC SMALL LETTER ALEUT KA */ }, \
+	{ 0x04E8, /* CYRILLIC CAPITAL LETTER BARRED O */ \
+	  0x04E9, /* CYRILLIC SMALL LETTER BARRED O */ }, \
+	{ 0x04EA, /* CYRILLIC CAPITAL LETTER BARRED O WITH DIAERESIS */ \
+	  0x04EB, /* CYRILLIC SMALL LETTER BARRED O WITH DIAERESIS */ }, \
+	{ 0x04A0, /* CYRILLIC CAPITAL LETTER BASHKIR KA */ \
+	  0x04A1, /* CYRILLIC SMALL LETTER BASHKIR KA */ }, \
+	{ 0x0411, /* CYRILLIC CAPITAL LETTER BE */ \
+	  0x0431, /* CYRILLIC SMALL LETTER BE */ }, \
+	{ 0x046A, /* CYRILLIC CAPITAL LETTER BIG YUS */ \
+	  0x046B, /* CYRILLIC SMALL LETTER BIG YUS */ }, \
+	{ 0xA66A, /* CYRILLIC CAPITAL LETTER BINOCULAR O */ \
+	  0xA66B, /* CYRILLIC SMALL LETTER BINOCULAR O */ }, \
+	{ 0xA65A, /* CYRILLIC CAPITAL LETTER BLENDED YUS */ \
+	  0xA65B, /* CYRILLIC SMALL LETTER BLENDED YUS */ }, \
+	{ 0xA64C, /* CYRILLIC CAPITAL LETTER BROAD OMEGA */ \
+	  0xA64D, /* CYRILLIC SMALL LETTER BROAD OMEGA */ }, \
+	{ 0x0406, /* CYRILLIC CAPITAL LETTER BYELORUSSIAN-UKRAINIAN I */ \
+	  0x0456, /* CYRILLIC SMALL LETTER BYELORUSSIAN-UKRAINIAN I */ }, \
+	{ 0xA686, /* CYRILLIC CAPITAL LETTER CCHE */ \
+	  0xA687, /* CYRILLIC SMALL LETTER CCHE */ }, \
+	{ 0x0427, /* CYRILLIC CAPITAL LETTER CHE */ \
+	  0x0447, /* CYRILLIC SMALL LETTER CHE */ }, \
+	{ 0x04B6, /* CYRILLIC CAPITAL LETTER CHE WITH DESCENDER */ \
+	  0x04B7, /* CYRILLIC SMALL LETTER CHE WITH DESCENDER */ }, \
+	{ 0x04F4, /* CYRILLIC CAPITAL LETTER CHE WITH DIAERESIS */ \
+	  0x04F5, /* CYRILLIC SMALL LETTER CHE WITH DIAERESIS */ }, \
+	{ 0x04B8, /* CYRILLIC CAPITAL LETTER CHE WITH VERTICAL STROKE */ \
+	  0x04B9, /* CYRILLIC SMALL LETTER CHE WITH VERTICAL STROKE */ }, \
+	{ 0xA658, /* CYRILLIC CAPITAL LETTER CLOSED LITTLE YUS */ \
+	  0xA659, /* CYRILLIC SMALL LETTER CLOSED LITTLE YUS */ }, \
+	{ 0xA69A, /* CYRILLIC CAPITAL LETTER CROSSED O */ \
+	  0xA69B, /* CYRILLIC SMALL LETTER CROSSED O */ }, \
+	{ 0x052C, /* CYRILLIC CAPITAL LETTER DCHE */ \
+	  0x052D, /* CYRILLIC SMALL LETTER DCHE */ }, \
+	{ 0x0414, /* CYRILLIC CAPITAL LETTER DE */ \
+	  0x0434, /* CYRILLIC SMALL LETTER DE */ }, \
+	{ 0x0402, /* CYRILLIC CAPITAL LETTER DJE */ \
+	  0x0452, /* CYRILLIC SMALL LETTER DJE */ }, \
+	{ 0xA648, /* CYRILLIC CAPITAL LETTER DJERV */ \
+	  0xA649, /* CYRILLIC SMALL LETTER DJERV */ }, \
+	{ 0xA66C, /* CYRILLIC CAPITAL LETTER DOUBLE MONOCULAR O */ \
+	  0xA66D, /* CYRILLIC SMALL LETTER DOUBLE MONOCULAR O */ }, \
+	{ 0xA698, /* CYRILLIC CAPITAL LETTER DOUBLE O */ \
+	  0xA699, /* CYRILLIC SMALL LETTER DOUBLE O */ }, \
+	{ 0xA680, /* CYRILLIC CAPITAL LETTER DWE */ \
+	  0xA681, /* CYRILLIC SMALL LETTER DWE */ }, \
+	{ 0x0405, /* CYRILLIC CAPITAL LETTER DZE */ \
+	  0x0455, /* CYRILLIC SMALL LETTER DZE */ }, \
+	{ 0xA642, /* CYRILLIC CAPITAL LETTER DZELO */ \
+	  0xA643, /* CYRILLIC SMALL LETTER DZELO */ }, \
+	{ 0x040F, /* CYRILLIC CAPITAL LETTER DZHE */ \
+	  0x045F, /* CYRILLIC SMALL LETTER DZHE */ }, \
+	{ 0xA682, /* CYRILLIC CAPITAL LETTER DZWE */ \
+	  0xA683, /* CYRILLIC SMALL LETTER DZWE */ }, \
+	{ 0xA688, /* CYRILLIC CAPITAL LETTER DZZE */ \
+	  0xA689, /* CYRILLIC SMALL LETTER DZZE */ }, \
+	{ 0x052A, /* CYRILLIC CAPITAL LETTER DZZHE */ \
+	  0x052B, /* CYRILLIC SMALL LETTER DZZHE */ }, \
+	{ 0x042D, /* CYRILLIC CAPITAL LETTER E */ \
+	  0x044D, /* CYRILLIC SMALL LETTER E */ }, \
+	{ 0x04EC, /* CYRILLIC CAPITAL LETTER E WITH DIAERESIS */ \
+	  0x04ED, /* CYRILLIC SMALL LETTER E WITH DIAERESIS */ }, \
+	{ 0x0424, /* CYRILLIC CAPITAL LETTER EF */ \
+	  0x0444, /* CYRILLIC SMALL LETTER EF */ }, \
+	{ 0x041B, /* CYRILLIC CAPITAL LETTER EL */ \
+	  0x043B, /* CYRILLIC SMALL LETTER EL */ }, \
+	{ 0x052E, /* CYRILLIC CAPITAL LETTER EL WITH DESCENDER */ \
+	  0x052F, /* CYRILLIC SMALL LETTER EL WITH DESCENDER */ }, \
+	{ 0x0512, /* CYRILLIC CAPITAL LETTER EL WITH HOOK */ \
+	  0x0513, /* CYRILLIC SMALL LETTER EL WITH HOOK */ }, \
+	{ 0x0520, /* CYRILLIC CAPITAL LETTER EL WITH MIDDLE HOOK */ \
+	  0x0521, /* CYRILLIC SMALL LETTER EL WITH MIDDLE HOOK */ }, \
+	{ 0x04C5, /* CYRILLIC CAPITAL LETTER EL WITH TAIL */ \
+	  0x04C6, /* CYRILLIC SMALL LETTER EL WITH TAIL */ }, \
+	{ 0x041C, /* CYRILLIC CAPITAL LETTER EM */ \
+	  0x043C, /* CYRILLIC SMALL LETTER EM */ }, \
+	{ 0x04CD, /* CYRILLIC CAPITAL LETTER EM WITH TAIL */ \
+	  0x04CE, /* CYRILLIC SMALL LETTER EM WITH TAIL */ }, \
+	{ 0x041D, /* CYRILLIC CAPITAL LETTER EN */ \
+	  0x043D, /* CYRILLIC SMALL LETTER EN */ }, \
+	{ 0x04A2, /* CYRILLIC CAPITAL LETTER EN WITH DESCENDER */ \
+	  0x04A3, /* CYRILLIC SMALL LETTER EN WITH DESCENDER */ }, \
+	{ 0x04C7, /* CYRILLIC CAPITAL LETTER EN WITH HOOK */ \
+	  0x04C8, /* CYRILLIC SMALL LETTER EN WITH HOOK */ }, \
+	{ 0x0528, /* CYRILLIC CAPITAL LETTER EN WITH LEFT HOOK */ \
+	  0x0529, /* CYRILLIC SMALL LETTER EN WITH LEFT HOOK */ }, \
+	{ 0x0522, /* CYRILLIC CAPITAL LETTER EN WITH MIDDLE HOOK */ \
+	  0x0523, /* CYRILLIC SMALL LETTER EN WITH MIDDLE HOOK */ }, \
+	{ 0x04C9, /* CYRILLIC CAPITAL LETTER EN WITH TAIL */ \
+	  0x04CA, /* CYRILLIC SMALL LETTER EN WITH TAIL */ }, \
+	{ 0x0420, /* CYRILLIC CAPITAL LETTER ER */ \
+	  0x0440, /* CYRILLIC SMALL LETTER ER */ }, \
+	{ 0x048E, /* CYRILLIC CAPITAL LETTER ER WITH TICK */ \
+	  0x048F, /* CYRILLIC SMALL LETTER ER WITH TICK */ }, \
+	{ 0x0421, /* CYRILLIC CAPITAL LETTER ES */ \
+	  0x0441, /* CYRILLIC SMALL LETTER ES */ }, \
+	{ 0x04AA, /* CYRILLIC CAPITAL LETTER ES WITH DESCENDER */ \
+	  0x04AB, /* CYRILLIC SMALL LETTER ES WITH DESCENDER */ }, \
+	{ 0x0472, /* CYRILLIC CAPITAL LETTER FITA */ \
+	  0x0473, /* CYRILLIC SMALL LETTER FITA */ }, \
+	{ 0x0413, /* CYRILLIC CAPITAL LETTER GHE */ \
+	  0x0433, /* CYRILLIC SMALL LETTER GHE */ }, \
+	{ 0x04F6, /* CYRILLIC CAPITAL LETTER GHE WITH DESCENDER */ \
+	  0x04F7, /* CYRILLIC SMALL LETTER GHE WITH DESCENDER */ }, \
+	{ 0x0494, /* CYRILLIC CAPITAL LETTER GHE WITH MIDDLE HOOK */ \
+	  0x0495, /* CYRILLIC SMALL LETTER GHE WITH MIDDLE HOOK */ }, \
+	{ 0x0492, /* CYRILLIC CAPITAL LETTER GHE WITH STROKE */ \
+	  0x0493, /* CYRILLIC SMALL LETTER GHE WITH STROKE */ }, \
+	{ 0x04FA, /* CYRILLIC CAPITAL LETTER GHE WITH STROKE AND HOOK */ \
+	  0x04FB, /* CYRILLIC SMALL LETTER GHE WITH STROKE AND HOOK */ }, \
+	{ 0x0490, /* CYRILLIC CAPITAL LETTER GHE WITH UPTURN */ \
+	  0x0491, /* CYRILLIC SMALL LETTER GHE WITH UPTURN */ }, \
+	{ 0x0403, /* CYRILLIC CAPITAL LETTER GJE */ \
+	  0x0453, /* CYRILLIC SMALL LETTER GJE */ }, \
+	{ 0x0425, /* CYRILLIC CAPITAL LETTER HA */ \
+	  0x0445, /* CYRILLIC SMALL LETTER HA */ }, \
+	{ 0x04B2, /* CYRILLIC CAPITAL LETTER HA WITH DESCENDER */ \
+	  0x04B3, /* CYRILLIC SMALL LETTER HA WITH DESCENDER */ }, \
+	{ 0x04FC, /* CYRILLIC CAPITAL LETTER HA WITH HOOK */ \
+	  0x04FD, /* CYRILLIC SMALL LETTER HA WITH HOOK */ }, \
+	{ 0x04FE, /* CYRILLIC CAPITAL LETTER HA WITH STROKE */ \
+	  0x04FF, /* CYRILLIC SMALL LETTER HA WITH STROKE */ }, \
+	{ 0x042A, /* CYRILLIC CAPITAL LETTER HARD SIGN */ \
+	  0x044A, /* CYRILLIC SMALL LETTER HARD SIGN */ }, \
+	{ 0xA694, /* CYRILLIC CAPITAL LETTER HWE */ \
+	  0xA695, /* CYRILLIC SMALL LETTER HWE */ }, \
+	{ 0x0418, /* CYRILLIC CAPITAL LETTER I */ \
+	  0x0438, /* CYRILLIC SMALL LETTER I */ }, \
+	{ 0x04E4, /* CYRILLIC CAPITAL LETTER I WITH DIAERESIS */ \
+	  0x04E5, /* CYRILLIC SMALL LETTER I WITH DIAERESIS */ }, \
+	{ 0x040D, /* CYRILLIC CAPITAL LETTER I WITH GRAVE */ \
+	  0x045D, /* CYRILLIC SMALL LETTER I WITH GRAVE */ }, \
+	{ 0x04E2, /* CYRILLIC CAPITAL LETTER I WITH MACRON */ \
+	  0x04E3, /* CYRILLIC SMALL LETTER I WITH MACRON */ }, \
+	{ 0x0415, /* CYRILLIC CAPITAL LETTER IE */ \
+	  0x0435, /* CYRILLIC SMALL LETTER IE */ }, \
+	{ 0x04D6, /* CYRILLIC CAPITAL LETTER IE WITH BREVE */ \
+	  0x04D7, /* CYRILLIC SMALL LETTER IE WITH BREVE */ }, \
+	{ 0x0400, /* CYRILLIC CAPITAL LETTER IE WITH GRAVE */ \
+	  0x0450, /* CYRILLIC SMALL LETTER IE WITH GRAVE */ }, \
+	{ 0x0401, /* CYRILLIC CAPITAL LETTER IO */ \
+	  0x0451, /* CYRILLIC SMALL LETTER IO */ }, \
+	{ 0xA646, /* CYRILLIC CAPITAL LETTER IOTA */ \
+	  0xA647, /* CYRILLIC SMALL LETTER IOTA */ }, \
+	{ 0xA656, /* CYRILLIC CAPITAL LETTER IOTIFIED A */ \
+	  0xA657, /* CYRILLIC SMALL LETTER IOTIFIED A */ }, \
+	{ 0x046C, /* CYRILLIC CAPITAL LETTER IOTIFIED BIG YUS */ \
+	  0x046D, /* CYRILLIC SMALL LETTER IOTIFIED BIG YUS */ }, \
+	{ 0xA65C, /* CYRILLIC CAPITAL LETTER IOTIFIED CLOSED LITTLE YUS */ \
+	  0xA65D, /* CYRILLIC SMALL LETTER IOTIFIED CLOSED LITTLE YUS */ }, \
+	{ 0x0464, /* CYRILLIC CAPITAL LETTER IOTIFIED E */ \
+	  0x0465, /* CYRILLIC SMALL LETTER IOTIFIED E */ }, \
+	{ 0x0468, /* CYRILLIC CAPITAL LETTER IOTIFIED LITTLE YUS */ \
+	  0x0469, /* CYRILLIC SMALL LETTER IOTIFIED LITTLE YUS */ }, \
+	{ 0xA652, /* CYRILLIC CAPITAL LETTER IOTIFIED YAT */ \
+	  0xA653, /* CYRILLIC SMALL LETTER IOTIFIED YAT */ }, \
+	{ 0x0474, /* CYRILLIC CAPITAL LETTER IZHITSA */ \
+	  0x0475, /* CYRILLIC SMALL LETTER IZHITSA */ }, \
+	{ 0x0476, /* CYRILLIC CAPITAL LETTER IZHITSA WITH DOUBLE GRAVE ACCENT */ \
+	  0x0477, /* CYRILLIC SMALL LETTER IZHITSA WITH DOUBLE GRAVE ACCENT */ }, \
+	{ 0x0408, /* CYRILLIC CAPITAL LETTER JE */ \
+	  0x0458, /* CYRILLIC SMALL LETTER JE */ }, \
+	{ 0x041A, /* CYRILLIC CAPITAL LETTER KA */ \
+	  0x043A, /* CYRILLIC SMALL LETTER KA */ }, \
+	{ 0x049A, /* CYRILLIC CAPITAL LETTER KA WITH DESCENDER */ \
+	  0x049B, /* CYRILLIC SMALL LETTER KA WITH DESCENDER */ }, \
+	{ 0x04C3, /* CYRILLIC CAPITAL LETTER KA WITH HOOK */ \
+	  0x04C4, /* CYRILLIC SMALL LETTER KA WITH HOOK */ }, \
+	{ 0x049E, /* CYRILLIC CAPITAL LETTER KA WITH STROKE */ \
+	  0x049F, /* CYRILLIC SMALL LETTER KA WITH STROKE */ }, \
+	{ 0x049C, /* CYRILLIC CAPITAL LETTER KA WITH VERTICAL STROKE */ \
+	  0x049D, /* CYRILLIC SMALL LETTER KA WITH VERTICAL STROKE */ }, \
+	{ 0x04CB, /* CYRILLIC CAPITAL LETTER KHAKASSIAN CHE */ \
+	  0x04CC, /* CYRILLIC SMALL LETTER KHAKASSIAN CHE */ }, \
+	{ 0x040C, /* CYRILLIC CAPITAL LETTER KJE */ \
+	  0x045C, /* CYRILLIC SMALL LETTER KJE */ }, \
+	{ 0x0500, /* CYRILLIC CAPITAL LETTER KOMI DE */ \
+	  0x0501, /* CYRILLIC SMALL LETTER KOMI DE */ }, \
+	{ 0x0502, /* CYRILLIC CAPITAL LETTER KOMI DJE */ \
+	  0x0503, /* CYRILLIC SMALL LETTER KOMI DJE */ }, \
+	{ 0x0506, /* CYRILLIC CAPITAL LETTER KOMI DZJE */ \
+	  0x0507, /* CYRILLIC SMALL LETTER KOMI DZJE */ }, \
+	{ 0x0508, /* CYRILLIC CAPITAL LETTER KOMI LJE */ \
+	  0x0509, /* CYRILLIC SMALL LETTER KOMI LJE */ }, \
+	{ 0x050A, /* CYRILLIC CAPITAL LETTER KOMI NJE */ \
+	  0x050B, /* CYRILLIC SMALL LETTER KOMI NJE */ }, \
+	{ 0x050C, /* CYRILLIC CAPITAL LETTER KOMI SJE */ \
+	  0x050D, /* CYRILLIC SMALL LETTER KOMI SJE */ }, \
+	{ 0x050E, /* CYRILLIC CAPITAL LETTER KOMI TJE */ \
+	  0x050F, /* CYRILLIC SMALL LETTER KOMI TJE */ }, \
+	{ 0x0504, /* CYRILLIC CAPITAL LETTER KOMI ZJE */ \
+	  0x0505, /* CYRILLIC SMALL LETTER KOMI ZJE */ }, \
+	{ 0x0480, /* CYRILLIC CAPITAL LETTER KOPPA */ \
+	  0x0481, /* CYRILLIC SMALL LETTER KOPPA */ }, \
+	{ 0x046E, /* CYRILLIC CAPITAL LETTER KSI */ \
+	  0x046F, /* CYRILLIC SMALL LETTER KSI */ }, \
+	{ 0x0514, /* CYRILLIC CAPITAL LETTER LHA */ \
+	  0x0515, /* CYRILLIC SMALL LETTER LHA */ }, \
+	{ 0x0466, /* CYRILLIC CAPITAL LETTER LITTLE YUS */ \
+	  0x0467, /* CYRILLIC SMALL LETTER LITTLE YUS */ }, \
+	{ 0x0409, /* CYRILLIC CAPITAL LETTER LJE */ \
+	  0x0459, /* CYRILLIC SMALL LETTER LJE */ }, \
+	{ 0xA668, /* CYRILLIC CAPITAL LETTER MONOCULAR O */ \
+	  0xA669, /* CYRILLIC SMALL LETTER MONOCULAR O */ }, \
+	{ 0xA64A, /* CYRILLIC CAPITAL LETTER MONOGRAPH UK */ \
+	  0xA64B, /* CYRILLIC SMALL LETTER MONOGRAPH UK */ }, \
+	{ 0xA64E, /* CYRILLIC CAPITAL LETTER NEUTRAL YER */ \
+	  0xA64F, /* CYRILLIC SMALL LETTER NEUTRAL YER */ }, \
+	{ 0x040A, /* CYRILLIC CAPITAL LETTER NJE */ \
+	  0x045A, /* CYRILLIC SMALL LETTER NJE */ }, \
+	{ 0x041E, /* CYRILLIC CAPITAL LETTER O */ \
+	  0x043E, /* CYRILLIC SMALL LETTER O */ }, \
+	{ 0x04E6, /* CYRILLIC CAPITAL LETTER O WITH DIAERESIS */ \
+	  0x04E7, /* CYRILLIC SMALL LETTER O WITH DIAERESIS */ }, \
+	{ 0x0460, /* CYRILLIC CAPITAL LETTER OMEGA */ \
+	  0x0461, /* CYRILLIC SMALL LETTER OMEGA */ }, \
+	{ 0x047C, /* CYRILLIC CAPITAL LETTER OMEGA WITH TITLO */ \
+	  0x047D, /* CYRILLIC SMALL LETTER OMEGA WITH TITLO */ }, \
+	{ 0x047E, /* CYRILLIC CAPITAL LETTER OT */ \
+	  0x047F, /* CYRILLIC SMALL LETTER OT */ }, \
+	{ 0x041F, /* CYRILLIC CAPITAL LETTER PE */ \
+	  0x043F, /* CYRILLIC SMALL LETTER PE */ }, \
+	{ 0x0524, /* CYRILLIC CAPITAL LETTER PE WITH DESCENDER */ \
+	  0x0525, /* CYRILLIC SMALL LETTER PE WITH DESCENDER */ }, \
+	{ 0x04A6, /* CYRILLIC CAPITAL LETTER PE WITH MIDDLE HOOK */ \
+	  0x04A7, /* CYRILLIC SMALL LETTER PE WITH MIDDLE HOOK */ }, \
+	{ 0x0470, /* CYRILLIC CAPITAL LETTER PSI */ \
+	  0x0471, /* CYRILLIC SMALL LETTER PSI */ }, \
+	{ 0x051A, /* CYRILLIC CAPITAL LETTER QA */ \
+	  0x051B, /* CYRILLIC SMALL LETTER QA */ }, \
+	{ 0xA644, /* CYRILLIC CAPITAL LETTER REVERSED DZE */ \
+	  0xA645, /* CYRILLIC SMALL LETTER REVERSED DZE */ }, \
+	{ 0xA660, /* CYRILLIC CAPITAL LETTER REVERSED TSE */ \
+	  0xA661, /* CYRILLIC SMALL LETTER REVERSED TSE */ }, \
+	{ 0xA654, /* CYRILLIC CAPITAL LETTER REVERSED YU */ \
+	  0xA655, /* CYRILLIC SMALL LETTER REVERSED YU */ }, \
+	{ 0x0510, /* CYRILLIC CAPITAL LETTER REVERSED ZE */ \
+	  0x0511, /* CYRILLIC SMALL LETTER REVERSED ZE */ }, \
+	{ 0x0516, /* CYRILLIC CAPITAL LETTER RHA */ \
+	  0x0517, /* CYRILLIC SMALL LETTER RHA */ }, \
+	{ 0x047A, /* CYRILLIC CAPITAL LETTER ROUND OMEGA */ \
+	  0x047B, /* CYRILLIC SMALL LETTER ROUND OMEGA */ }, \
+	{ 0x04D8, /* CYRILLIC CAPITAL LETTER SCHWA */ \
+	  0x04D9, /* CYRILLIC SMALL LETTER SCHWA */ }, \
+	{ 0x04DA, /* CYRILLIC CAPITAL LETTER SCHWA WITH DIAERESIS */ \
+	  0x04DB, /* CYRILLIC SMALL LETTER SCHWA WITH DIAERESIS */ }, \
+	{ 0x048C, /* CYRILLIC CAPITAL LETTER SEMISOFT SIGN */ \
+	  0x048D, /* CYRILLIC SMALL LETTER SEMISOFT SIGN */ }, \
+	{ 0x0428, /* CYRILLIC CAPITAL LETTER SHA */ \
+	  0x0448, /* CYRILLIC SMALL LETTER SHA */ }, \
+	{ 0x0429, /* CYRILLIC CAPITAL LETTER SHCHA */ \
+	  0x0449, /* CYRILLIC SMALL LETTER SHCHA */ }, \
+	{ 0x04BA, /* CYRILLIC CAPITAL LETTER SHHA */ \
+	  0x04BB, /* CYRILLIC SMALL LETTER SHHA */ }, \
+	{ 0x0526, /* CYRILLIC CAPITAL LETTER SHHA WITH DESCENDER */ \
+	  0x0527, /* CYRILLIC SMALL LETTER SHHA WITH DESCENDER */ }, \
+	{ 0x048A, /* CYRILLIC CAPITAL LETTER SHORT I WITH TAIL */ \
+	  0x048B, /* CYRILLIC SMALL LETTER SHORT I WITH TAIL */ }, \
+	{ 0x040E, /* CYRILLIC CAPITAL LETTER SHORT U */ \
+	  0x045E, /* CYRILLIC SMALL LETTER SHORT U */ }, \
+	{ 0xA696, /* CYRILLIC CAPITAL LETTER SHWE */ \
+	  0xA697, /* CYRILLIC SMALL LETTER SHWE */ }, \
+	{ 0xA662, /* CYRILLIC CAPITAL LETTER SOFT DE */ \
+	  0xA663, /* CYRILLIC SMALL LETTER SOFT DE */ }, \
+	{ 0xA664, /* CYRILLIC CAPITAL LETTER SOFT EL */ \
+	  0xA665, /* CYRILLIC SMALL LETTER SOFT EL */ }, \
+	{ 0xA666, /* CYRILLIC CAPITAL LETTER SOFT EM */ \
+	  0xA667, /* CYRILLIC SMALL LETTER SOFT EM */ }, \
+	{ 0x042C, /* CYRILLIC CAPITAL LETTER SOFT SIGN */ \
+	  0x044C, /* CYRILLIC SMALL LETTER SOFT SIGN */ }, \
+	{ 0x04AE, /* CYRILLIC CAPITAL LETTER STRAIGHT U */ \
+	  0x04AF, /* CYRILLIC SMALL LETTER STRAIGHT U */ }, \
+	{ 0x04B0, /* CYRILLIC CAPITAL LETTER STRAIGHT U WITH STROKE */ \
+	  0x04B1, /* CYRILLIC SMALL LETTER STRAIGHT U WITH STROKE */ }, \
+	{ 0xA692, /* CYRILLIC CAPITAL LETTER TCHE */ \
+	  0xA693, /* CYRILLIC SMALL LETTER TCHE */ }, \
+	{ 0x0422, /* CYRILLIC CAPITAL LETTER TE */ \
+	  0x0442, /* CYRILLIC SMALL LETTER TE */ }, \
+	{ 0x04AC, /* CYRILLIC CAPITAL LETTER TE WITH DESCENDER */ \
+	  0x04AD, /* CYRILLIC SMALL LETTER TE WITH DESCENDER */ }, \
+	{ 0xA68A, /* CYRILLIC CAPITAL LETTER TE WITH MIDDLE HOOK */ \
+	  0xA68B, /* CYRILLIC SMALL LETTER TE WITH MIDDLE HOOK */ }, \
+	{ 0x0426, /* CYRILLIC CAPITAL LETTER TSE */ \
+	  0x0446, /* CYRILLIC SMALL LETTER TSE */ }, \
+	{ 0x040B, /* CYRILLIC CAPITAL LETTER TSHE */ \
+	  0x045B, /* CYRILLIC SMALL LETTER TSHE */ }, \
+	{ 0xA690, /* CYRILLIC CAPITAL LETTER TSSE */ \
+	  0xA691, /* CYRILLIC SMALL LETTER TSSE */ }, \
+	{ 0xA68E, /* CYRILLIC CAPITAL LETTER TSWE */ \
+	  0xA68F, /* CYRILLIC SMALL LETTER TSWE */ }, \
+	{ 0xA68C, /* CYRILLIC CAPITAL LETTER TWE */ \
+	  0xA68D, /* CYRILLIC SMALL LETTER TWE */ }, \
+	{ 0x0423, /* CYRILLIC CAPITAL LETTER U */ \
+	  0x0443, /* CYRILLIC SMALL LETTER U */ }, \
+	{ 0x04F0, /* CYRILLIC CAPITAL LETTER U WITH DIAERESIS */ \
+	  0x04F1, /* CYRILLIC SMALL LETTER U WITH DIAERESIS */ }, \
+	{ 0x04F2, /* CYRILLIC CAPITAL LETTER U WITH DOUBLE ACUTE */ \
+	  0x04F3, /* CYRILLIC SMALL LETTER U WITH DOUBLE ACUTE */ }, \
+	{ 0x04EE, /* CYRILLIC CAPITAL LETTER U WITH MACRON */ \
+	  0x04EF, /* CYRILLIC SMALL LETTER U WITH MACRON */ }, \
+	{ 0x0478, /* CYRILLIC CAPITAL LETTER UK */ \
+	  0x0479, /* CYRILLIC SMALL LETTER UK */ }, \
+	{ 0x0404, /* CYRILLIC CAPITAL LETTER UKRAINIAN IE */ \
+	  0x0454, /* CYRILLIC SMALL LETTER UKRAINIAN IE */ }, \
+	{ 0x0412, /* CYRILLIC CAPITAL LETTER VE */ \
+	  0x0432, /* CYRILLIC SMALL LETTER VE */ }, \
+	{ 0x051C, /* CYRILLIC CAPITAL LETTER WE */ \
+	  0x051D, /* CYRILLIC SMALL LETTER WE */ }, \
+	{ 0x042F, /* CYRILLIC CAPITAL LETTER YA */ \
+	  0x044F, /* CYRILLIC SMALL LETTER YA */ }, \
+	{ 0x0518, /* CYRILLIC CAPITAL LETTER YAE */ \
+	  0x0519, /* CYRILLIC SMALL LETTER YAE */ }, \
+	{ 0x0462, /* CYRILLIC CAPITAL LETTER YAT */ \
+	  0x0463, /* CYRILLIC SMALL LETTER YAT */ }, \
+	{ 0x042B, /* CYRILLIC CAPITAL LETTER YERU */ \
+	  0x044B, /* CYRILLIC SMALL LETTER YERU */ }, \
+	{ 0xA650, /* CYRILLIC CAPITAL LETTER YERU WITH BACK YER */ \
+	  0xA651, /* CYRILLIC SMALL LETTER YERU WITH BACK YER */ }, \
+	{ 0x04F8, /* CYRILLIC CAPITAL LETTER YERU WITH DIAERESIS */ \
+	  0x04F9, /* CYRILLIC SMALL LETTER YERU WITH DIAERESIS */ }, \
+	{ 0x0407, /* CYRILLIC CAPITAL LETTER YI */ \
+	  0x0457, /* CYRILLIC SMALL LETTER YI */ }, \
+	{ 0xA65E, /* CYRILLIC CAPITAL LETTER YN */ \
+	  0xA65F, /* CYRILLIC SMALL LETTER YN */ }, \
+	{ 0x042E, /* CYRILLIC CAPITAL LETTER YU */ \
+	  0x044E, /* CYRILLIC SMALL LETTER YU */ }, \
+	{ 0x0417, /* CYRILLIC CAPITAL LETTER ZE */ \
+	  0x0437, /* CYRILLIC SMALL LETTER ZE */ }, \
+	{ 0x0498, /* CYRILLIC CAPITAL LETTER ZE WITH DESCENDER */ \
+	  0x0499, /* CYRILLIC SMALL LETTER ZE WITH DESCENDER */ }, \
+	{ 0x04DE, /* CYRILLIC CAPITAL LETTER ZE WITH DIAERESIS */ \
+	  0x04DF, /* CYRILLIC SMALL LETTER ZE WITH DIAERESIS */ }, \
+	{ 0xA640, /* CYRILLIC CAPITAL LETTER ZEMLYA */ \
+	  0xA641, /* CYRILLIC SMALL LETTER ZEMLYA */ }, \
+	{ 0x0416, /* CYRILLIC CAPITAL LETTER ZHE */ \
+	  0x0436, /* CYRILLIC SMALL LETTER ZHE */ }, \
+	{ 0x04C1, /* CYRILLIC CAPITAL LETTER ZHE WITH BREVE */ \
+	  0x04C2, /* CYRILLIC SMALL LETTER ZHE WITH BREVE */ }, \
+	{ 0x0496, /* CYRILLIC CAPITAL LETTER ZHE WITH DESCENDER */ \
+	  0x0497, /* CYRILLIC SMALL LETTER ZHE WITH DESCENDER */ }, \
+	{ 0x04DC, /* CYRILLIC CAPITAL LETTER ZHE WITH DIAERESIS */ \
+	  0x04DD, /* CYRILLIC SMALL LETTER ZHE WITH DIAERESIS */ }, \
+	{ 0xA684, /* CYRILLIC CAPITAL LETTER ZHWE */ \
+	  0xA685, /* CYRILLIC SMALL LETTER ZHWE */ }, \
+	{ 0xFF21, /* FULLWIDTH LATIN CAPITAL LETTER A */ \
+	  0xFF41, /* FULLWIDTH LATIN SMALL LETTER A */ }, \
+	{ 0xFF22, /* FULLWIDTH LATIN CAPITAL LETTER B */ \
+	  0xFF42, /* FULLWIDTH LATIN SMALL LETTER B */ }, \
+	{ 0xFF23, /* FULLWIDTH LATIN CAPITAL LETTER C */ \
+	  0xFF43, /* FULLWIDTH LATIN SMALL LETTER C */ }, \
+	{ 0xFF24, /* FULLWIDTH LATIN CAPITAL LETTER D */ \
+	  0xFF44, /* FULLWIDTH LATIN SMALL LETTER D */ }, \
+	{ 0xFF25, /* FULLWIDTH LATIN CAPITAL LETTER E */ \
+	  0xFF45, /* FULLWIDTH LATIN SMALL LETTER E */ }, \
+	{ 0xFF26, /* FULLWIDTH LATIN CAPITAL LETTER F */ \
+	  0xFF46, /* FULLWIDTH LATIN SMALL LETTER F */ }, \
+	{ 0xFF27, /* FULLWIDTH LATIN CAPITAL LETTER G */ \
+	  0xFF47, /* FULLWIDTH LATIN SMALL LETTER G */ }, \
+	{ 0xFF28, /* FULLWIDTH LATIN CAPITAL LETTER H */ \
+	  0xFF48, /* FULLWIDTH LATIN SMALL LETTER H */ }, \
+	{ 0xFF29, /* FULLWIDTH LATIN CAPITAL LETTER I */ \
+	  0xFF49, /* FULLWIDTH LATIN SMALL LETTER I */ }, \
+	{ 0xFF2A, /* FULLWIDTH LATIN CAPITAL LETTER J */ \
+	  0xFF4A, /* FULLWIDTH LATIN SMALL LETTER J */ }, \
+	{ 0xFF2B, /* FULLWIDTH LATIN CAPITAL LETTER K */ \
+	  0xFF4B, /* FULLWIDTH LATIN SMALL LETTER K */ }, \
+	{ 0xFF2C, /* FULLWIDTH LATIN CAPITAL LETTER L */ \
+	  0xFF4C, /* FULLWIDTH LATIN SMALL LETTER L */ }, \
+	{ 0xFF2D, /* FULLWIDTH LATIN CAPITAL LETTER M */ \
+	  0xFF4D, /* FULLWIDTH LATIN SMALL LETTER M */ }, \
+	{ 0xFF2E, /* FULLWIDTH LATIN CAPITAL LETTER N */ \
+	  0xFF4E, /* FULLWIDTH LATIN SMALL LETTER N */ }, \
+	{ 0xFF2F, /* FULLWIDTH LATIN CAPITAL LETTER O */ \
+	  0xFF4F, /* FULLWIDTH LATIN SMALL LETTER O */ }, \
+	{ 0xFF30, /* FULLWIDTH LATIN CAPITAL LETTER P */ \
+	  0xFF50, /* FULLWIDTH LATIN SMALL LETTER P */ }, \
+	{ 0xFF31, /* FULLWIDTH LATIN CAPITAL LETTER Q */ \
+	  0xFF51, /* FULLWIDTH LATIN SMALL LETTER Q */ }, \
+	{ 0xFF32, /* FULLWIDTH LATIN CAPITAL LETTER R */ \
+	  0xFF52, /* FULLWIDTH LATIN SMALL LETTER R */ }, \
+	{ 0xFF33, /* FULLWIDTH LATIN CAPITAL LETTER S */ \
+	  0xFF53, /* FULLWIDTH LATIN SMALL LETTER S */ }, \
+	{ 0xFF34, /* FULLWIDTH LATIN CAPITAL LETTER T */ \
+	  0xFF54, /* FULLWIDTH LATIN SMALL LETTER T */ }, \
+	{ 0xFF35, /* FULLWIDTH LATIN CAPITAL LETTER U */ \
+	  0xFF55, /* FULLWIDTH LATIN SMALL LETTER U */ }, \
+	{ 0xFF36, /* FULLWIDTH LATIN CAPITAL LETTER V */ \
+	  0xFF56, /* FULLWIDTH LATIN SMALL LETTER V */ }, \
+	{ 0xFF37, /* FULLWIDTH LATIN CAPITAL LETTER W */ \
+	  0xFF57, /* FULLWIDTH LATIN SMALL LETTER W */ }, \
+	{ 0xFF38, /* FULLWIDTH LATIN CAPITAL LETTER X */ \
+	  0xFF58, /* FULLWIDTH LATIN SMALL LETTER X */ }, \
+	{ 0xFF39, /* FULLWIDTH LATIN CAPITAL LETTER Y */ \
+	  0xFF59, /* FULLWIDTH LATIN SMALL LETTER Y */ }, \
+	{ 0xFF3A, /* FULLWIDTH LATIN CAPITAL LETTER Z */ \
+	  0xFF5A, /* FULLWIDTH LATIN SMALL LETTER Z */ }, \
+	{ 0x10CD, /* GEORGIAN CAPITAL LETTER AEN */ \
+	  0x2D2D, /* GEORGIAN SMALL LETTER AEN */ }, \
+	{ 0x10A0, /* GEORGIAN CAPITAL LETTER AN */ \
+	  0x2D00, /* GEORGIAN SMALL LETTER AN */ }, \
+	{ 0x10A1, /* GEORGIAN CAPITAL LETTER BAN */ \
+	  0x2D01, /* GEORGIAN SMALL LETTER BAN */ }, \
+	{ 0x10BA, /* GEORGIAN CAPITAL LETTER CAN */ \
+	  0x2D1A, /* GEORGIAN SMALL LETTER CAN */ }, \
+	{ 0x10BD, /* GEORGIAN CAPITAL LETTER CHAR */ \
+	  0x2D1D, /* GEORGIAN SMALL LETTER CHAR */ }, \
+	{ 0x10B9, /* GEORGIAN CAPITAL LETTER CHIN */ \
+	  0x2D19, /* GEORGIAN SMALL LETTER CHIN */ }, \
+	{ 0x10BC, /* GEORGIAN CAPITAL LETTER CIL */ \
+	  0x2D1C, /* GEORGIAN SMALL LETTER CIL */ }, \
+	{ 0x10A3, /* GEORGIAN CAPITAL LETTER DON */ \
+	  0x2D03, /* GEORGIAN SMALL LETTER DON */ }, \
+	{ 0x10A4, /* GEORGIAN CAPITAL LETTER EN */ \
+	  0x2D04, /* GEORGIAN SMALL LETTER EN */ }, \
+	{ 0x10A2, /* GEORGIAN CAPITAL LETTER GAN */ \
+	  0x2D02, /* GEORGIAN SMALL LETTER GAN */ }, \
+	{ 0x10B6, /* GEORGIAN CAPITAL LETTER GHAN */ \
+	  0x2D16, /* GEORGIAN SMALL LETTER GHAN */ }, \
+	{ 0x10C0, /* GEORGIAN CAPITAL LETTER HAE */ \
+	  0x2D20, /* GEORGIAN SMALL LETTER HAE */ }, \
+	{ 0x10C4, /* GEORGIAN CAPITAL LETTER HAR */ \
+	  0x2D24, /* GEORGIAN SMALL LETTER HAR */ }, \
+	{ 0x10C1, /* GEORGIAN CAPITAL LETTER HE */ \
+	  0x2D21, /* GEORGIAN SMALL LETTER HE */ }, \
+	{ 0x10C2, /* GEORGIAN CAPITAL LETTER HIE */ \
+	  0x2D22, /* GEORGIAN SMALL LETTER HIE */ }, \
+	{ 0x10C5, /* GEORGIAN CAPITAL LETTER HOE */ \
+	  0x2D25, /* GEORGIAN SMALL LETTER HOE */ }, \
+	{ 0x10A8, /* GEORGIAN CAPITAL LETTER IN */ \
+	  0x2D08, /* GEORGIAN SMALL LETTER IN */ }, \
+	{ 0x10BF, /* GEORGIAN CAPITAL LETTER JHAN */ \
+	  0x2D1F, /* GEORGIAN SMALL LETTER JHAN */ }, \
+	{ 0x10BB, /* GEORGIAN CAPITAL LETTER JIL */ \
+	  0x2D1B, /* GEORGIAN SMALL LETTER JIL */ }, \
+	{ 0x10A9, /* GEORGIAN CAPITAL LETTER KAN */ \
+	  0x2D09, /* GEORGIAN SMALL LETTER KAN */ }, \
+	{ 0x10B5, /* GEORGIAN CAPITAL LETTER KHAR */ \
+	  0x2D15, /* GEORGIAN SMALL LETTER KHAR */ }, \
+	{ 0x10AA, /* GEORGIAN CAPITAL LETTER LAS */ \
+	  0x2D0A, /* GEORGIAN SMALL LETTER LAS */ }, \
+	{ 0x10AB, /* GEORGIAN CAPITAL LETTER MAN */ \
+	  0x2D0B, /* GEORGIAN SMALL LETTER MAN */ }, \
+	{ 0x10AC, /* GEORGIAN CAPITAL LETTER NAR */ \
+	  0x2D0C, /* GEORGIAN SMALL LETTER NAR */ }, \
+	{ 0x10AD, /* GEORGIAN CAPITAL LETTER ON */ \
+	  0x2D0D, /* GEORGIAN SMALL LETTER ON */ }, \
+	{ 0x10AE, /* GEORGIAN CAPITAL LETTER PAR */ \
+	  0x2D0E, /* GEORGIAN SMALL LETTER PAR */ }, \
+	{ 0x10B4, /* GEORGIAN CAPITAL LETTER PHAR */ \
+	  0x2D14, /* GEORGIAN SMALL LETTER PHAR */ }, \
+	{ 0x10B7, /* GEORGIAN CAPITAL LETTER QAR */ \
+	  0x2D17, /* GEORGIAN SMALL LETTER QAR */ }, \
+	{ 0x10B0, /* GEORGIAN CAPITAL LETTER RAE */ \
+	  0x2D10, /* GEORGIAN SMALL LETTER RAE */ }, \
+	{ 0x10B1, /* GEORGIAN CAPITAL LETTER SAN */ \
+	  0x2D11, /* GEORGIAN SMALL LETTER SAN */ }, \
+	{ 0x10B8, /* GEORGIAN CAPITAL LETTER SHIN */ \
+	  0x2D18, /* GEORGIAN SMALL LETTER SHIN */ }, \
+	{ 0x10A7, /* GEORGIAN CAPITAL LETTER TAN */ \
+	  0x2D07, /* GEORGIAN SMALL LETTER TAN */ }, \
+	{ 0x10B2, /* GEORGIAN CAPITAL LETTER TAR */ \
+	  0x2D12, /* GEORGIAN SMALL LETTER TAR */ }, \
+	{ 0x10B3, /* GEORGIAN CAPITAL LETTER UN */ \
+	  0x2D13, /* GEORGIAN SMALL LETTER UN */ }, \
+	{ 0x10A5, /* GEORGIAN CAPITAL LETTER VIN */ \
+	  0x2D05, /* GEORGIAN SMALL LETTER VIN */ }, \
+	{ 0x10C3, /* GEORGIAN CAPITAL LETTER WE */ \
+	  0x2D23, /* GEORGIAN SMALL LETTER WE */ }, \
+	{ 0x10BE, /* GEORGIAN CAPITAL LETTER XAN */ \
+	  0x2D1E, /* GEORGIAN SMALL LETTER XAN */ }, \
+	{ 0x10C7, /* GEORGIAN CAPITAL LETTER YN */ \
+	  0x2D27, /* GEORGIAN SMALL LETTER YN */ }, \
+	{ 0x10A6, /* GEORGIAN CAPITAL LETTER ZEN */ \
+	  0x2D06, /* GEORGIAN SMALL LETTER ZEN */ }, \
+	{ 0x10AF, /* GEORGIAN CAPITAL LETTER ZHAR */ \
+	  0x2D0F, /* GEORGIAN SMALL LETTER ZHAR */ }, \
+	{ 0x2C00, /* GLAGOLITIC CAPITAL LETTER AZU */ \
+	  0x2C30, /* GLAGOLITIC SMALL LETTER AZU */ }, \
+	{ 0x2C28, /* GLAGOLITIC CAPITAL LETTER BIG YUS */ \
+	  0x2C58, /* GLAGOLITIC SMALL LETTER BIG YUS */ }, \
+	{ 0x2C01, /* GLAGOLITIC CAPITAL LETTER BUKY */ \
+	  0x2C31, /* GLAGOLITIC SMALL LETTER BUKY */ }, \
+	{ 0x2C1D, /* GLAGOLITIC CAPITAL LETTER CHRIVI */ \
+	  0x2C4D, /* GLAGOLITIC SMALL LETTER CHRIVI */ }, \
+	{ 0x2C0C, /* GLAGOLITIC CAPITAL LETTER DJERVI */ \
+	  0x2C3C, /* GLAGOLITIC SMALL LETTER DJERVI */ }, \
+	{ 0x2C04, /* GLAGOLITIC CAPITAL LETTER DOBRO */ \
+	  0x2C34, /* GLAGOLITIC SMALL LETTER DOBRO */ }, \
+	{ 0x2C07, /* GLAGOLITIC CAPITAL LETTER DZELO */ \
+	  0x2C37, /* GLAGOLITIC SMALL LETTER DZELO */ }, \
+	{ 0x2C2A, /* GLAGOLITIC CAPITAL LETTER FITA */ \
+	  0x2C5A, /* GLAGOLITIC SMALL LETTER FITA */ }, \
+	{ 0x2C17, /* GLAGOLITIC CAPITAL LETTER FRITU */ \
+	  0x2C47, /* GLAGOLITIC SMALL LETTER FRITU */ }, \
+	{ 0x2C03, /* GLAGOLITIC CAPITAL LETTER GLAGOLI */ \
+	  0x2C33, /* GLAGOLITIC SMALL LETTER GLAGOLI */ }, \
+	{ 0x2C18, /* GLAGOLITIC CAPITAL LETTER HERU */ \
+	  0x2C48, /* GLAGOLITIC SMALL LETTER HERU */ }, \
+	{ 0x2C0B, /* GLAGOLITIC CAPITAL LETTER I */ \
+	  0x2C3B, /* GLAGOLITIC SMALL LETTER I */ }, \
+	{ 0x2C0A, /* GLAGOLITIC CAPITAL LETTER INITIAL IZHE */ \
+	  0x2C3A, /* GLAGOLITIC SMALL LETTER INITIAL IZHE */ }, \
+	{ 0x2C29, /* GLAGOLITIC CAPITAL LETTER IOTATED BIG YUS */ \
+	  0x2C59, /* GLAGOLITIC SMALL LETTER IOTATED BIG YUS */ }, \
+	{ 0x2C27, /* GLAGOLITIC CAPITAL LETTER IOTATED SMALL YUS */ \
+	  0x2C57, /* GLAGOLITIC SMALL LETTER IOTATED SMALL YUS */ }, \
+	{ 0x2C09, /* GLAGOLITIC CAPITAL LETTER IZHE */ \
+	  0x2C39, /* GLAGOLITIC SMALL LETTER IZHE */ }, \
+	{ 0x2C2B, /* GLAGOLITIC CAPITAL LETTER IZHITSA */ \
+	  0x2C5B, /* GLAGOLITIC SMALL LETTER IZHITSA */ }, \
+	{ 0x2C0D, /* GLAGOLITIC CAPITAL LETTER KAKO */ \
+	  0x2C3D, /* GLAGOLITIC SMALL LETTER KAKO */ }, \
+	{ 0x2C2E, /* GLAGOLITIC CAPITAL LETTER LATINATE MYSLITE */ \
+	  0x2C5E, /* GLAGOLITIC SMALL LETTER LATINATE MYSLITE */ }, \
+	{ 0x2C0E, /* GLAGOLITIC CAPITAL LETTER LJUDIJE */ \
+	  0x2C3E, /* GLAGOLITIC SMALL LETTER LJUDIJE */ }, \
+	{ 0x2C0F, /* GLAGOLITIC CAPITAL LETTER MYSLITE */ \
+	  0x2C3F, /* GLAGOLITIC SMALL LETTER MYSLITE */ }, \
+	{ 0x2C10, /* GLAGOLITIC CAPITAL LETTER NASHI */ \
+	  0x2C40, /* GLAGOLITIC SMALL LETTER NASHI */ }, \
+	{ 0x2C11, /* GLAGOLITIC CAPITAL LETTER ONU */ \
+	  0x2C41, /* GLAGOLITIC SMALL LETTER ONU */ }, \
+	{ 0x2C19, /* GLAGOLITIC CAPITAL LETTER OTU */ \
+	  0x2C49, /* GLAGOLITIC SMALL LETTER OTU */ }, \
+	{ 0x2C1A, /* GLAGOLITIC CAPITAL LETTER PE */ \
+	  0x2C4A, /* GLAGOLITIC SMALL LETTER PE */ }, \
+	{ 0x2C12, /* GLAGOLITIC CAPITAL LETTER POKOJI */ \
+	  0x2C42, /* GLAGOLITIC SMALL LETTER POKOJI */ }, \
+	{ 0x2C13, /* GLAGOLITIC CAPITAL LETTER RITSI */ \
+	  0x2C43, /* GLAGOLITIC SMALL LETTER RITSI */ }, \
+	{ 0x2C1E, /* GLAGOLITIC CAPITAL LETTER SHA */ \
+	  0x2C4E, /* GLAGOLITIC SMALL LETTER SHA */ }, \
+	{ 0x2C1B, /* GLAGOLITIC CAPITAL LETTER SHTA */ \
+	  0x2C4B, /* GLAGOLITIC SMALL LETTER SHTA */ }, \
+	{ 0x2C2C, /* GLAGOLITIC CAPITAL LETTER SHTAPIC */ \
+	  0x2C5C, /* GLAGOLITIC SMALL LETTER SHTAPIC */ }, \
+	{ 0x2C14, /* GLAGOLITIC CAPITAL LETTER SLOVO */ \
+	  0x2C44, /* GLAGOLITIC SMALL LETTER SLOVO */ }, \
+	{ 0x2C24, /* GLAGOLITIC CAPITAL LETTER SMALL YUS */ \
+	  0x2C54, /* GLAGOLITIC SMALL LETTER SMALL YUS */ }, \
+	{ 0x2C25, /* GLAGOLITIC CAPITAL LETTER SMALL YUS WITH TAIL */ \
+	  0x2C55, /* GLAGOLITIC SMALL LETTER SMALL YUS WITH TAIL */ }, \
+	{ 0x2C22, /* GLAGOLITIC CAPITAL LETTER SPIDERY HA */ \
+	  0x2C52, /* GLAGOLITIC SMALL LETTER SPIDERY HA */ }, \
+	{ 0x2C2D, /* GLAGOLITIC CAPITAL LETTER TROKUTASTI A */ \
+	  0x2C5D, /* GLAGOLITIC SMALL LETTER TROKUTASTI A */ }, \
+	{ 0x2C1C, /* GLAGOLITIC CAPITAL LETTER TSI */ \
+	  0x2C4C, /* GLAGOLITIC SMALL LETTER TSI */ }, \
+	{ 0x2C15, /* GLAGOLITIC CAPITAL LETTER TVRIDO */ \
+	  0x2C45, /* GLAGOLITIC SMALL LETTER TVRIDO */ }, \
+	{ 0x2C16, /* GLAGOLITIC CAPITAL LETTER UKU */ \
+	  0x2C46, /* GLAGOLITIC SMALL LETTER UKU */ }, \
+	{ 0x2C02, /* GLAGOLITIC CAPITAL LETTER VEDE */ \
+	  0x2C32, /* GLAGOLITIC SMALL LETTER VEDE */ }, \
+	{ 0x2C21, /* GLAGOLITIC CAPITAL LETTER YATI */ \
+	  0x2C51, /* GLAGOLITIC SMALL LETTER YATI */ }, \
+	{ 0x2C20, /* GLAGOLITIC CAPITAL LETTER YERI */ \
+	  0x2C50, /* GLAGOLITIC SMALL LETTER YERI */ }, \
+	{ 0x2C1F, /* GLAGOLITIC CAPITAL LETTER YERU */ \
+	  0x2C4F, /* GLAGOLITIC SMALL LETTER YERU */ }, \
+	{ 0x2C05, /* GLAGOLITIC CAPITAL LETTER YESTU */ \
+	  0x2C35, /* GLAGOLITIC SMALL LETTER YESTU */ }, \
+	{ 0x2C26, /* GLAGOLITIC CAPITAL LETTER YO */ \
+	  0x2C56, /* GLAGOLITIC SMALL LETTER YO */ }, \
+	{ 0x2C23, /* GLAGOLITIC CAPITAL LETTER YU */ \
+	  0x2C53, /* GLAGOLITIC SMALL LETTER YU */ }, \
+	{ 0x2C08, /* GLAGOLITIC CAPITAL LETTER ZEMLJA */ \
+	  0x2C38, /* GLAGOLITIC SMALL LETTER ZEMLJA */ }, \
+	{ 0x2C06, /* GLAGOLITIC CAPITAL LETTER ZHIVETE */ \
+	  0x2C36, /* GLAGOLITIC SMALL LETTER ZHIVETE */ }, \
+	{ 0x0391, /* GREEK CAPITAL LETTER ALPHA */ \
+	  0x03B1, /* GREEK SMALL LETTER ALPHA */ }, \
+	{ 0x1F09, /* GREEK CAPITAL LETTER ALPHA WITH DASIA */ \
+	  0x1F01, /* GREEK SMALL LETTER ALPHA WITH DASIA */ }, \
+	{ 0x1F0D, /* GREEK CAPITAL LETTER ALPHA WITH DASIA AND OXIA */ \
+	  0x1F05, /* GREEK SMALL LETTER ALPHA WITH DASIA AND OXIA */ }, \
+	{ 0x1F0F, /* GREEK CAPITAL LETTER ALPHA WITH DASIA AND PERISPOMENI */ \
+	  0x1F07, /* GREEK SMALL LETTER ALPHA WITH DASIA AND PERISPOMENI */ }, \
+	{ 0x1F0B, /* GREEK CAPITAL LETTER ALPHA WITH DASIA AND VARIA */ \
+	  0x1F03, /* GREEK SMALL LETTER ALPHA WITH DASIA AND VARIA */ }, \
+	{ 0x1FB9, /* GREEK CAPITAL LETTER ALPHA WITH MACRON */ \
+	  0x1FB1, /* GREEK SMALL LETTER ALPHA WITH MACRON */ }, \
+	{ 0x1FBB, /* GREEK CAPITAL LETTER ALPHA WITH OXIA */ \
+	  0x1F71, /* GREEK SMALL LETTER ALPHA WITH OXIA */ }, \
+	{ 0x1F08, /* GREEK CAPITAL LETTER ALPHA WITH PSILI */ \
+	  0x1F00, /* GREEK SMALL LETTER ALPHA WITH PSILI */ }, \
+	{ 0x1F0C, /* GREEK CAPITAL LETTER ALPHA WITH PSILI AND OXIA */ \
+	  0x1F04, /* GREEK SMALL LETTER ALPHA WITH PSILI AND OXIA */ }, \
+	{ 0x1F0E, /* GREEK CAPITAL LETTER ALPHA WITH PSILI AND PERISPOMENI */ \
+	  0x1F06, /* GREEK SMALL LETTER ALPHA WITH PSILI AND PERISPOMENI */ }, \
+	{ 0x1F0A, /* GREEK CAPITAL LETTER ALPHA WITH PSILI AND VARIA */ \
+	  0x1F02, /* GREEK SMALL LETTER ALPHA WITH PSILI AND VARIA */ }, \
+	{ 0x0386, /* GREEK CAPITAL LETTER ALPHA WITH TONOS */ \
+	  0x03AC, /* GREEK SMALL LETTER ALPHA WITH TONOS */ }, \
+	{ 0x1FBA, /* GREEK CAPITAL LETTER ALPHA WITH VARIA */ \
+	  0x1F70, /* GREEK SMALL LETTER ALPHA WITH VARIA */ }, \
+	{ 0x1FB8, /* GREEK CAPITAL LETTER ALPHA WITH VRACHY */ \
+	  0x1FB0, /* GREEK SMALL LETTER ALPHA WITH VRACHY */ }, \
+	{ 0x0372, /* GREEK CAPITAL LETTER ARCHAIC SAMPI */ \
+	  0x0373, /* GREEK SMALL LETTER ARCHAIC SAMPI */ }, \
+	{ 0x0392, /* GREEK CAPITAL LETTER BETA */ \
+	  0x03B2, /* GREEK SMALL LETTER BETA */ }, \
+	{ 0x03A7, /* GREEK CAPITAL LETTER CHI */ \
+	  0x03C7, /* GREEK SMALL LETTER CHI */ }, \
+	{ 0x0394, /* GREEK CAPITAL LETTER DELTA */ \
+	  0x03B4, /* GREEK SMALL LETTER DELTA */ }, \
+	{ 0x0395, /* GREEK CAPITAL LETTER EPSILON */ \
+	  0x03B5, /* GREEK SMALL LETTER EPSILON */ }, \
+	{ 0x1F19, /* GREEK CAPITAL LETTER EPSILON WITH DASIA */ \
+	  0x1F11, /* GREEK SMALL LETTER EPSILON WITH DASIA */ }, \
+	{ 0x1F1D, /* GREEK CAPITAL LETTER EPSILON WITH DASIA AND OXIA */ \
+	  0x1F15, /* GREEK SMALL LETTER EPSILON WITH DASIA AND OXIA */ }, \
+	{ 0x1F1B, /* GREEK CAPITAL LETTER EPSILON WITH DASIA AND VARIA */ \
+	  0x1F13, /* GREEK SMALL LETTER EPSILON WITH DASIA AND VARIA */ }, \
+	{ 0x1FC9, /* GREEK CAPITAL LETTER EPSILON WITH OXIA */ \
+	  0x1F73, /* GREEK SMALL LETTER EPSILON WITH OXIA */ }, \
+	{ 0x1F18, /* GREEK CAPITAL LETTER EPSILON WITH PSILI */ \
+	  0x1F10, /* GREEK SMALL LETTER EPSILON WITH PSILI */ }, \
+	{ 0x1F1C, /* GREEK CAPITAL LETTER EPSILON WITH PSILI AND OXIA */ \
+	  0x1F14, /* GREEK SMALL LETTER EPSILON WITH PSILI AND OXIA */ }, \
+	{ 0x1F1A, /* GREEK CAPITAL LETTER EPSILON WITH PSILI AND VARIA */ \
+	  0x1F12, /* GREEK SMALL LETTER EPSILON WITH PSILI AND VARIA */ }, \
+	{ 0x0388, /* GREEK CAPITAL LETTER EPSILON WITH TONOS */ \
+	  0x03AD, /* GREEK SMALL LETTER EPSILON WITH TONOS */ }, \
+	{ 0x1FC8, /* GREEK CAPITAL LETTER EPSILON WITH VARIA */ \
+	  0x1F72, /* GREEK SMALL LETTER EPSILON WITH VARIA */ }, \
+	{ 0x0397, /* GREEK CAPITAL LETTER ETA */ \
+	  0x03B7, /* GREEK SMALL LETTER ETA */ }, \
+	{ 0x1F29, /* GREEK CAPITAL LETTER ETA WITH DASIA */ \
+	  0x1F21, /* GREEK SMALL LETTER ETA WITH DASIA */ }, \
+	{ 0x1F2D, /* GREEK CAPITAL LETTER ETA WITH DASIA AND OXIA */ \
+	  0x1F25, /* GREEK SMALL LETTER ETA WITH DASIA AND OXIA */ }, \
+	{ 0x1F2F, /* GREEK CAPITAL LETTER ETA WITH DASIA AND PERISPOMENI */ \
+	  0x1F27, /* GREEK SMALL LETTER ETA WITH DASIA AND PERISPOMENI */ }, \
+	{ 0x1F2B, /* GREEK CAPITAL LETTER ETA WITH DASIA AND VARIA */ \
+	  0x1F23, /* GREEK SMALL LETTER ETA WITH DASIA AND VARIA */ }, \
+	{ 0x1FCB, /* GREEK CAPITAL LETTER ETA WITH OXIA */ \
+	  0x1F75, /* GREEK SMALL LETTER ETA WITH OXIA */ }, \
+	{ 0x1F28, /* GREEK CAPITAL LETTER ETA WITH PSILI */ \
+	  0x1F20, /* GREEK SMALL LETTER ETA WITH PSILI */ }, \
+	{ 0x1F2C, /* GREEK CAPITAL LETTER ETA WITH PSILI AND OXIA */ \
+	  0x1F24, /* GREEK SMALL LETTER ETA WITH PSILI AND OXIA */ }, \
+	{ 0x1F2E, /* GREEK CAPITAL LETTER ETA WITH PSILI AND PERISPOMENI */ \
+	  0x1F26, /* GREEK SMALL LETTER ETA WITH PSILI AND PERISPOMENI */ }, \
+	{ 0x1F2A, /* GREEK CAPITAL LETTER ETA WITH PSILI AND VARIA */ \
+	  0x1F22, /* GREEK SMALL LETTER ETA WITH PSILI AND VARIA */ }, \
+	{ 0x0389, /* GREEK CAPITAL LETTER ETA WITH TONOS */ \
+	  0x03AE, /* GREEK SMALL LETTER ETA WITH TONOS */ }, \
+	{ 0x1FCA, /* GREEK CAPITAL LETTER ETA WITH VARIA */ \
+	  0x1F74, /* GREEK SMALL LETTER ETA WITH VARIA */ }, \
+	{ 0x0393, /* GREEK CAPITAL LETTER GAMMA */ \
+	  0x03B3, /* GREEK SMALL LETTER GAMMA */ }, \
+	{ 0x0370, /* GREEK CAPITAL LETTER HETA */ \
+	  0x0371, /* GREEK SMALL LETTER HETA */ }, \
+	{ 0x0399, /* GREEK CAPITAL LETTER IOTA */ \
+	  0x03B9, /* GREEK SMALL LETTER IOTA */ }, \
+	{ 0x1F39, /* GREEK CAPITAL LETTER IOTA WITH DASIA */ \
+	  0x1F31, /* GREEK SMALL LETTER IOTA WITH DASIA */ }, \
+	{ 0x1F3D, /* GREEK CAPITAL LETTER IOTA WITH DASIA AND OXIA */ \
+	  0x1F35, /* GREEK SMALL LETTER IOTA WITH DASIA AND OXIA */ }, \
+	{ 0x1F3F, /* GREEK CAPITAL LETTER IOTA WITH DASIA AND PERISPOMENI */ \
+	  0x1F37, /* GREEK SMALL LETTER IOTA WITH DASIA AND PERISPOMENI */ }, \
+	{ 0x1F3B, /* GREEK CAPITAL LETTER IOTA WITH DASIA AND VARIA */ \
+	  0x1F33, /* GREEK SMALL LETTER IOTA WITH DASIA AND VARIA */ }, \
+	{ 0x03AA, /* GREEK CAPITAL LETTER IOTA WITH DIALYTIKA */ \
+	  0x03CA, /* GREEK SMALL LETTER IOTA WITH DIALYTIKA */ }, \
+	{ 0x1FD9, /* GREEK CAPITAL LETTER IOTA WITH MACRON */ \
+	  0x1FD1, /* GREEK SMALL LETTER IOTA WITH MACRON */ }, \
+	{ 0x1FDB, /* GREEK CAPITAL LETTER IOTA WITH OXIA */ \
+	  0x1F77, /* GREEK SMALL LETTER IOTA WITH OXIA */ }, \
+	{ 0x1F38, /* GREEK CAPITAL LETTER IOTA WITH PSILI */ \
+	  0x1F30, /* GREEK SMALL LETTER IOTA WITH PSILI */ }, \
+	{ 0x1F3C, /* GREEK CAPITAL LETTER IOTA WITH PSILI AND OXIA */ \
+	  0x1F34, /* GREEK SMALL LETTER IOTA WITH PSILI AND OXIA */ }, \
+	{ 0x1F3E, /* GREEK CAPITAL LETTER IOTA WITH PSILI AND PERISPOMENI */ \
+	  0x1F36, /* GREEK SMALL LETTER IOTA WITH PSILI AND PERISPOMENI */ }, \
+	{ 0x1F3A, /* GREEK CAPITAL LETTER IOTA WITH PSILI AND VARIA */ \
+	  0x1F32, /* GREEK SMALL LETTER IOTA WITH PSILI AND VARIA */ }, \
+	{ 0x038A, /* GREEK CAPITAL LETTER IOTA WITH TONOS */ \
+	  0x03AF, /* GREEK SMALL LETTER IOTA WITH TONOS */ }, \
+	{ 0x1FDA, /* GREEK CAPITAL LETTER IOTA WITH VARIA */ \
+	  0x1F76, /* GREEK SMALL LETTER IOTA WITH VARIA */ }, \
+	{ 0x1FD8, /* GREEK CAPITAL LETTER IOTA WITH VRACHY */ \
+	  0x1FD0, /* GREEK SMALL LETTER IOTA WITH VRACHY */ }, \
+	{ 0x039A, /* GREEK CAPITAL LETTER KAPPA */ \
+	  0x03BA, /* GREEK SMALL LETTER KAPPA */ }, \
+	{ 0x039B, /* GREEK CAPITAL LETTER LAMDA */ \
+	  0x03BB, /* GREEK SMALL LETTER LAMDA */ }, \
+	{ 0x039C, /* GREEK CAPITAL LETTER MU */ \
+	  0x03BC, /* GREEK SMALL LETTER MU */ }, \
+	{ 0x039D, /* GREEK CAPITAL LETTER NU */ \
+	  0x03BD, /* GREEK SMALL LETTER NU */ }, \
+	{ 0x03A9, /* GREEK CAPITAL LETTER OMEGA */ \
+	  0x03C9, /* GREEK SMALL LETTER OMEGA */ }, \
+	{ 0x1F69, /* GREEK CAPITAL LETTER OMEGA WITH DASIA */ \
+	  0x1F61, /* GREEK SMALL LETTER OMEGA WITH DASIA */ }, \
+	{ 0x1F6D, /* GREEK CAPITAL LETTER OMEGA WITH DASIA AND OXIA */ \
+	  0x1F65, /* GREEK SMALL LETTER OMEGA WITH DASIA AND OXIA */ }, \
+	{ 0x1F6F, /* GREEK CAPITAL LETTER OMEGA WITH DASIA AND PERISPOMENI */ \
+	  0x1F67, /* GREEK SMALL LETTER OMEGA WITH DASIA AND PERISPOMENI */ }, \
+	{ 0x1F6B, /* GREEK CAPITAL LETTER OMEGA WITH DASIA AND VARIA */ \
+	  0x1F63, /* GREEK SMALL LETTER OMEGA WITH DASIA AND VARIA */ }, \
+	{ 0x1FFB, /* GREEK CAPITAL LETTER OMEGA WITH OXIA */ \
+	  0x1F7D, /* GREEK SMALL LETTER OMEGA WITH OXIA */ }, \
+	{ 0x1F68, /* GREEK CAPITAL LETTER OMEGA WITH PSILI */ \
+	  0x1F60, /* GREEK SMALL LETTER OMEGA WITH PSILI */ }, \
+	{ 0x1F6C, /* GREEK CAPITAL LETTER OMEGA WITH PSILI AND OXIA */ \
+	  0x1F64, /* GREEK SMALL LETTER OMEGA WITH PSILI AND OXIA */ }, \
+	{ 0x1F6E, /* GREEK CAPITAL LETTER OMEGA WITH PSILI AND PERISPOMENI */ \
+	  0x1F66, /* GREEK SMALL LETTER OMEGA WITH PSILI AND PERISPOMENI */ }, \
+	{ 0x1F6A, /* GREEK CAPITAL LETTER OMEGA WITH PSILI AND VARIA */ \
+	  0x1F62, /* GREEK SMALL LETTER OMEGA WITH PSILI AND VARIA */ }, \
+	{ 0x038F, /* GREEK CAPITAL LETTER OMEGA WITH TONOS */ \
+	  0x03CE, /* GREEK SMALL LETTER OMEGA WITH TONOS */ }, \
+	{ 0x1FFA, /* GREEK CAPITAL LETTER OMEGA WITH VARIA */ \
+	  0x1F7C, /* GREEK SMALL LETTER OMEGA WITH VARIA */ }, \
+	{ 0x039F, /* GREEK CAPITAL LETTER OMICRON */ \
+	  0x03BF, /* GREEK SMALL LETTER OMICRON */ }, \
+	{ 0x1F49, /* GREEK CAPITAL LETTER OMICRON WITH DASIA */ \
+	  0x1F41, /* GREEK SMALL LETTER OMICRON WITH DASIA */ }, \
+	{ 0x1F4D, /* GREEK CAPITAL LETTER OMICRON WITH DASIA AND OXIA */ \
+	  0x1F45, /* GREEK SMALL LETTER OMICRON WITH DASIA AND OXIA */ }, \
+	{ 0x1F4B, /* GREEK CAPITAL LETTER OMICRON WITH DASIA AND VARIA */ \
+	  0x1F43, /* GREEK SMALL LETTER OMICRON WITH DASIA AND VARIA */ }, \
+	{ 0x1FF9, /* GREEK CAPITAL LETTER OMICRON WITH OXIA */ \
+	  0x1F79, /* GREEK SMALL LETTER OMICRON WITH OXIA */ }, \
+	{ 0x1F48, /* GREEK CAPITAL LETTER OMICRON WITH PSILI */ \
+	  0x1F40, /* GREEK SMALL LETTER OMICRON WITH PSILI */ }, \
+	{ 0x1F4C, /* GREEK CAPITAL LETTER OMICRON WITH PSILI AND OXIA */ \
+	  0x1F44, /* GREEK SMALL LETTER OMICRON WITH PSILI AND OXIA */ }, \
+	{ 0x1F4A, /* GREEK CAPITAL LETTER OMICRON WITH PSILI AND VARIA */ \
+	  0x1F42, /* GREEK SMALL LETTER OMICRON WITH PSILI AND VARIA */ }, \
+	{ 0x038C, /* GREEK CAPITAL LETTER OMICRON WITH TONOS */ \
+	  0x03CC, /* GREEK SMALL LETTER OMICRON WITH TONOS */ }, \
+	{ 0x1FF8, /* GREEK CAPITAL LETTER OMICRON WITH VARIA */ \
+	  0x1F78, /* GREEK SMALL LETTER OMICRON WITH VARIA */ }, \
+	{ 0x0376, /* GREEK CAPITAL LETTER PAMPHYLIAN DIGAMMA */ \
+	  0x0377, /* GREEK SMALL LETTER PAMPHYLIAN DIGAMMA */ }, \
+	{ 0x03A6, /* GREEK CAPITAL LETTER PHI */ \
+	  0x03C6, /* GREEK SMALL LETTER PHI */ }, \
+	{ 0x03A0, /* GREEK CAPITAL LETTER PI */ \
+	  0x03C0, /* GREEK SMALL LETTER PI */ }, \
+	{ 0x03A8, /* GREEK CAPITAL LETTER PSI */ \
+	  0x03C8, /* GREEK SMALL LETTER PSI */ }, \
+	{ 0x03A1, /* GREEK CAPITAL LETTER RHO */ \
+	  0x03C1, /* GREEK SMALL LETTER RHO */ }, \
+	{ 0x1FEC, /* GREEK CAPITAL LETTER RHO WITH DASIA */ \
+	  0x1FE5, /* GREEK SMALL LETTER RHO WITH DASIA */ }, \
+	{ 0x03FA, /* GREEK CAPITAL LETTER SAN */ \
+	  0x03FB, /* GREEK SMALL LETTER SAN */ }, \
+	{ 0x03F7, /* GREEK CAPITAL LETTER SHO */ \
+	  0x03F8, /* GREEK SMALL LETTER SHO */ }, \
+	{ 0x03A3, /* GREEK CAPITAL LETTER SIGMA */ \
+	  0x03C3, /* GREEK SMALL LETTER SIGMA */ }, \
+	{ 0x03A4, /* GREEK CAPITAL LETTER TAU */ \
+	  0x03C4, /* GREEK SMALL LETTER TAU */ }, \
+	{ 0x0398, /* GREEK CAPITAL LETTER THETA */ \
+	  0x03B8, /* GREEK SMALL LETTER THETA */ }, \
+	{ 0x03A5, /* GREEK CAPITAL LETTER UPSILON */ \
+	  0x03C5, /* GREEK SMALL LETTER UPSILON */ }, \
+	{ 0x1F59, /* GREEK CAPITAL LETTER UPSILON WITH DASIA */ \
+	  0x1F51, /* GREEK SMALL LETTER UPSILON WITH DASIA */ }, \
+	{ 0x1F5D, /* GREEK CAPITAL LETTER UPSILON WITH DASIA AND OXIA */ \
+	  0x1F55, /* GREEK SMALL LETTER UPSILON WITH DASIA AND OXIA */ }, \
+	{ 0x1F5F, /* GREEK CAPITAL LETTER UPSILON WITH DASIA AND PERISPOMENI */ \
+	  0x1F57, /* GREEK SMALL LETTER UPSILON WITH DASIA AND PERISPOMENI */ }, \
+	{ 0x1F5B, /* GREEK CAPITAL LETTER UPSILON WITH DASIA AND VARIA */ \
+	  0x1F53, /* GREEK SMALL LETTER UPSILON WITH DASIA AND VARIA */ }, \
+	{ 0x03AB, /* GREEK CAPITAL LETTER UPSILON WITH DIALYTIKA */ \
+	  0x03CB, /* GREEK SMALL LETTER UPSILON WITH DIALYTIKA */ }, \
+	{ 0x1FE9, /* GREEK CAPITAL LETTER UPSILON WITH MACRON */ \
+	  0x1FE1, /* GREEK SMALL LETTER UPSILON WITH MACRON */ }, \
+	{ 0x1FEB, /* GREEK CAPITAL LETTER UPSILON WITH OXIA */ \
+	  0x1F7B, /* GREEK SMALL LETTER UPSILON WITH OXIA */ }, \
+	{ 0x038E, /* GREEK CAPITAL LETTER UPSILON WITH TONOS */ \
+	  0x03CD, /* GREEK SMALL LETTER UPSILON WITH TONOS */ }, \
+	{ 0x1FEA, /* GREEK CAPITAL LETTER UPSILON WITH VARIA */ \
+	  0x1F7A, /* GREEK SMALL LETTER UPSILON WITH VARIA */ }, \
+	{ 0x1FE8, /* GREEK CAPITAL LETTER UPSILON WITH VRACHY */ \
+	  0x1FE0, /* GREEK SMALL LETTER UPSILON WITH VRACHY */ }, \
+	{ 0x039E, /* GREEK CAPITAL LETTER XI */ \
+	  0x03BE, /* GREEK SMALL LETTER XI */ }, \
+	{ 0x0396, /* GREEK CAPITAL LETTER ZETA */ \
+	  0x03B6, /* GREEK SMALL LETTER ZETA */ }, \
+	{ 0x0041, /* LATIN CAPITAL LETTER A */ \
+	  0x0061, /* LATIN SMALL LETTER A */ }, \
+	{ 0x00C1, /* LATIN CAPITAL LETTER A WITH ACUTE */ \
+	  0x00E1, /* LATIN SMALL LETTER A WITH ACUTE */ }, \
+	{ 0x0102, /* LATIN CAPITAL LETTER A WITH BREVE */ \
+	  0x0103, /* LATIN SMALL LETTER A WITH BREVE */ }, \
+	{ 0x1EAE, /* LATIN CAPITAL LETTER A WITH BREVE AND ACUTE */ \
+	  0x1EAF, /* LATIN SMALL LETTER A WITH BREVE AND ACUTE */ }, \
+	{ 0x1EB6, /* LATIN CAPITAL LETTER A WITH BREVE AND DOT BELOW */ \
+	  0x1EB7, /* LATIN SMALL LETTER A WITH BREVE AND DOT BELOW */ }, \
+	{ 0x1EB0, /* LATIN CAPITAL LETTER A WITH BREVE AND GRAVE */ \
+	  0x1EB1, /* LATIN SMALL LETTER A WITH BREVE AND GRAVE */ }, \
+	{ 0x1EB2, /* LATIN CAPITAL LETTER A WITH BREVE AND HOOK ABOVE */ \
+	  0x1EB3, /* LATIN SMALL LETTER A WITH BREVE AND HOOK ABOVE */ }, \
+	{ 0x1EB4, /* LATIN CAPITAL LETTER A WITH BREVE AND TILDE */ \
+	  0x1EB5, /* LATIN SMALL LETTER A WITH BREVE AND TILDE */ }, \
+	{ 0x01CD, /* LATIN CAPITAL LETTER A WITH CARON */ \
+	  0x01CE, /* LATIN SMALL LETTER A WITH CARON */ }, \
+	{ 0x00C2, /* LATIN CAPITAL LETTER A WITH CIRCUMFLEX */ \
+	  0x00E2, /* LATIN SMALL LETTER A WITH CIRCUMFLEX */ }, \
+	{ 0x1EA4, /* LATIN CAPITAL LETTER A WITH CIRCUMFLEX AND ACUTE */ \
+	  0x1EA5, /* LATIN SMALL LETTER A WITH CIRCUMFLEX AND ACUTE */ }, \
+	{ 0x1EAC, /* LATIN CAPITAL LETTER A WITH CIRCUMFLEX AND DOT BELOW */ \
+	  0x1EAD, /* LATIN SMALL LETTER A WITH CIRCUMFLEX AND DOT BELOW */ }, \
+	{ 0x1EA6, /* LATIN CAPITAL LETTER A WITH CIRCUMFLEX AND GRAVE */ \
+	  0x1EA7, /* LATIN SMALL LETTER A WITH CIRCUMFLEX AND GRAVE */ }, \
+	{ 0x1EA8, /* LATIN CAPITAL LETTER A WITH CIRCUMFLEX AND HOOK ABOVE */ \
+	  0x1EA9, /* LATIN SMALL LETTER A WITH CIRCUMFLEX AND HOOK ABOVE */ }, \
+	{ 0x1EAA, /* LATIN CAPITAL LETTER A WITH CIRCUMFLEX AND TILDE */ \
+	  0x1EAB, /* LATIN SMALL LETTER A WITH CIRCUMFLEX AND TILDE */ }, \
+	{ 0x00C4, /* LATIN CAPITAL LETTER A WITH DIAERESIS */ \
+	  0x00E4, /* LATIN SMALL LETTER A WITH DIAERESIS */ }, \
+	{ 0x01DE, /* LATIN CAPITAL LETTER A WITH DIAERESIS AND MACRON */ \
+	  0x01DF, /* LATIN SMALL LETTER A WITH DIAERESIS AND MACRON */ }, \
+	{ 0x0226, /* LATIN CAPITAL LETTER A WITH DOT ABOVE */ \
+	  0x0227, /* LATIN SMALL LETTER A WITH DOT ABOVE */ }, \
+	{ 0x01E0, /* LATIN CAPITAL LETTER A WITH DOT ABOVE AND MACRON */ \
+	  0x01E1, /* LATIN SMALL LETTER A WITH DOT ABOVE AND MACRON */ }, \
+	{ 0x1EA0, /* LATIN CAPITAL LETTER A WITH DOT BELOW */ \
+	  0x1EA1, /* LATIN SMALL LETTER A WITH DOT BELOW */ }, \
+	{ 0x0200, /* LATIN CAPITAL LETTER A WITH DOUBLE GRAVE */ \
+	  0x0201, /* LATIN SMALL LETTER A WITH DOUBLE GRAVE */ }, \
+	{ 0x00C0, /* LATIN CAPITAL LETTER A WITH GRAVE */ \
+	  0x00E0, /* LATIN SMALL LETTER A WITH GRAVE */ }, \
+	{ 0x1EA2, /* LATIN CAPITAL LETTER A WITH HOOK ABOVE */ \
+	  0x1EA3, /* LATIN SMALL LETTER A WITH HOOK ABOVE */ }, \
+	{ 0x0202, /* LATIN CAPITAL LETTER A WITH INVERTED BREVE */ \
+	  0x0203, /* LATIN SMALL LETTER A WITH INVERTED BREVE */ }, \
+	{ 0x0100, /* LATIN CAPITAL LETTER A WITH MACRON */ \
+	  0x0101, /* LATIN SMALL LETTER A WITH MACRON */ }, \
+	{ 0x0104, /* LATIN CAPITAL LETTER A WITH OGONEK */ \
+	  0x0105, /* LATIN SMALL LETTER A WITH OGONEK */ }, \
+	{ 0x00C5, /* LATIN CAPITAL LETTER A WITH RING ABOVE */ \
+	  0x00E5, /* LATIN SMALL LETTER A WITH RING ABOVE */ }, \
+	{ 0x01FA, /* LATIN CAPITAL LETTER A WITH RING ABOVE AND ACUTE */ \
+	  0x01FB, /* LATIN SMALL LETTER A WITH RING ABOVE AND ACUTE */ }, \
+	{ 0x1E00, /* LATIN CAPITAL LETTER A WITH RING BELOW */ \
+	  0x1E01, /* LATIN SMALL LETTER A WITH RING BELOW */ }, \
+	{ 0x023A, /* LATIN CAPITAL LETTER A WITH STROKE */ \
+	  0x2C65, /* LATIN SMALL LETTER A WITH STROKE */ }, \
+	{ 0x00C3, /* LATIN CAPITAL LETTER A WITH TILDE */ \
+	  0x00E3, /* LATIN SMALL LETTER A WITH TILDE */ }, \
+	{ 0xA732, /* LATIN CAPITAL LETTER AA */ \
+	  0xA733, /* LATIN SMALL LETTER AA */ }, \
+	{ 0x00C6, /* LATIN CAPITAL LETTER AE */ \
+	  0x00E6, /* LATIN SMALL LETTER AE */ }, \
+	{ 0x01FC, /* LATIN CAPITAL LETTER AE WITH ACUTE */ \
+	  0x01FD, /* LATIN SMALL LETTER AE WITH ACUTE */ }, \
+	{ 0x01E2, /* LATIN CAPITAL LETTER AE WITH MACRON */ \
+	  0x01E3, /* LATIN SMALL LETTER AE WITH MACRON */ }, \
+	{ 0x2C6D, /* LATIN CAPITAL LETTER ALPHA */ \
+	  0x0251, /* LATIN SMALL LETTER ALPHA */ }, \
+	{ 0xA734, /* LATIN CAPITAL LETTER AO */ \
+	  0xA735, /* LATIN SMALL LETTER AO */ }, \
+	{ 0xA736, /* LATIN CAPITAL LETTER AU */ \
+	  0xA737, /* LATIN SMALL LETTER AU */ }, \
+	{ 0xA738, /* LATIN CAPITAL LETTER AV */ \
+	  0xA739, /* LATIN SMALL LETTER AV */ }, \
+	{ 0xA73A, /* LATIN CAPITAL LETTER AV WITH HORIZONTAL BAR */ \
+	  0xA73B, /* LATIN SMALL LETTER AV WITH HORIZONTAL BAR */ }, \
+	{ 0xA73C, /* LATIN CAPITAL LETTER AY */ \
+	  0xA73D, /* LATIN SMALL LETTER AY */ }, \
+	{ 0x0042, /* LATIN CAPITAL LETTER B */ \
+	  0x0062, /* LATIN SMALL LETTER B */ }, \
+	{ 0x1E02, /* LATIN CAPITAL LETTER B WITH DOT ABOVE */ \
+	  0x1E03, /* LATIN SMALL LETTER B WITH DOT ABOVE */ }, \
+	{ 0x1E04, /* LATIN CAPITAL LETTER B WITH DOT BELOW */ \
+	  0x1E05, /* LATIN SMALL LETTER B WITH DOT BELOW */ }, \
+	{ 0xA796, /* LATIN CAPITAL LETTER B WITH FLOURISH */ \
+	  0xA797, /* LATIN SMALL LETTER B WITH FLOURISH */ }, \
+	{ 0x0181, /* LATIN CAPITAL LETTER B WITH HOOK */ \
+	  0x0253, /* LATIN SMALL LETTER B WITH HOOK */ }, \
+	{ 0x1E06, /* LATIN CAPITAL LETTER B WITH LINE BELOW */ \
+	  0x1E07, /* LATIN SMALL LETTER B WITH LINE BELOW */ }, \
+	{ 0x0243, /* LATIN CAPITAL LETTER B WITH STROKE */ \
+	  0x0180, /* LATIN SMALL LETTER B WITH STROKE */ }, \
+	{ 0x0182, /* LATIN CAPITAL LETTER B WITH TOPBAR */ \
+	  0x0183, /* LATIN SMALL LETTER B WITH TOPBAR */ }, \
+	{ 0xA7B4, /* LATIN CAPITAL LETTER BETA */ \
+	  0xA7B5, /* LATIN SMALL LETTER BETA */ }, \
+	{ 0xA746, /* LATIN CAPITAL LETTER BROKEN L */ \
+	  0xA747, /* LATIN SMALL LETTER BROKEN L */ }, \
+	{ 0x0043, /* LATIN CAPITAL LETTER C */ \
+	  0x0063, /* LATIN SMALL LETTER C */ }, \
+	{ 0x0106, /* LATIN CAPITAL LETTER C WITH ACUTE */ \
+	  0x0107, /* LATIN SMALL LETTER C WITH ACUTE */ }, \
+	{ 0xA792, /* LATIN CAPITAL LETTER C WITH BAR */ \
+	  0xA793, /* LATIN SMALL LETTER C WITH BAR */ }, \
+	{ 0x010C, /* LATIN CAPITAL LETTER C WITH CARON */ \
+	  0x010D, /* LATIN SMALL LETTER C WITH CARON */ }, \
+	{ 0x00C7, /* LATIN CAPITAL LETTER C WITH CEDILLA */ \
+	  0x00E7, /* LATIN SMALL LETTER C WITH CEDILLA */ }, \
+	{ 0x1E08, /* LATIN CAPITAL LETTER C WITH CEDILLA AND ACUTE */ \
+	  0x1E09, /* LATIN SMALL LETTER C WITH CEDILLA AND ACUTE */ }, \
+	{ 0x0108, /* LATIN CAPITAL LETTER C WITH CIRCUMFLEX */ \
+	  0x0109, /* LATIN SMALL LETTER C WITH CIRCUMFLEX */ }, \
+	{ 0x010A, /* LATIN CAPITAL LETTER C WITH DOT ABOVE */ \
+	  0x010B, /* LATIN SMALL LETTER C WITH DOT ABOVE */ }, \
+	{ 0x0187, /* LATIN CAPITAL LETTER C WITH HOOK */ \
+	  0x0188, /* LATIN SMALL LETTER C WITH HOOK */ }, \
+	{ 0x023B, /* LATIN CAPITAL LETTER C WITH STROKE */ \
+	  0x023C, /* LATIN SMALL LETTER C WITH STROKE */ }, \
+	{ 0xA7B3, /* LATIN CAPITAL LETTER CHI */ \
+	  0xAB53, /* LATIN SMALL LETTER CHI */ }, \
+	{ 0xA76E, /* LATIN CAPITAL LETTER CON */ \
+	  0xA76F, /* LATIN SMALL LETTER CON */ }, \
+	{ 0xA72C, /* LATIN CAPITAL LETTER CUATRILLO */ \
+	  0xA72D, /* LATIN SMALL LETTER CUATRILLO */ }, \
+	{ 0xA72E, /* LATIN CAPITAL LETTER CUATRILLO WITH COMMA */ \
+	  0xA72F, /* LATIN SMALL LETTER CUATRILLO WITH COMMA */ }, \
+	{ 0x0044, /* LATIN CAPITAL LETTER D */ \
+	  0x0064, /* LATIN SMALL LETTER D */ }, \
+	{ 0x010E, /* LATIN CAPITAL LETTER D WITH CARON */ \
+	  0x010F, /* LATIN SMALL LETTER D WITH CARON */ }, \
+	{ 0x1E10, /* LATIN CAPITAL LETTER D WITH CEDILLA */ \
+	  0x1E11, /* LATIN SMALL LETTER D WITH CEDILLA */ }, \
+	{ 0x1E12, /* LATIN CAPITAL LETTER D WITH CIRCUMFLEX BELOW */ \
+	  0x1E13, /* LATIN SMALL LETTER D WITH CIRCUMFLEX BELOW */ }, \
+	{ 0x1E0A, /* LATIN CAPITAL LETTER D WITH DOT ABOVE */ \
+	  0x1E0B, /* LATIN SMALL LETTER D WITH DOT ABOVE */ }, \
+	{ 0x1E0C, /* LATIN CAPITAL LETTER D WITH DOT BELOW */ \
+	  0x1E0D, /* LATIN SMALL LETTER D WITH DOT BELOW */ }, \
+	{ 0x018A, /* LATIN CAPITAL LETTER D WITH HOOK */ \
+	  0x0257, /* LATIN SMALL LETTER D WITH HOOK */ }, \
+	{ 0x1E0E, /* LATIN CAPITAL LETTER D WITH LINE BELOW */ \
+	  0x1E0F, /* LATIN SMALL LETTER D WITH LINE BELOW */ }, \
+	{ 0x0110, /* LATIN CAPITAL LETTER D WITH STROKE */ \
+	  0x0111, /* LATIN SMALL LETTER D WITH STROKE */ }, \
+	{ 0x018B, /* LATIN CAPITAL LETTER D WITH TOPBAR */ \
+	  0x018C, /* LATIN SMALL LETTER D WITH TOPBAR */ }, \
+	{ 0x01F1, /* LATIN CAPITAL LETTER DZ */ \
+	  0x01F3, /* LATIN SMALL LETTER DZ */ }, \
+	{ 0x01C4, /* LATIN CAPITAL LETTER DZ WITH CARON */ \
+	  0x01C6, /* LATIN SMALL LETTER DZ WITH CARON */ }, \
+	{ 0x0045, /* LATIN CAPITAL LETTER E */ \
+	  0x0065, /* LATIN SMALL LETTER E */ }, \
+	{ 0x00C9, /* LATIN CAPITAL LETTER E WITH ACUTE */ \
+	  0x00E9, /* LATIN SMALL LETTER E WITH ACUTE */ }, \
+	{ 0x0114, /* LATIN CAPITAL LETTER E WITH BREVE */ \
+	  0x0115, /* LATIN SMALL LETTER E WITH BREVE */ }, \
+	{ 0x011A, /* LATIN CAPITAL LETTER E WITH CARON */ \
+	  0x011B, /* LATIN SMALL LETTER E WITH CARON */ }, \
+	{ 0x0228, /* LATIN CAPITAL LETTER E WITH CEDILLA */ \
+	  0x0229, /* LATIN SMALL LETTER E WITH CEDILLA */ }, \
+	{ 0x1E1C, /* LATIN CAPITAL LETTER E WITH CEDILLA AND BREVE */ \
+	  0x1E1D, /* LATIN SMALL LETTER E WITH CEDILLA AND BREVE */ }, \
+	{ 0x00CA, /* LATIN CAPITAL LETTER E WITH CIRCUMFLEX */ \
+	  0x00EA, /* LATIN SMALL LETTER E WITH CIRCUMFLEX */ }, \
+	{ 0x1EBE, /* LATIN CAPITAL LETTER E WITH CIRCUMFLEX AND ACUTE */ \
+	  0x1EBF, /* LATIN SMALL LETTER E WITH CIRCUMFLEX AND ACUTE */ }, \
+	{ 0x1EC6, /* LATIN CAPITAL LETTER E WITH CIRCUMFLEX AND DOT BELOW */ \
+	  0x1EC7, /* LATIN SMALL LETTER E WITH CIRCUMFLEX AND DOT BELOW */ }, \
+	{ 0x1EC0, /* LATIN CAPITAL LETTER E WITH CIRCUMFLEX AND GRAVE */ \
+	  0x1EC1, /* LATIN SMALL LETTER E WITH CIRCUMFLEX AND GRAVE */ }, \
+	{ 0x1EC2, /* LATIN CAPITAL LETTER E WITH CIRCUMFLEX AND HOOK ABOVE */ \
+	  0x1EC3, /* LATIN SMALL LETTER E WITH CIRCUMFLEX AND HOOK ABOVE */ }, \
+	{ 0x1EC4, /* LATIN CAPITAL LETTER E WITH CIRCUMFLEX AND TILDE */ \
+	  0x1EC5, /* LATIN SMALL LETTER E WITH CIRCUMFLEX AND TILDE */ }, \
+	{ 0x1E18, /* LATIN CAPITAL LETTER E WITH CIRCUMFLEX BELOW */ \
+	  0x1E19, /* LATIN SMALL LETTER E WITH CIRCUMFLEX BELOW */ }, \
+	{ 0x00CB, /* LATIN CAPITAL LETTER E WITH DIAERESIS */ \
+	  0x00EB, /* LATIN SMALL LETTER E WITH DIAERESIS */ }, \
+	{ 0x0116, /* LATIN CAPITAL LETTER E WITH DOT ABOVE */ \
+	  0x0117, /* LATIN SMALL LETTER E WITH DOT ABOVE */ }, \
+	{ 0x1EB8, /* LATIN CAPITAL LETTER E WITH DOT BELOW */ \
+	  0x1EB9, /* LATIN SMALL LETTER E WITH DOT BELOW */ }, \
+	{ 0x0204, /* LATIN CAPITAL LETTER E WITH DOUBLE GRAVE */ \
+	  0x0205, /* LATIN SMALL LETTER E WITH DOUBLE GRAVE */ }, \
+	{ 0x00C8, /* LATIN CAPITAL LETTER E WITH GRAVE */ \
+	  0x00E8, /* LATIN SMALL LETTER E WITH GRAVE */ }, \
+	{ 0x1EBA, /* LATIN CAPITAL LETTER E WITH HOOK ABOVE */ \
+	  0x1EBB, /* LATIN SMALL LETTER E WITH HOOK ABOVE */ }, \
+	{ 0x0206, /* LATIN CAPITAL LETTER E WITH INVERTED BREVE */ \
+	  0x0207, /* LATIN SMALL LETTER E WITH INVERTED BREVE */ }, \
+	{ 0x0112, /* LATIN CAPITAL LETTER E WITH MACRON */ \
+	  0x0113, /* LATIN SMALL LETTER E WITH MACRON */ }, \
+	{ 0x1E16, /* LATIN CAPITAL LETTER E WITH MACRON AND ACUTE */ \
+	  0x1E17, /* LATIN SMALL LETTER E WITH MACRON AND ACUTE */ }, \
+	{ 0x1E14, /* LATIN CAPITAL LETTER E WITH MACRON AND GRAVE */ \
+	  0x1E15, /* LATIN SMALL LETTER E WITH MACRON AND GRAVE */ }, \
+	{ 0x0118, /* LATIN CAPITAL LETTER E WITH OGONEK */ \
+	  0x0119, /* LATIN SMALL LETTER E WITH OGONEK */ }, \
+	{ 0x0246, /* LATIN CAPITAL LETTER E WITH STROKE */ \
+	  0x0247, /* LATIN SMALL LETTER E WITH STROKE */ }, \
+	{ 0x1EBC, /* LATIN CAPITAL LETTER E WITH TILDE */ \
+	  0x1EBD, /* LATIN SMALL LETTER E WITH TILDE */ }, \
+	{ 0x1E1A, /* LATIN CAPITAL LETTER E WITH TILDE BELOW */ \
+	  0x1E1B, /* LATIN SMALL LETTER E WITH TILDE BELOW */ }, \
+	{ 0xA724, /* LATIN CAPITAL LETTER EGYPTOLOGICAL AIN */ \
+	  0xA725, /* LATIN SMALL LETTER EGYPTOLOGICAL AIN */ }, \
+	{ 0xA722, /* LATIN CAPITAL LETTER EGYPTOLOGICAL ALEF */ \
+	  0xA723, /* LATIN SMALL LETTER EGYPTOLOGICAL ALEF */ }, \
+	{ 0x014A, /* LATIN CAPITAL LETTER ENG */ \
+	  0x014B, /* LATIN SMALL LETTER ENG */ }, \
+	{ 0x01A9, /* LATIN CAPITAL LETTER ESH */ \
+	  0x0283, /* LATIN SMALL LETTER ESH */ }, \
+	{ 0xA76A, /* LATIN CAPITAL LETTER ET */ \
+	  0xA76B, /* LATIN SMALL LETTER ET */ }, \
+	{ 0x00D0, /* LATIN CAPITAL LETTER ETH */ \
+	  0x00F0, /* LATIN SMALL LETTER ETH */ }, \
+	{ 0x01B7, /* LATIN CAPITAL LETTER EZH */ \
+	  0x0292, /* LATIN SMALL LETTER EZH */ }, \
+	{ 0x01B8, /* LATIN CAPITAL LETTER EZH REVERSED */ \
+	  0x01B9, /* LATIN SMALL LETTER EZH REVERSED */ }, \
+	{ 0x01EE, /* LATIN CAPITAL LETTER EZH WITH CARON */ \
+	  0x01EF, /* LATIN SMALL LETTER EZH WITH CARON */ }, \
+	{ 0x0046, /* LATIN CAPITAL LETTER F */ \
+	  0x0066, /* LATIN SMALL LETTER F */ }, \
+	{ 0x1E1E, /* LATIN CAPITAL LETTER F WITH DOT ABOVE */ \
+	  0x1E1F, /* LATIN SMALL LETTER F WITH DOT ABOVE */ }, \
+	{ 0x0191, /* LATIN CAPITAL LETTER F WITH HOOK */ \
+	  0x0192, /* LATIN SMALL LETTER F WITH HOOK */ }, \
+	{ 0xA798, /* LATIN CAPITAL LETTER F WITH STROKE */ \
+	  0xA799, /* LATIN SMALL LETTER F WITH STROKE */ }, \
+	{ 0x0047, /* LATIN CAPITAL LETTER G */ \
+	  0x0067, /* LATIN SMALL LETTER G */ }, \
+	{ 0x01F4, /* LATIN CAPITAL LETTER G WITH ACUTE */ \
+	  0x01F5, /* LATIN SMALL LETTER G WITH ACUTE */ }, \
+	{ 0x011E, /* LATIN CAPITAL LETTER G WITH BREVE */ \
+	  0x011F, /* LATIN SMALL LETTER G WITH BREVE */ }, \
+	{ 0x01E6, /* LATIN CAPITAL LETTER G WITH CARON */ \
+	  0x01E7, /* LATIN SMALL LETTER G WITH CARON */ }, \
+	{ 0x0122, /* LATIN CAPITAL LETTER G WITH CEDILLA */ \
+	  0x0123, /* LATIN SMALL LETTER G WITH CEDILLA */ }, \
+	{ 0x011C, /* LATIN CAPITAL LETTER G WITH CIRCUMFLEX */ \
+	  0x011D, /* LATIN SMALL LETTER G WITH CIRCUMFLEX */ }, \
+	{ 0x0120, /* LATIN CAPITAL LETTER G WITH DOT ABOVE */ \
+	  0x0121, /* LATIN SMALL LETTER G WITH DOT ABOVE */ }, \
+	{ 0x0193, /* LATIN CAPITAL LETTER G WITH HOOK */ \
+	  0x0260, /* LATIN SMALL LETTER G WITH HOOK */ }, \
+	{ 0x1E20, /* LATIN CAPITAL LETTER G WITH MACRON */ \
+	  0x1E21, /* LATIN SMALL LETTER G WITH MACRON */ }, \
+	{ 0xA7A0, /* LATIN CAPITAL LETTER G WITH OBLIQUE STROKE */ \
+	  0xA7A1, /* LATIN SMALL LETTER G WITH OBLIQUE STROKE */ }, \
+	{ 0x01E4, /* LATIN CAPITAL LETTER G WITH STROKE */ \
+	  0x01E5, /* LATIN SMALL LETTER G WITH STROKE */ }, \
+	{ 0x0194, /* LATIN CAPITAL LETTER GAMMA */ \
+	  0x0263, /* LATIN SMALL LETTER GAMMA */ }, \
+	{ 0x0241, /* LATIN CAPITAL LETTER GLOTTAL STOP */ \
+	  0x0242, /* LATIN SMALL LETTER GLOTTAL STOP */ }, \
+	{ 0x0048, /* LATIN CAPITAL LETTER H */ \
+	  0x0068, /* LATIN SMALL LETTER H */ }, \
+	{ 0x1E2A, /* LATIN CAPITAL LETTER H WITH BREVE BELOW */ \
+	  0x1E2B, /* LATIN SMALL LETTER H WITH BREVE BELOW */ }, \
+	{ 0x021E, /* LATIN CAPITAL LETTER H WITH CARON */ \
+	  0x021F, /* LATIN SMALL LETTER H WITH CARON */ }, \
+	{ 0x1E28, /* LATIN CAPITAL LETTER H WITH CEDILLA */ \
+	  0x1E29, /* LATIN SMALL LETTER H WITH CEDILLA */ }, \
+	{ 0x0124, /* LATIN CAPITAL LETTER H WITH CIRCUMFLEX */ \
+	  0x0125, /* LATIN SMALL LETTER H WITH CIRCUMFLEX */ }, \
+	{ 0x2C67, /* LATIN CAPITAL LETTER H WITH DESCENDER */ \
+	  0x2C68, /* LATIN SMALL LETTER H WITH DESCENDER */ }, \
+	{ 0x1E26, /* LATIN CAPITAL LETTER H WITH DIAERESIS */ \
+	  0x1E27, /* LATIN SMALL LETTER H WITH DIAERESIS */ }, \
+	{ 0x1E22, /* LATIN CAPITAL LETTER H WITH DOT ABOVE */ \
+	  0x1E23, /* LATIN SMALL LETTER H WITH DOT ABOVE */ }, \
+	{ 0x1E24, /* LATIN CAPITAL LETTER H WITH DOT BELOW */ \
+	  0x1E25, /* LATIN SMALL LETTER H WITH DOT BELOW */ }, \
+	{ 0xA7AA, /* LATIN CAPITAL LETTER H WITH HOOK */ \
+	  0x0266, /* LATIN SMALL LETTER H WITH HOOK */ }, \
+	{ 0x0126, /* LATIN CAPITAL LETTER H WITH STROKE */ \
+	  0x0127, /* LATIN SMALL LETTER H WITH STROKE */ }, \
+	{ 0x2C75, /* LATIN CAPITAL LETTER HALF H */ \
+	  0x2C76, /* LATIN SMALL LETTER HALF H */ }, \
+	{ 0xA726, /* LATIN CAPITAL LETTER HENG */ \
+	  0xA727, /* LATIN SMALL LETTER HENG */ }, \
+	{ 0x0049, /* LATIN CAPITAL LETTER I */ \
+	  0x0069, /* LATIN SMALL LETTER I */ }, \
+	{ 0x00CD, /* LATIN CAPITAL LETTER I WITH ACUTE */ \
+	  0x00ED, /* LATIN SMALL LETTER I WITH ACUTE */ }, \
+	{ 0x012C, /* LATIN CAPITAL LETTER I WITH BREVE */ \
+	  0x012D, /* LATIN SMALL LETTER I WITH BREVE */ }, \
+	{ 0x01CF, /* LATIN CAPITAL LETTER I WITH CARON */ \
+	  0x01D0, /* LATIN SMALL LETTER I WITH CARON */ }, \
+	{ 0x00CE, /* LATIN CAPITAL LETTER I WITH CIRCUMFLEX */ \
+	  0x00EE, /* LATIN SMALL LETTER I WITH CIRCUMFLEX */ }, \
+	{ 0x00CF, /* LATIN CAPITAL LETTER I WITH DIAERESIS */ \
+	  0x00EF, /* LATIN SMALL LETTER I WITH DIAERESIS */ }, \
+	{ 0x1E2E, /* LATIN CAPITAL LETTER I WITH DIAERESIS AND ACUTE */ \
+	  0x1E2F, /* LATIN SMALL LETTER I WITH DIAERESIS AND ACUTE */ }, \
+	{ 0x1ECA, /* LATIN CAPITAL LETTER I WITH DOT BELOW */ \
+	  0x1ECB, /* LATIN SMALL LETTER I WITH DOT BELOW */ }, \
+	{ 0x0208, /* LATIN CAPITAL LETTER I WITH DOUBLE GRAVE */ \
+	  0x0209, /* LATIN SMALL LETTER I WITH DOUBLE GRAVE */ }, \
+	{ 0x00CC, /* LATIN CAPITAL LETTER I WITH GRAVE */ \
+	  0x00EC, /* LATIN SMALL LETTER I WITH GRAVE */ }, \
+	{ 0x1EC8, /* LATIN CAPITAL LETTER I WITH HOOK ABOVE */ \
+	  0x1EC9, /* LATIN SMALL LETTER I WITH HOOK ABOVE */ }, \
+	{ 0x020A, /* LATIN CAPITAL LETTER I WITH INVERTED BREVE */ \
+	  0x020B, /* LATIN SMALL LETTER I WITH INVERTED BREVE */ }, \
+	{ 0x012A, /* LATIN CAPITAL LETTER I WITH MACRON */ \
+	  0x012B, /* LATIN SMALL LETTER I WITH MACRON */ }, \
+	{ 0x012E, /* LATIN CAPITAL LETTER I WITH OGONEK */ \
+	  0x012F, /* LATIN SMALL LETTER I WITH OGONEK */ }, \
+	{ 0x0197, /* LATIN CAPITAL LETTER I WITH STROKE */ \
+	  0x0268, /* LATIN SMALL LETTER I WITH STROKE */ }, \
+	{ 0x0128, /* LATIN CAPITAL LETTER I WITH TILDE */ \
+	  0x0129, /* LATIN SMALL LETTER I WITH TILDE */ }, \
+	{ 0x1E2C, /* LATIN CAPITAL LETTER I WITH TILDE BELOW */ \
+	  0x1E2D, /* LATIN SMALL LETTER I WITH TILDE BELOW */ }, \
+	{ 0xA779, /* LATIN CAPITAL LETTER INSULAR D */ \
+	  0xA77A, /* LATIN SMALL LETTER INSULAR D */ }, \
+	{ 0xA77B, /* LATIN CAPITAL LETTER INSULAR F */ \
+	  0xA77C, /* LATIN SMALL LETTER INSULAR F */ }, \
+	{ 0xA77D, /* LATIN CAPITAL LETTER INSULAR G */ \
+	  0x1D79, /* LATIN SMALL LETTER INSULAR G */ }, \
+	{ 0xA782, /* LATIN CAPITAL LETTER INSULAR R */ \
+	  0xA783, /* LATIN SMALL LETTER INSULAR R */ }, \
+	{ 0xA784, /* LATIN CAPITAL LETTER INSULAR S */ \
+	  0xA785, /* LATIN SMALL LETTER INSULAR S */ }, \
+	{ 0xA786, /* LATIN CAPITAL LETTER INSULAR T */ \
+	  0xA787, /* LATIN SMALL LETTER INSULAR T */ }, \
+	{ 0x0196, /* LATIN CAPITAL LETTER IOTA */ \
+	  0x0269, /* LATIN SMALL LETTER IOTA */ }, \
+	{ 0xA76C, /* LATIN CAPITAL LETTER IS */ \
+	  0xA76D, /* LATIN SMALL LETTER IS */ }, \
+	{ 0x004A, /* LATIN CAPITAL LETTER J */ \
+	  0x006A, /* LATIN SMALL LETTER J */ }, \
+	{ 0x0134, /* LATIN CAPITAL LETTER J WITH CIRCUMFLEX */ \
+	  0x0135, /* LATIN SMALL LETTER J WITH CIRCUMFLEX */ }, \
+	{ 0xA7B2, /* LATIN CAPITAL LETTER J WITH CROSSED-TAIL */ \
+	  0x029D, /* LATIN SMALL LETTER J WITH CROSSED-TAIL */ }, \
+	{ 0x0248, /* LATIN CAPITAL LETTER J WITH STROKE */ \
+	  0x0249, /* LATIN SMALL LETTER J WITH STROKE */ }, \
+	{ 0x004B, /* LATIN CAPITAL LETTER K */ \
+	  0x006B, /* LATIN SMALL LETTER K */ }, \
+	{ 0x1E30, /* LATIN CAPITAL LETTER K WITH ACUTE */ \
+	  0x1E31, /* LATIN SMALL LETTER K WITH ACUTE */ }, \
+	{ 0x01E8, /* LATIN CAPITAL LETTER K WITH CARON */ \
+	  0x01E9, /* LATIN SMALL LETTER K WITH CARON */ }, \
+	{ 0x0136, /* LATIN CAPITAL LETTER K WITH CEDILLA */ \
+	  0x0137, /* LATIN SMALL LETTER K WITH CEDILLA */ }, \
+	{ 0x2C69, /* LATIN CAPITAL LETTER K WITH DESCENDER */ \
+	  0x2C6A, /* LATIN SMALL LETTER K WITH DESCENDER */ }, \
+	{ 0xA742, /* LATIN CAPITAL LETTER K WITH DIAGONAL STROKE */ \
+	  0xA743, /* LATIN SMALL LETTER K WITH DIAGONAL STROKE */ }, \
+	{ 0x1E32, /* LATIN CAPITAL LETTER K WITH DOT BELOW */ \
+	  0x1E33, /* LATIN SMALL LETTER K WITH DOT BELOW */ }, \
+	{ 0x0198, /* LATIN CAPITAL LETTER K WITH HOOK */ \
+	  0x0199, /* LATIN SMALL LETTER K WITH HOOK */ }, \
+	{ 0x1E34, /* LATIN CAPITAL LETTER K WITH LINE BELOW */ \
+	  0x1E35, /* LATIN SMALL LETTER K WITH LINE BELOW */ }, \
+	{ 0xA7A2, /* LATIN CAPITAL LETTER K WITH OBLIQUE STROKE */ \
+	  0xA7A3, /* LATIN SMALL LETTER K WITH OBLIQUE STROKE */ }, \
+	{ 0xA740, /* LATIN CAPITAL LETTER K WITH STROKE */ \
+	  0xA741, /* LATIN SMALL LETTER K WITH STROKE */ }, \
+	{ 0xA744, /* LATIN CAPITAL LETTER K WITH STROKE AND DIAGONAL STROKE */ \
+	  0xA745, /* LATIN SMALL LETTER K WITH STROKE AND DIAGONAL STROKE */ }, \
+	{ 0x004C, /* LATIN CAPITAL LETTER L */ \
+	  0x006C, /* LATIN SMALL LETTER L */ }, \
+	{ 0x0139, /* LATIN CAPITAL LETTER L WITH ACUTE */ \
+	  0x013A, /* LATIN SMALL LETTER L WITH ACUTE */ }, \
+	{ 0x023D, /* LATIN CAPITAL LETTER L WITH BAR */ \
+	  0x019A, /* LATIN SMALL LETTER L WITH BAR */ }, \
+	{ 0xA7AD, /* LATIN CAPITAL LETTER L WITH BELT */ \
+	  0x026C, /* LATIN SMALL LETTER L WITH BELT */ }, \
+	{ 0x013D, /* LATIN CAPITAL LETTER L WITH CARON */ \
+	  0x013E, /* LATIN SMALL LETTER L WITH CARON */ }, \
+	{ 0x013B, /* LATIN CAPITAL LETTER L WITH CEDILLA */ \
+	  0x013C, /* LATIN SMALL LETTER L WITH CEDILLA */ }, \
+	{ 0x1E3C, /* LATIN CAPITAL LETTER L WITH CIRCUMFLEX BELOW */ \
+	  0x1E3D, /* LATIN SMALL LETTER L WITH CIRCUMFLEX BELOW */ }, \
+	{ 0x1E36, /* LATIN CAPITAL LETTER L WITH DOT BELOW */ \
+	  0x1E37, /* LATIN SMALL LETTER L WITH DOT BELOW */ }, \
+	{ 0x1E38, /* LATIN CAPITAL LETTER L WITH DOT BELOW AND MACRON */ \
+	  0x1E39, /* LATIN SMALL LETTER L WITH DOT BELOW AND MACRON */ }, \
+	{ 0x2C60, /* LATIN CAPITAL LETTER L WITH DOUBLE BAR */ \
+	  0x2C61, /* LATIN SMALL LETTER L WITH DOUBLE BAR */ }, \
+	{ 0xA748, /* LATIN CAPITAL LETTER L WITH HIGH STROKE */ \
+	  0xA749, /* LATIN SMALL LETTER L WITH HIGH STROKE */ }, \
+	{ 0x1E3A, /* LATIN CAPITAL LETTER L WITH LINE BELOW */ \
+	  0x1E3B, /* LATIN SMALL LETTER L WITH LINE BELOW */ }, \
+	{ 0x013F, /* LATIN CAPITAL LETTER L WITH MIDDLE DOT */ \
+	  0x0140, /* LATIN SMALL LETTER L WITH MIDDLE DOT */ }, \
+	{ 0x2C62, /* LATIN CAPITAL LETTER L WITH MIDDLE TILDE */ \
+	  0x026B, /* LATIN SMALL LETTER L WITH MIDDLE TILDE */ }, \
+	{ 0x0141, /* LATIN CAPITAL LETTER L WITH STROKE */ \
+	  0x0142, /* LATIN SMALL LETTER L WITH STROKE */ }, \
+	{ 0x01C7, /* LATIN CAPITAL LETTER LJ */ \
+	  0x01C9, /* LATIN SMALL LETTER LJ */ }, \
+	{ 0x004D, /* LATIN CAPITAL LETTER M */ \
+	  0x006D, /* LATIN SMALL LETTER M */ }, \
+	{ 0x1E3E, /* LATIN CAPITAL LETTER M WITH ACUTE */ \
+	  0x1E3F, /* LATIN SMALL LETTER M WITH ACUTE */ }, \
+	{ 0x1E40, /* LATIN CAPITAL LETTER M WITH DOT ABOVE */ \
+	  0x1E41, /* LATIN SMALL LETTER M WITH DOT ABOVE */ }, \
+	{ 0x1E42, /* LATIN CAPITAL LETTER M WITH DOT BELOW */ \
+	  0x1E43, /* LATIN SMALL LETTER M WITH DOT BELOW */ }, \
+	{ 0x2C6E, /* LATIN CAPITAL LETTER M WITH HOOK */ \
+	  0x0271, /* LATIN SMALL LETTER M WITH HOOK */ }, \
+	{ 0x1EFA, /* LATIN CAPITAL LETTER MIDDLE-WELSH LL */ \
+	  0x1EFB, /* LATIN SMALL LETTER MIDDLE-WELSH LL */ }, \
+	{ 0x1EFC, /* LATIN CAPITAL LETTER MIDDLE-WELSH V */ \
+	  0x1EFD, /* LATIN SMALL LETTER MIDDLE-WELSH V */ }, \
+	{ 0x004E, /* LATIN CAPITAL LETTER N */ \
+	  0x006E, /* LATIN SMALL LETTER N */ }, \
+	{ 0x0143, /* LATIN CAPITAL LETTER N WITH ACUTE */ \
+	  0x0144, /* LATIN SMALL LETTER N WITH ACUTE */ }, \
+	{ 0x0147, /* LATIN CAPITAL LETTER N WITH CARON */ \
+	  0x0148, /* LATIN SMALL LETTER N WITH CARON */ }, \
+	{ 0x0145, /* LATIN CAPITAL LETTER N WITH CEDILLA */ \
+	  0x0146, /* LATIN SMALL LETTER N WITH CEDILLA */ }, \
+	{ 0x1E4A, /* LATIN CAPITAL LETTER N WITH CIRCUMFLEX BELOW */ \
+	  0x1E4B, /* LATIN SMALL LETTER N WITH CIRCUMFLEX BELOW */ }, \
+	{ 0xA790, /* LATIN CAPITAL LETTER N WITH DESCENDER */ \
+	  0xA791, /* LATIN SMALL LETTER N WITH DESCENDER */ }, \
+	{ 0x1E44, /* LATIN CAPITAL LETTER N WITH DOT ABOVE */ \
+	  0x1E45, /* LATIN SMALL LETTER N WITH DOT ABOVE */ }, \
+	{ 0x1E46, /* LATIN CAPITAL LETTER N WITH DOT BELOW */ \
+	  0x1E47, /* LATIN SMALL LETTER N WITH DOT BELOW */ }, \
+	{ 0x01F8, /* LATIN CAPITAL LETTER N WITH GRAVE */ \
+	  0x01F9, /* LATIN SMALL LETTER N WITH GRAVE */ }, \
+	{ 0x019D, /* LATIN CAPITAL LETTER N WITH LEFT HOOK */ \
+	  0x0272, /* LATIN SMALL LETTER N WITH LEFT HOOK */ }, \
+	{ 0x1E48, /* LATIN CAPITAL LETTER N WITH LINE BELOW */ \
+	  0x1E49, /* LATIN SMALL LETTER N WITH LINE BELOW */ }, \
+	{ 0x0220, /* LATIN CAPITAL LETTER N WITH LONG RIGHT LEG */ \
+	  0x019E, /* LATIN SMALL LETTER N WITH LONG RIGHT LEG */ }, \
+	{ 0xA7A4, /* LATIN CAPITAL LETTER N WITH OBLIQUE STROKE */ \
+	  0xA7A5, /* LATIN SMALL LETTER N WITH OBLIQUE STROKE */ }, \
+	{ 0x00D1, /* LATIN CAPITAL LETTER N WITH TILDE */ \
+	  0x00F1, /* LATIN SMALL LETTER N WITH TILDE */ }, \
+	{ 0x01CA, /* LATIN CAPITAL LETTER NJ */ \
+	  0x01CC, /* LATIN SMALL LETTER NJ */ }, \
+	{ 0x004F, /* LATIN CAPITAL LETTER O */ \
+	  0x006F, /* LATIN SMALL LETTER O */ }, \
+	{ 0x00D3, /* LATIN CAPITAL LETTER O WITH ACUTE */ \
+	  0x00F3, /* LATIN SMALL LETTER O WITH ACUTE */ }, \
+	{ 0x014E, /* LATIN CAPITAL LETTER O WITH BREVE */ \
+	  0x014F, /* LATIN SMALL LETTER O WITH BREVE */ }, \
+	{ 0x01D1, /* LATIN CAPITAL LETTER O WITH CARON */ \
+	  0x01D2, /* LATIN SMALL LETTER O WITH CARON */ }, \
+	{ 0x00D4, /* LATIN CAPITAL LETTER O WITH CIRCUMFLEX */ \
+	  0x00F4, /* LATIN SMALL LETTER O WITH CIRCUMFLEX */ }, \
+	{ 0x1ED0, /* LATIN CAPITAL LETTER O WITH CIRCUMFLEX AND ACUTE */ \
+	  0x1ED1, /* LATIN SMALL LETTER O WITH CIRCUMFLEX AND ACUTE */ }, \
+	{ 0x1ED8, /* LATIN CAPITAL LETTER O WITH CIRCUMFLEX AND DOT BELOW */ \
+	  0x1ED9, /* LATIN SMALL LETTER O WITH CIRCUMFLEX AND DOT BELOW */ }, \
+	{ 0x1ED2, /* LATIN CAPITAL LETTER O WITH CIRCUMFLEX AND GRAVE */ \
+	  0x1ED3, /* LATIN SMALL LETTER O WITH CIRCUMFLEX AND GRAVE */ }, \
+	{ 0x1ED4, /* LATIN CAPITAL LETTER O WITH CIRCUMFLEX AND HOOK ABOVE */ \
+	  0x1ED5, /* LATIN SMALL LETTER O WITH CIRCUMFLEX AND HOOK ABOVE */ }, \
+	{ 0x1ED6, /* LATIN CAPITAL LETTER O WITH CIRCUMFLEX AND TILDE */ \
+	  0x1ED7, /* LATIN SMALL LETTER O WITH CIRCUMFLEX AND TILDE */ }, \
+	{ 0x00D6, /* LATIN CAPITAL LETTER O WITH DIAERESIS */ \
+	  0x00F6, /* LATIN SMALL LETTER O WITH DIAERESIS */ }, \
+	{ 0x022A, /* LATIN CAPITAL LETTER O WITH DIAERESIS AND MACRON */ \
+	  0x022B, /* LATIN SMALL LETTER O WITH DIAERESIS AND MACRON */ }, \
+	{ 0x022E, /* LATIN CAPITAL LETTER O WITH DOT ABOVE */ \
+	  0x022F, /* LATIN SMALL LETTER O WITH DOT ABOVE */ }, \
+	{ 0x0230, /* LATIN CAPITAL LETTER O WITH DOT ABOVE AND MACRON */ \
+	  0x0231, /* LATIN SMALL LETTER O WITH DOT ABOVE AND MACRON */ }, \
+	{ 0x1ECC, /* LATIN CAPITAL LETTER O WITH DOT BELOW */ \
+	  0x1ECD, /* LATIN SMALL LETTER O WITH DOT BELOW */ }, \
+	{ 0x0150, /* LATIN CAPITAL LETTER O WITH DOUBLE ACUTE */ \
+	  0x0151, /* LATIN SMALL LETTER O WITH DOUBLE ACUTE */ }, \
+	{ 0x020C, /* LATIN CAPITAL LETTER O WITH DOUBLE GRAVE */ \
+	  0x020D, /* LATIN SMALL LETTER O WITH DOUBLE GRAVE */ }, \
+	{ 0x00D2, /* LATIN CAPITAL LETTER O WITH GRAVE */ \
+	  0x00F2, /* LATIN SMALL LETTER O WITH GRAVE */ }, \
+	{ 0x1ECE, /* LATIN CAPITAL LETTER O WITH HOOK ABOVE */ \
+	  0x1ECF, /* LATIN SMALL LETTER O WITH HOOK ABOVE */ }, \
+	{ 0x01A0, /* LATIN CAPITAL LETTER O WITH HORN */ \
+	  0x01A1, /* LATIN SMALL LETTER O WITH HORN */ }, \
+	{ 0x1EDA, /* LATIN CAPITAL LETTER O WITH HORN AND ACUTE */ \
+	  0x1EDB, /* LATIN SMALL LETTER O WITH HORN AND ACUTE */ }, \
+	{ 0x1EE2, /* LATIN CAPITAL LETTER O WITH HORN AND DOT BELOW */ \
+	  0x1EE3, /* LATIN SMALL LETTER O WITH HORN AND DOT BELOW */ }, \
+	{ 0x1EDC, /* LATIN CAPITAL LETTER O WITH HORN AND GRAVE */ \
+	  0x1EDD, /* LATIN SMALL LETTER O WITH HORN AND GRAVE */ }, \
+	{ 0x1EDE, /* LATIN CAPITAL LETTER O WITH HORN AND HOOK ABOVE */ \
+	  0x1EDF, /* LATIN SMALL LETTER O WITH HORN AND HOOK ABOVE */ }, \
+	{ 0x1EE0, /* LATIN CAPITAL LETTER O WITH HORN AND TILDE */ \
+	  0x1EE1, /* LATIN SMALL LETTER O WITH HORN AND TILDE */ }, \
+	{ 0x020E, /* LATIN CAPITAL LETTER O WITH INVERTED BREVE */ \
+	  0x020F, /* LATIN SMALL LETTER O WITH INVERTED BREVE */ }, \
+	{ 0xA74A, /* LATIN CAPITAL LETTER O WITH LONG STROKE OVERLAY */ \
+	  0xA74B, /* LATIN SMALL LETTER O WITH LONG STROKE OVERLAY */ }, \
+	{ 0xA74C, /* LATIN CAPITAL LETTER O WITH LOOP */ \
+	  0xA74D, /* LATIN SMALL LETTER O WITH LOOP */ }, \
+	{ 0x014C, /* LATIN CAPITAL LETTER O WITH MACRON */ \
+	  0x014D, /* LATIN SMALL LETTER O WITH MACRON */ }, \
+	{ 0x1E52, /* LATIN CAPITAL LETTER O WITH MACRON AND ACUTE */ \
+	  0x1E53, /* LATIN SMALL LETTER O WITH MACRON AND ACUTE */ }, \
+	{ 0x1E50, /* LATIN CAPITAL LETTER O WITH MACRON AND GRAVE */ \
+	  0x1E51, /* LATIN SMALL LETTER O WITH MACRON AND GRAVE */ }, \
+	{ 0x01EA, /* LATIN CAPITAL LETTER O WITH OGONEK */ \
+	  0x01EB, /* LATIN SMALL LETTER O WITH OGONEK */ }, \
+	{ 0x01EC, /* LATIN CAPITAL LETTER O WITH OGONEK AND MACRON */ \
+	  0x01ED, /* LATIN SMALL LETTER O WITH OGONEK AND MACRON */ }, \
+	{ 0x00D8, /* LATIN CAPITAL LETTER O WITH STROKE */ \
+	  0x00F8, /* LATIN SMALL LETTER O WITH STROKE */ }, \
+	{ 0x01FE, /* LATIN CAPITAL LETTER O WITH STROKE AND ACUTE */ \
+	  0x01FF, /* LATIN SMALL LETTER O WITH STROKE AND ACUTE */ }, \
+	{ 0x00D5, /* LATIN CAPITAL LETTER O WITH TILDE */ \
+	  0x00F5, /* LATIN SMALL LETTER O WITH TILDE */ }, \
+	{ 0x1E4C, /* LATIN CAPITAL LETTER O WITH TILDE AND ACUTE */ \
+	  0x1E4D, /* LATIN SMALL LETTER O WITH TILDE AND ACUTE */ }, \
+	{ 0x1E4E, /* LATIN CAPITAL LETTER O WITH TILDE AND DIAERESIS */ \
+	  0x1E4F, /* LATIN SMALL LETTER O WITH TILDE AND DIAERESIS */ }, \
+	{ 0x022C, /* LATIN CAPITAL LETTER O WITH TILDE AND MACRON */ \
+	  0x022D, /* LATIN SMALL LETTER O WITH TILDE AND MACRON */ }, \
+	{ 0x01A2, /* LATIN CAPITAL LETTER OI */ \
+	  0x01A3, /* LATIN SMALL LETTER OI */ }, \
+	{ 0xA7B6, /* LATIN CAPITAL LETTER OMEGA */ \
+	  0xA7B7, /* LATIN SMALL LETTER OMEGA */ }, \
+	{ 0xA74E, /* LATIN CAPITAL LETTER OO */ \
+	  0xA74F, /* LATIN SMALL LETTER OO */ }, \
+	{ 0x0190, /* LATIN CAPITAL LETTER OPEN E */ \
+	  0x025B, /* LATIN SMALL LETTER OPEN E */ }, \
+	{ 0x0186, /* LATIN CAPITAL LETTER OPEN O */ \
+	  0x0254, /* LATIN SMALL LETTER OPEN O */ }, \
+	{ 0x0222, /* LATIN CAPITAL LETTER OU */ \
+	  0x0223, /* LATIN SMALL LETTER OU */ }, \
+	{ 0x0050, /* LATIN CAPITAL LETTER P */ \
+	  0x0070, /* LATIN SMALL LETTER P */ }, \
+	{ 0x1E54, /* LATIN CAPITAL LETTER P WITH ACUTE */ \
+	  0x1E55, /* LATIN SMALL LETTER P WITH ACUTE */ }, \
+	{ 0x1E56, /* LATIN CAPITAL LETTER P WITH DOT ABOVE */ \
+	  0x1E57, /* LATIN SMALL LETTER P WITH DOT ABOVE */ }, \
+	{ 0xA752, /* LATIN CAPITAL LETTER P WITH FLOURISH */ \
+	  0xA753, /* LATIN SMALL LETTER P WITH FLOURISH */ }, \
+	{ 0x01A4, /* LATIN CAPITAL LETTER P WITH HOOK */ \
+	  0x01A5, /* LATIN SMALL LETTER P WITH HOOK */ }, \
+	{ 0xA754, /* LATIN CAPITAL LETTER P WITH SQUIRREL TAIL */ \
+	  0xA755, /* LATIN SMALL LETTER P WITH SQUIRREL TAIL */ }, \
+	{ 0x2C63, /* LATIN CAPITAL LETTER P WITH STROKE */ \
+	  0x1D7D, /* LATIN SMALL LETTER P WITH STROKE */ }, \
+	{ 0xA750, /* LATIN CAPITAL LETTER P WITH STROKE THROUGH DESCENDER */ \
+	  0xA751, /* LATIN SMALL LETTER P WITH STROKE THROUGH DESCENDER */ }, \
+	{ 0x0051, /* LATIN CAPITAL LETTER Q */ \
+	  0x0071, /* LATIN SMALL LETTER Q */ }, \
+	{ 0xA758, /* LATIN CAPITAL LETTER Q WITH DIAGONAL STROKE */ \
+	  0xA759, /* LATIN SMALL LETTER Q WITH DIAGONAL STROKE */ }, \
+	{ 0xA756, /* LATIN CAPITAL LETTER Q WITH STROKE THROUGH DESCENDER */ \
+	  0xA757, /* LATIN SMALL LETTER Q WITH STROKE THROUGH DESCENDER */ }, \
+	{ 0x0052, /* LATIN CAPITAL LETTER R */ \
+	  0x0072, /* LATIN SMALL LETTER R */ }, \
+	{ 0xA75A, /* LATIN CAPITAL LETTER R ROTUNDA */ \
+	  0xA75B, /* LATIN SMALL LETTER R ROTUNDA */ }, \
+	{ 0x0154, /* LATIN CAPITAL LETTER R WITH ACUTE */ \
+	  0x0155, /* LATIN SMALL LETTER R WITH ACUTE */ }, \
+	{ 0x0158, /* LATIN CAPITAL LETTER R WITH CARON */ \
+	  0x0159, /* LATIN SMALL LETTER R WITH CARON */ }, \
+	{ 0x0156, /* LATIN CAPITAL LETTER R WITH CEDILLA */ \
+	  0x0157, /* LATIN SMALL LETTER R WITH CEDILLA */ }, \
+	{ 0x1E58, /* LATIN CAPITAL LETTER R WITH DOT ABOVE */ \
+	  0x1E59, /* LATIN SMALL LETTER R WITH DOT ABOVE */ }, \
+	{ 0x1E5A, /* LATIN CAPITAL LETTER R WITH DOT BELOW */ \
+	  0x1E5B, /* LATIN SMALL LETTER R WITH DOT BELOW */ }, \
+	{ 0x1E5C, /* LATIN CAPITAL LETTER R WITH DOT BELOW AND MACRON */ \
+	  0x1E5D, /* LATIN SMALL LETTER R WITH DOT BELOW AND MACRON */ }, \
+	{ 0x0210, /* LATIN CAPITAL LETTER R WITH DOUBLE GRAVE */ \
+	  0x0211, /* LATIN SMALL LETTER R WITH DOUBLE GRAVE */ }, \
+	{ 0x0212, /* LATIN CAPITAL LETTER R WITH INVERTED BREVE */ \
+	  0x0213, /* LATIN SMALL LETTER R WITH INVERTED BREVE */ }, \
+	{ 0x1E5E, /* LATIN CAPITAL LETTER R WITH LINE BELOW */ \
+	  0x1E5F, /* LATIN SMALL LETTER R WITH LINE BELOW */ }, \
+	{ 0xA7A6, /* LATIN CAPITAL LETTER R WITH OBLIQUE STROKE */ \
+	  0xA7A7, /* LATIN SMALL LETTER R WITH OBLIQUE STROKE */ }, \
+	{ 0x024C, /* LATIN CAPITAL LETTER R WITH STROKE */ \
+	  0x024D, /* LATIN SMALL LETTER R WITH STROKE */ }, \
+	{ 0x2C64, /* LATIN CAPITAL LETTER R WITH TAIL */ \
+	  0x027D, /* LATIN SMALL LETTER R WITH TAIL */ }, \
+	{ 0xA73E, /* LATIN CAPITAL LETTER REVERSED C WITH DOT */ \
+	  0xA73F, /* LATIN SMALL LETTER REVERSED C WITH DOT */ }, \
+	{ 0x018E, /* LATIN CAPITAL LETTER REVERSED E */ \
+	  0x0258, /* LATIN SMALL LETTER REVERSED E */ }, \
+	{ 0xA7AB, /* LATIN CAPITAL LETTER REVERSED OPEN E */ \
+	  0x025C, /* LATIN SMALL LETTER REVERSED OPEN E */ }, \
+	{ 0xA75C, /* LATIN CAPITAL LETTER RUM ROTUNDA */ \
+	  0xA75D, /* LATIN SMALL LETTER RUM ROTUNDA */ }, \
+	{ 0x0053, /* LATIN CAPITAL LETTER S */ \
+	  0x0073, /* LATIN SMALL LETTER S */ }, \
+	{ 0x015A, /* LATIN CAPITAL LETTER S WITH ACUTE */ \
+	  0x015B, /* LATIN SMALL LETTER S WITH ACUTE */ }, \
+	{ 0x1E64, /* LATIN CAPITAL LETTER S WITH ACUTE AND DOT ABOVE */ \
+	  0x1E65, /* LATIN SMALL LETTER S WITH ACUTE AND DOT ABOVE */ }, \
+	{ 0x0160, /* LATIN CAPITAL LETTER S WITH CARON */ \
+	  0x0161, /* LATIN SMALL LETTER S WITH CARON */ }, \
+	{ 0x1E66, /* LATIN CAPITAL LETTER S WITH CARON AND DOT ABOVE */ \
+	  0x1E67, /* LATIN SMALL LETTER S WITH CARON AND DOT ABOVE */ }, \
+	{ 0x015E, /* LATIN CAPITAL LETTER S WITH CEDILLA */ \
+	  0x015F, /* LATIN SMALL LETTER S WITH CEDILLA */ }, \
+	{ 0x015C, /* LATIN CAPITAL LETTER S WITH CIRCUMFLEX */ \
+	  0x015D, /* LATIN SMALL LETTER S WITH CIRCUMFLEX */ }, \
+	{ 0x0218, /* LATIN CAPITAL LETTER S WITH COMMA BELOW */ \
+	  0x0219, /* LATIN SMALL LETTER S WITH COMMA BELOW */ }, \
+	{ 0x1E60, /* LATIN CAPITAL LETTER S WITH DOT ABOVE */ \
+	  0x1E61, /* LATIN SMALL LETTER S WITH DOT ABOVE */ }, \
+	{ 0x1E62, /* LATIN CAPITAL LETTER S WITH DOT BELOW */ \
+	  0x1E63, /* LATIN SMALL LETTER S WITH DOT BELOW */ }, \
+	{ 0x1E68, /* LATIN CAPITAL LETTER S WITH DOT BELOW AND DOT ABOVE */ \
+	  0x1E69, /* LATIN SMALL LETTER S WITH DOT BELOW AND DOT ABOVE */ }, \
+	{ 0xA7A8, /* LATIN CAPITAL LETTER S WITH OBLIQUE STROKE */ \
+	  0xA7A9, /* LATIN SMALL LETTER S WITH OBLIQUE STROKE */ }, \
+	{ 0x2C7E, /* LATIN CAPITAL LETTER S WITH SWASH TAIL */ \
+	  0x023F, /* LATIN SMALL LETTER S WITH SWASH TAIL */ }, \
+	{ 0xA78B, /* LATIN CAPITAL LETTER SALTILLO */ \
+	  0xA78C, /* LATIN SMALL LETTER SALTILLO */ }, \
+	{ 0x018F, /* LATIN CAPITAL LETTER SCHWA */ \
+	  0x0259, /* LATIN SMALL LETTER SCHWA */ }, \
+	{ 0xA7AC, /* LATIN CAPITAL LETTER SCRIPT G */ \
+	  0x0261, /* LATIN SMALL LETTER SCRIPT G */ }, \
+	{ 0x1E9E, /* LATIN CAPITAL LETTER SHARP S */ \
+	  0x00DF, /* LATIN SMALL LETTER SHARP S */ }, \
+	{ 0x0054, /* LATIN CAPITAL LETTER T */ \
+	  0x0074, /* LATIN SMALL LETTER T */ }, \
+	{ 0x0164, /* LATIN CAPITAL LETTER T WITH CARON */ \
+	  0x0165, /* LATIN SMALL LETTER T WITH CARON */ }, \
+	{ 0x0162, /* LATIN CAPITAL LETTER T WITH CEDILLA */ \
+	  0x0163, /* LATIN SMALL LETTER T WITH CEDILLA */ }, \
+	{ 0x1E70, /* LATIN CAPITAL LETTER T WITH CIRCUMFLEX BELOW */ \
+	  0x1E71, /* LATIN SMALL LETTER T WITH CIRCUMFLEX BELOW */ }, \
+	{ 0x021A, /* LATIN CAPITAL LETTER T WITH COMMA BELOW */ \
+	  0x021B, /* LATIN SMALL LETTER T WITH COMMA BELOW */ }, \
+	{ 0x023E, /* LATIN CAPITAL LETTER T WITH DIAGONAL STROKE */ \
+	  0x2C66, /* LATIN SMALL LETTER T WITH DIAGONAL STROKE */ }, \
+	{ 0x1E6A, /* LATIN CAPITAL LETTER T WITH DOT ABOVE */ \
+	  0x1E6B, /* LATIN SMALL LETTER T WITH DOT ABOVE */ }, \
+	{ 0x1E6C, /* LATIN CAPITAL LETTER T WITH DOT BELOW */ \
+	  0x1E6D, /* LATIN SMALL LETTER T WITH DOT BELOW */ }, \
+	{ 0x01AC, /* LATIN CAPITAL LETTER T WITH HOOK */ \
+	  0x01AD, /* LATIN SMALL LETTER T WITH HOOK */ }, \
+	{ 0x1E6E, /* LATIN CAPITAL LETTER T WITH LINE BELOW */ \
+	  0x1E6F, /* LATIN SMALL LETTER T WITH LINE BELOW */ }, \
+	{ 0x01AE, /* LATIN CAPITAL LETTER T WITH RETROFLEX HOOK */ \
+	  0x0288, /* LATIN SMALL LETTER T WITH RETROFLEX HOOK */ }, \
+	{ 0x0166, /* LATIN CAPITAL LETTER T WITH STROKE */ \
+	  0x0167, /* LATIN SMALL LETTER T WITH STROKE */ }, \
+	{ 0x00DE, /* LATIN CAPITAL LETTER THORN */ \
+	  0x00FE, /* LATIN SMALL LETTER THORN */ }, \
+	{ 0xA764, /* LATIN CAPITAL LETTER THORN WITH STROKE */ \
+	  0xA765, /* LATIN SMALL LETTER THORN WITH STROKE */ }, \
+	{ 0xA766, /* LATIN CAPITAL LETTER THORN WITH STROKE THROUGH DESCENDER */ \
+	  0xA767, /* LATIN SMALL LETTER THORN WITH STROKE THROUGH DESCENDER */ }, \
+	{ 0x01BC, /* LATIN CAPITAL LETTER TONE FIVE */ \
+	  0x01BD, /* LATIN SMALL LETTER TONE FIVE */ }, \
+	{ 0x0184, /* LATIN CAPITAL LETTER TONE SIX */ \
+	  0x0185, /* LATIN SMALL LETTER TONE SIX */ }, \
+	{ 0x01A7, /* LATIN CAPITAL LETTER TONE TWO */ \
+	  0x01A8, /* LATIN SMALL LETTER TONE TWO */ }, \
+	{ 0xA72A, /* LATIN CAPITAL LETTER TRESILLO */ \
+	  0xA72B, /* LATIN SMALL LETTER TRESILLO */ }, \
+	{ 0x2C6F, /* LATIN CAPITAL LETTER TURNED A */ \
+	  0x0250, /* LATIN SMALL LETTER TURNED A */ }, \
+	{ 0x2C70, /* LATIN CAPITAL LETTER TURNED ALPHA */ \
+	  0x0252, /* LATIN SMALL LETTER TURNED ALPHA */ }, \
+	{ 0xA78D, /* LATIN CAPITAL LETTER TURNED H */ \
+	  0x0265, /* LATIN SMALL LETTER TURNED H */ }, \
+	{ 0xA77E, /* LATIN CAPITAL LETTER TURNED INSULAR G */ \
+	  0xA77F, /* LATIN SMALL LETTER TURNED INSULAR G */ }, \
+	{ 0xA7B0, /* LATIN CAPITAL LETTER TURNED K */ \
+	  0x029E, /* LATIN SMALL LETTER TURNED K */ }, \
+	{ 0xA780, /* LATIN CAPITAL LETTER TURNED L */ \
+	  0xA781, /* LATIN SMALL LETTER TURNED L */ }, \
+	{ 0x019C, /* LATIN CAPITAL LETTER TURNED M */ \
+	  0x026F, /* LATIN SMALL LETTER TURNED M */ }, \
+	{ 0xA7B1, /* LATIN CAPITAL LETTER TURNED T */ \
+	  0x0287, /* LATIN SMALL LETTER TURNED T */ }, \
+	{ 0x0245, /* LATIN CAPITAL LETTER TURNED V */ \
+	  0x028C, /* LATIN SMALL LETTER TURNED V */ }, \
+	{ 0xA728, /* LATIN CAPITAL LETTER TZ */ \
+	  0xA729, /* LATIN SMALL LETTER TZ */ }, \
+	{ 0x0055, /* LATIN CAPITAL LETTER U */ \
+	  0x0075, /* LATIN SMALL LETTER U */ }, \
+	{ 0x0244, /* LATIN CAPITAL LETTER U BAR */ \
+	  0x0289, /* LATIN SMALL LETTER U BAR */ }, \
+	{ 0x00DA, /* LATIN CAPITAL LETTER U WITH ACUTE */ \
+	  0x00FA, /* LATIN SMALL LETTER U WITH ACUTE */ }, \
+	{ 0x016C, /* LATIN CAPITAL LETTER U WITH BREVE */ \
+	  0x016D, /* LATIN SMALL LETTER U WITH BREVE */ }, \
+	{ 0x01D3, /* LATIN CAPITAL LETTER U WITH CARON */ \
+	  0x01D4, /* LATIN SMALL LETTER U WITH CARON */ }, \
+	{ 0x00DB, /* LATIN CAPITAL LETTER U WITH CIRCUMFLEX */ \
+	  0x00FB, /* LATIN SMALL LETTER U WITH CIRCUMFLEX */ }, \
+	{ 0x1E76, /* LATIN CAPITAL LETTER U WITH CIRCUMFLEX BELOW */ \
+	  0x1E77, /* LATIN SMALL LETTER U WITH CIRCUMFLEX BELOW */ }, \
+	{ 0x00DC, /* LATIN CAPITAL LETTER U WITH DIAERESIS */ \
+	  0x00FC, /* LATIN SMALL LETTER U WITH DIAERESIS */ }, \
+	{ 0x01D7, /* LATIN CAPITAL LETTER U WITH DIAERESIS AND ACUTE */ \
+	  0x01D8, /* LATIN SMALL LETTER U WITH DIAERESIS AND ACUTE */ }, \
+	{ 0x01D9, /* LATIN CAPITAL LETTER U WITH DIAERESIS AND CARON */ \
+	  0x01DA, /* LATIN SMALL LETTER U WITH DIAERESIS AND CARON */ }, \
+	{ 0x01DB, /* LATIN CAPITAL LETTER U WITH DIAERESIS AND GRAVE */ \
+	  0x01DC, /* LATIN SMALL LETTER U WITH DIAERESIS AND GRAVE */ }, \
+	{ 0x01D5, /* LATIN CAPITAL LETTER U WITH DIAERESIS AND MACRON */ \
+	  0x01D6, /* LATIN SMALL LETTER U WITH DIAERESIS AND MACRON */ }, \
+	{ 0x1E72, /* LATIN CAPITAL LETTER U WITH DIAERESIS BELOW */ \
+	  0x1E73, /* LATIN SMALL LETTER U WITH DIAERESIS BELOW */ }, \
+	{ 0x1EE4, /* LATIN CAPITAL LETTER U WITH DOT BELOW */ \
+	  0x1EE5, /* LATIN SMALL LETTER U WITH DOT BELOW */ }, \
+	{ 0x0170, /* LATIN CAPITAL LETTER U WITH DOUBLE ACUTE */ \
+	  0x0171, /* LATIN SMALL LETTER U WITH DOUBLE ACUTE */ }, \
+	{ 0x0214, /* LATIN CAPITAL LETTER U WITH DOUBLE GRAVE */ \
+	  0x0215, /* LATIN SMALL LETTER U WITH DOUBLE GRAVE */ }, \
+	{ 0x00D9, /* LATIN CAPITAL LETTER U WITH GRAVE */ \
+	  0x00F9, /* LATIN SMALL LETTER U WITH GRAVE */ }, \
+	{ 0x1EE6, /* LATIN CAPITAL LETTER U WITH HOOK ABOVE */ \
+	  0x1EE7, /* LATIN SMALL LETTER U WITH HOOK ABOVE */ }, \
+	{ 0x01AF, /* LATIN CAPITAL LETTER U WITH HORN */ \
+	  0x01B0, /* LATIN SMALL LETTER U WITH HORN */ }, \
+	{ 0x1EE8, /* LATIN CAPITAL LETTER U WITH HORN AND ACUTE */ \
+	  0x1EE9, /* LATIN SMALL LETTER U WITH HORN AND ACUTE */ }, \
+	{ 0x1EF0, /* LATIN CAPITAL LETTER U WITH HORN AND DOT BELOW */ \
+	  0x1EF1, /* LATIN SMALL LETTER U WITH HORN AND DOT BELOW */ }, \
+	{ 0x1EEA, /* LATIN CAPITAL LETTER U WITH HORN AND GRAVE */ \
+	  0x1EEB, /* LATIN SMALL LETTER U WITH HORN AND GRAVE */ }, \
+	{ 0x1EEC, /* LATIN CAPITAL LETTER U WITH HORN AND HOOK ABOVE */ \
+	  0x1EED, /* LATIN SMALL LETTER U WITH HORN AND HOOK ABOVE */ }, \
+	{ 0x1EEE, /* LATIN CAPITAL LETTER U WITH HORN AND TILDE */ \
+	  0x1EEF, /* LATIN SMALL LETTER U WITH HORN AND TILDE */ }, \
+	{ 0x0216, /* LATIN CAPITAL LETTER U WITH INVERTED BREVE */ \
+	  0x0217, /* LATIN SMALL LETTER U WITH INVERTED BREVE */ }, \
+	{ 0x016A, /* LATIN CAPITAL LETTER U WITH MACRON */ \
+	  0x016B, /* LATIN SMALL LETTER U WITH MACRON */ }, \
+	{ 0x1E7A, /* LATIN CAPITAL LETTER U WITH MACRON AND DIAERESIS */ \
+	  0x1E7B, /* LATIN SMALL LETTER U WITH MACRON AND DIAERESIS */ }, \
+	{ 0x0172, /* LATIN CAPITAL LETTER U WITH OGONEK */ \
+	  0x0173, /* LATIN SMALL LETTER U WITH OGONEK */ }, \
+	{ 0x016E, /* LATIN CAPITAL LETTER U WITH RING ABOVE */ \
+	  0x016F, /* LATIN SMALL LETTER U WITH RING ABOVE */ }, \
+	{ 0xA7B8, /* LATIN CAPITAL LETTER U WITH STROKE */ \
+	  0xA7B9, /* LATIN SMALL LETTER U WITH STROKE */ }, \
+	{ 0x0168, /* LATIN CAPITAL LETTER U WITH TILDE */ \
+	  0x0169, /* LATIN SMALL LETTER U WITH TILDE */ }, \
+	{ 0x1E78, /* LATIN CAPITAL LETTER U WITH TILDE AND ACUTE */ \
+	  0x1E79, /* LATIN SMALL LETTER U WITH TILDE AND ACUTE */ }, \
+	{ 0x1E74, /* LATIN CAPITAL LETTER U WITH TILDE BELOW */ \
+	  0x1E75, /* LATIN SMALL LETTER U WITH TILDE BELOW */ }, \
+	{ 0x01B1, /* LATIN CAPITAL LETTER UPSILON */ \
+	  0x028A, /* LATIN SMALL LETTER UPSILON */ }, \
+	{ 0x0056, /* LATIN CAPITAL LETTER V */ \
+	  0x0076, /* LATIN SMALL LETTER V */ }, \
+	{ 0xA75E, /* LATIN CAPITAL LETTER V WITH DIAGONAL STROKE */ \
+	  0xA75F, /* LATIN SMALL LETTER V WITH DIAGONAL STROKE */ }, \
+	{ 0x1E7E, /* LATIN CAPITAL LETTER V WITH DOT BELOW */ \
+	  0x1E7F, /* LATIN SMALL LETTER V WITH DOT BELOW */ }, \
+	{ 0x01B2, /* LATIN CAPITAL LETTER V WITH HOOK */ \
+	  0x028B, /* LATIN SMALL LETTER V WITH HOOK */ }, \
+	{ 0x1E7C, /* LATIN CAPITAL LETTER V WITH TILDE */ \
+	  0x1E7D, /* LATIN SMALL LETTER V WITH TILDE */ }, \
+	{ 0xA768, /* LATIN CAPITAL LETTER VEND */ \
+	  0xA769, /* LATIN SMALL LETTER VEND */ }, \
+	{ 0xA762, /* LATIN CAPITAL LETTER VISIGOTHIC Z */ \
+	  0xA763, /* LATIN SMALL LETTER VISIGOTHIC Z */ }, \
+	{ 0xA79A, /* LATIN CAPITAL LETTER VOLAPUK AE */ \
+	  0xA79B, /* LATIN SMALL LETTER VOLAPUK AE */ }, \
+	{ 0xA79C, /* LATIN CAPITAL LETTER VOLAPUK OE */ \
+	  0xA79D, /* LATIN SMALL LETTER VOLAPUK OE */ }, \
+	{ 0xA79E, /* LATIN CAPITAL LETTER VOLAPUK UE */ \
+	  0xA79F, /* LATIN SMALL LETTER VOLAPUK UE */ }, \
+	{ 0xA760, /* LATIN CAPITAL LETTER VY */ \
+	  0xA761, /* LATIN SMALL LETTER VY */ }, \
+	{ 0x0057, /* LATIN CAPITAL LETTER W */ \
+	  0x0077, /* LATIN SMALL LETTER W */ }, \
+	{ 0x1E82, /* LATIN CAPITAL LETTER W WITH ACUTE */ \
+	  0x1E83, /* LATIN SMALL LETTER W WITH ACUTE */ }, \
+	{ 0x0174, /* LATIN CAPITAL LETTER W WITH CIRCUMFLEX */ \
+	  0x0175, /* LATIN SMALL LETTER W WITH CIRCUMFLEX */ }, \
+	{ 0x1E84, /* LATIN CAPITAL LETTER W WITH DIAERESIS */ \
+	  0x1E85, /* LATIN SMALL LETTER W WITH DIAERESIS */ }, \
+	{ 0x1E86, /* LATIN CAPITAL LETTER W WITH DOT ABOVE */ \
+	  0x1E87, /* LATIN SMALL LETTER W WITH DOT ABOVE */ }, \
+	{ 0x1E88, /* LATIN CAPITAL LETTER W WITH DOT BELOW */ \
+	  0x1E89, /* LATIN SMALL LETTER W WITH DOT BELOW */ }, \
+	{ 0x1E80, /* LATIN CAPITAL LETTER W WITH GRAVE */ \
+	  0x1E81, /* LATIN SMALL LETTER W WITH GRAVE */ }, \
+	{ 0x2C72, /* LATIN CAPITAL LETTER W WITH HOOK */ \
+	  0x2C73, /* LATIN SMALL LETTER W WITH HOOK */ }, \
+	{ 0x0058, /* LATIN CAPITAL LETTER X */ \
+	  0x0078, /* LATIN SMALL LETTER X */ }, \
+	{ 0x1E8C, /* LATIN CAPITAL LETTER X WITH DIAERESIS */ \
+	  0x1E8D, /* LATIN SMALL LETTER X WITH DIAERESIS */ }, \
+	{ 0x1E8A, /* LATIN CAPITAL LETTER X WITH DOT ABOVE */ \
+	  0x1E8B, /* LATIN SMALL LETTER X WITH DOT ABOVE */ }, \
+	{ 0x0059, /* LATIN CAPITAL LETTER Y */ \
+	  0x0079, /* LATIN SMALL LETTER Y */ }, \
+	{ 0x00DD, /* LATIN CAPITAL LETTER Y WITH ACUTE */ \
+	  0x00FD, /* LATIN SMALL LETTER Y WITH ACUTE */ }, \
+	{ 0x0176, /* LATIN CAPITAL LETTER Y WITH CIRCUMFLEX */ \
+	  0x0177, /* LATIN SMALL LETTER Y WITH CIRCUMFLEX */ }, \
+	{ 0x0178, /* LATIN CAPITAL LETTER Y WITH DIAERESIS */ \
+	  0x00FF, /* LATIN SMALL LETTER Y WITH DIAERESIS */ }, \
+	{ 0x1E8E, /* LATIN CAPITAL LETTER Y WITH DOT ABOVE */ \
+	  0x1E8F, /* LATIN SMALL LETTER Y WITH DOT ABOVE */ }, \
+	{ 0x1EF4, /* LATIN CAPITAL LETTER Y WITH DOT BELOW */ \
+	  0x1EF5, /* LATIN SMALL LETTER Y WITH DOT BELOW */ }, \
+	{ 0x1EF2, /* LATIN CAPITAL LETTER Y WITH GRAVE */ \
+	  0x1EF3, /* LATIN SMALL LETTER Y WITH GRAVE */ }, \
+	{ 0x01B3, /* LATIN CAPITAL LETTER Y WITH HOOK */ \
+	  0x01B4, /* LATIN SMALL LETTER Y WITH HOOK */ }, \
+	{ 0x1EF6, /* LATIN CAPITAL LETTER Y WITH HOOK ABOVE */ \
+	  0x1EF7, /* LATIN SMALL LETTER Y WITH HOOK ABOVE */ }, \
+	{ 0x1EFE, /* LATIN CAPITAL LETTER Y WITH LOOP */ \
+	  0x1EFF, /* LATIN SMALL LETTER Y WITH LOOP */ }, \
+	{ 0x0232, /* LATIN CAPITAL LETTER Y WITH MACRON */ \
+	  0x0233, /* LATIN SMALL LETTER Y WITH MACRON */ }, \
+	{ 0x024E, /* LATIN CAPITAL LETTER Y WITH STROKE */ \
+	  0x024F, /* LATIN SMALL LETTER Y WITH STROKE */ }, \
+	{ 0x1EF8, /* LATIN CAPITAL LETTER Y WITH TILDE */ \
+	  0x1EF9, /* LATIN SMALL LETTER Y WITH TILDE */ }, \
+	{ 0x021C, /* LATIN CAPITAL LETTER YOGH */ \
+	  0x021D, /* LATIN SMALL LETTER YOGH */ }, \
+	{ 0x005A, /* LATIN CAPITAL LETTER Z */ \
+	  0x007A, /* LATIN SMALL LETTER Z */ }, \
+	{ 0x0179, /* LATIN CAPITAL LETTER Z WITH ACUTE */ \
+	  0x017A, /* LATIN SMALL LETTER Z WITH ACUTE */ }, \
+	{ 0x017D, /* LATIN CAPITAL LETTER Z WITH CARON */ \
+	  0x017E, /* LATIN SMALL LETTER Z WITH CARON */ }, \
+	{ 0x1E90, /* LATIN CAPITAL LETTER Z WITH CIRCUMFLEX */ \
+	  0x1E91, /* LATIN SMALL LETTER Z WITH CIRCUMFLEX */ }, \
+	{ 0x2C6B, /* LATIN CAPITAL LETTER Z WITH DESCENDER */ \
+	  0x2C6C, /* LATIN SMALL LETTER Z WITH DESCENDER */ }, \
+	{ 0x017B, /* LATIN CAPITAL LETTER Z WITH DOT ABOVE */ \
+	  0x017C, /* LATIN SMALL LETTER Z WITH DOT ABOVE */ }, \
+	{ 0x1E92, /* LATIN CAPITAL LETTER Z WITH DOT BELOW */ \
+	  0x1E93, /* LATIN SMALL LETTER Z WITH DOT BELOW */ }, \
+	{ 0x0224, /* LATIN CAPITAL LETTER Z WITH HOOK */ \
+	  0x0225, /* LATIN SMALL LETTER Z WITH HOOK */ }, \
+	{ 0x1E94, /* LATIN CAPITAL LETTER Z WITH LINE BELOW */ \
+	  0x1E95, /* LATIN SMALL LETTER Z WITH LINE BELOW */ }, \
+	{ 0x01B5, /* LATIN CAPITAL LETTER Z WITH STROKE */ \
+	  0x01B6, /* LATIN SMALL LETTER Z WITH STROKE */ }, \
+	{ 0x2C7F, /* LATIN CAPITAL LETTER Z WITH SWASH TAIL */ \
+	  0x0240, /* LATIN SMALL LETTER Z WITH SWASH TAIL */ }, \
+	{ 0x0000, /* END OF LIST CAPITAL LETTERS */ \
+	  0x0000, /* END OF LIST SMALL LETTERS */ }, \
+}
+
+/*
+ * Correspondence table for small and capital letters of codepage 437.
+ * Letters A-Z are handled in code.
+ */
+#define CP437_CAPITALIZATION_TABLE { \
+	{ 0x00c4, 0x00e4, }, \
+	{ 0x00c5, 0x00e5, }, \
+	{ 0x00c6, 0x00e6, }, \
+	{ 0x00c7, 0x00e7, }, \
+	{ 0x00c9, 0x00e9, }, \
+	{ 0x00d1, 0x00f1, }, \
+	{ 0x00d6, 0x00f6, }, \
+	{ 0x00dc, 0x00fc, }, \
+	{ 0x03a3, 0x03c3, }, \
+	{ 0x03a6, 0x03c6, }, \
+	{ 0x0000, 0x0000, }, \
+}
+
+/*
+ * Correspondence table for small and capital letters of codepage 1250.
+ * Letters A-Z are handled in code.
+ */
+#define CP1250_CAPITALIZATION_TABLE { \
+	{ 0x00c1, 0x00e1, }, \
+	{ 0x00c2, 0x00e2, }, \
+	{ 0x00c4, 0x00e4, }, \
+	{ 0x00c7, 0x00e7, }, \
+	{ 0x00c9, 0x00e9, }, \
+	{ 0x00cb, 0x00eb, }, \
+	{ 0x00cd, 0x00ed, }, \
+	{ 0x00ce, 0x00ee, }, \
+	{ 0x00d3, 0x00f3, }, \
+	{ 0x00d4, 0x00f4, }, \
+	{ 0x00d6, 0x00f6, }, \
+	{ 0x00da, 0x00fa, }, \
+	{ 0x00dc, 0x00fc, }, \
+	{ 0x00dd, 0x00fd, }, \
+	{ 0x0102, 0x0103, }, \
+	{ 0x0104, 0x0105, }, \
+	{ 0x0106, 0x0107, }, \
+	{ 0x010c, 0x010d, }, \
+	{ 0x010e, 0x010f, }, \
+	{ 0x0110, 0x0111, }, \
+	{ 0x0118, 0x0119, }, \
+	{ 0x011a, 0x011b, }, \
+	{ 0x0139, 0x013a, }, \
+	{ 0x013d, 0x013e, }, \
+	{ 0x0141, 0x0142, }, \
+	{ 0x0143, 0x0144, }, \
+	{ 0x0147, 0x0148, }, \
+	{ 0x0150, 0x0151, }, \
+	{ 0x0154, 0x0155, }, \
+	{ 0x0158, 0x0159, }, \
+	{ 0x015a, 0x015b, }, \
+	{ 0x015e, 0x015f, }, \
+	{ 0x0160, 0x0161, }, \
+	{ 0x0162, 0x0163, }, \
+	{ 0x0164, 0x0165, }, \
+	{ 0x016e, 0x016f, }, \
+	{ 0x0170, 0x0171, }, \
+	{ 0x0179, 0x017a, }, \
+	{ 0x017b, 0x017c, }, \
+	{ 0x017d, 0x017e, }, \
+	{ 0x0000, 0x0000, }, \
+}
-- 
2.18.0

[U-Boot] [PATCH v3 1/5] efi_loader: capitalization table

[Alexander Graf]

On 03.09.18 18:35, Heinrich Schuchardt wrote:
> This patch provides a define to initialize a table that maps lower to
> capital letters for Unicode code point 0x0000 - 0xffff.
> 
> Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
> ---
> v3:
> 	remove codes < 0x80 from tables for code pages 437 and 1250
> 	sort tables for code pages 437 and 1250
> v2:
> 	add shorter tables for code pages 437 and 1250
> ---
>  MAINTAINERS              |    1 +
>  include/capitalization.h | 1978 ++++++++++++++++++++++++++++++++++++++
>  2 files changed, 1979 insertions(+)
>  create mode 100644 include/capitalization.h
> 


[...]

> +/*
> + * Correspondence table for small and capital letters of codepage 437.
> + * Letters A-Z are handled in code.
> + */
> +#define CP437_CAPITALIZATION_TABLE { \
> +	{ 0x00c4, 0x00e4, }, \

Please keep the comments that indicate which character these are. Makes
the list much more appealing :) and more obvious.


Alex

> +	{ 0x00c5, 0x00e5, }, \
> +	{ 0x00c6, 0x00e6, }, \
> +	{ 0x00c7, 0x00e7, }, \
> +	{ 0x00c9, 0x00e9, }, \
> +	{ 0x00d1, 0x00f1, }, \
> +	{ 0x00d6, 0x00f6, }, \
> +	{ 0x00dc, 0x00fc, }, \
> +	{ 0x03a3, 0x03c3, }, \
> +	{ 0x03a6, 0x03c6, }, \
> +	{ 0x0000, 0x0000, }, \
> +}
> +
> +/*
> + * Correspondence table for small and capital letters of codepage 1250.
> + * Letters A-Z are handled in code.
> + */
> +#define CP1250_CAPITALIZATION_TABLE { \
> +	{ 0x00c1, 0x00e1, }, \
> +	{ 0x00c2, 0x00e2, }, \
> +	{ 0x00c4, 0x00e4, }, \
> +	{ 0x00c7, 0x00e7, }, \
> +	{ 0x00c9, 0x00e9, }, \
> +	{ 0x00cb, 0x00eb, }, \
> +	{ 0x00cd, 0x00ed, }, \
> +	{ 0x00ce, 0x00ee, }, \
> +	{ 0x00d3, 0x00f3, }, \
> +	{ 0x00d4, 0x00f4, }, \
> +	{ 0x00d6, 0x00f6, }, \
> +	{ 0x00da, 0x00fa, }, \
> +	{ 0x00dc, 0x00fc, }, \
> +	{ 0x00dd, 0x00fd, }, \
> +	{ 0x0102, 0x0103, }, \
> +	{ 0x0104, 0x0105, }, \
> +	{ 0x0106, 0x0107, }, \
> +	{ 0x010c, 0x010d, }, \
> +	{ 0x010e, 0x010f, }, \
> +	{ 0x0110, 0x0111, }, \
> +	{ 0x0118, 0x0119, }, \
> +	{ 0x011a, 0x011b, }, \
> +	{ 0x0139, 0x013a, }, \
> +	{ 0x013d, 0x013e, }, \
> +	{ 0x0141, 0x0142, }, \
> +	{ 0x0143, 0x0144, }, \
> +	{ 0x0147, 0x0148, }, \
> +	{ 0x0150, 0x0151, }, \
> +	{ 0x0154, 0x0155, }, \
> +	{ 0x0158, 0x0159, }, \
> +	{ 0x015a, 0x015b, }, \
> +	{ 0x015e, 0x015f, }, \
> +	{ 0x0160, 0x0161, }, \
> +	{ 0x0162, 0x0163, }, \
> +	{ 0x0164, 0x0165, }, \
> +	{ 0x016e, 0x016f, }, \
> +	{ 0x0170, 0x0171, }, \
> +	{ 0x0179, 0x017a, }, \
> +	{ 0x017b, 0x017c, }, \
> +	{ 0x017d, 0x017e, }, \
> +	{ 0x0000, 0x0000, }, \
> +}
> 

[U-Boot] [PATCH v3 2/5] lib: charset: upper/lower case conversion

[Heinrich Schuchardt]

Provide functions for upper and lower case conversion.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
---
v3:
	no change
v2:
	add configuration switch EFI_UNICODE_CAPITALIZATION
---
 include/charset.h      | 16 ++++++++++++++
 lib/charset.c          | 47 ++++++++++++++++++++++++++++++++++++++++++
 lib/efi_loader/Kconfig | 10 +++++++++
 3 files changed, 73 insertions(+)

diff --git a/include/charset.h b/include/charset.h
index 90870513ef..686db5a1fe 100644
--- a/include/charset.h
+++ b/include/charset.h
@@ -142,6 +142,22 @@ int utf16_utf8_strncpy(char **dst, const u16 *src, size_t count);
  */
 #define utf16_utf8_strcpy(d, s) utf16_utf8_strncpy((d), (s), SIZE_MAX)
 
+/**
+ * utf_to_lower() - convert a Unicode letter to lower case
+ *
+ * @code:		letter to convert
+ * Return:		lower case letter or unchanged letter
+ */
+s32 utf_to_lower(const s32 code);
+
+/**
+ * utf_to_upper() - convert a Unicode letter to upper case
+ *
+ * @code:		letter to convert
+ * Return:		upper case letter or unchanged letter
+ */
+s32 utf_to_upper(const s32 code);
+
 /**
  * u16_strlen - count non-zero words
  *
diff --git a/lib/charset.c b/lib/charset.c
index 39c8329830..72c808ce64 100644
--- a/lib/charset.c
+++ b/lib/charset.c
@@ -6,8 +6,18 @@
  */
 
 #include <charset.h>
+#include <capitalization.h>
 #include <malloc.h>
 
+static struct capitalization_table capitalization_table[] =
+#ifdef CONFIG_EFI_UNICODE_CAPITALIZATION
+	UNICODE_CAPITALIZATION_TABLE;
+#elif CONFIG_FAT_DEFAULT_CODEPAGE == 1250
+	CP1250_CAPITALIZATION_TABLE;
+#else
+	CP437_CAPITALIZATION_TABLE;
+#endif
+
 s32 utf8_get(const char **src)
 {
 	s32 code = 0;
@@ -241,6 +251,43 @@ int utf16_utf8_strncpy(char **dst, const u16 *src, size_t count)
 	return 0;
 }
 
+s32 utf_to_lower(const s32 code)
+{
+	struct capitalization_table *pos = capitalization_table;
+	s32 ret = code;
+
+	if (code <= 0x7f) {
+		if (code >= 'A' && code <= 'Z')
+			ret += 0x20;
+		return ret;
+	}
+	for (; pos->upper; ++pos) {
+		if (pos->upper == code) {
+			ret = pos->lower;
+			break;
+		}
+	}
+	return ret;
+}
+
+s32 utf_to_upper(const s32 code)
+{
+	struct capitalization_table *pos = capitalization_table;
+	s32 ret = code;
+
+	if (code <= 0x7f) {
+		if (code >= 'a' && code <= 'z')
+			ret -= 0x20;
+		return ret;
+	}
+	for (; pos->lower; ++pos) {
+		if (pos->lower == code) {
+			ret = pos->upper;
+			break;
+		}
+	}
+	return ret;
+}
 
 size_t u16_strlen(const u16 *in)
 {
diff --git a/lib/efi_loader/Kconfig b/lib/efi_loader/Kconfig
index ce6a09f0b4..dbf7339fe0 100644
--- a/lib/efi_loader/Kconfig
+++ b/lib/efi_loader/Kconfig
@@ -15,6 +15,16 @@ config EFI_LOADER
 	  interfaces to a loaded EFI application, enabling it to reuse U-Boot's
 	  device drivers.
 
+config EFI_UNICODE_CAPITALIZATION
+	bool "Support Unicode capitalization"
+	depends on EFI_LOADER
+	default y
+	help
+	  Select this option to enable correct handling of the capitalization of
+	  Unicode codepoints in the range 0x0000-0xffff. If this option is not
+	  set, only the the correct handling of the letters of the codepage
+	  used by the FAT file system is ensured.
+
 config EFI_LOADER_BOUNCE_BUFFER
 	bool "EFI Applications use bounce buffers for DMA operations"
 	depends on EFI_LOADER && ARM64
-- 
2.18.0

[U-Boot] [PATCH v3 3/5] test: tests for utf_to_lower() utf_to_upper().

[Heinrich Schuchardt]

Provide unit tests for utf_to_lower() utf_to_upper().

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
---
v3:
	add test for latin letter O with diaresis
v2:
	use ut_assert*() for testing
---
 test/unicode_ut.c | 34 ++++++++++++++++++++++++++++++++++
 1 file changed, 34 insertions(+)

diff --git a/test/unicode_ut.c b/test/unicode_ut.c
index d0dbd7985a..b94b4a651f 100644
--- a/test/unicode_ut.c
+++ b/test/unicode_ut.c
@@ -500,6 +500,40 @@ int ut_utf16_utf8_strncpy(struct unit_test_state *uts)
 }
 UNICODE_TEST(ut_utf16_utf8_strncpy);
 
+int ut_utf_to_lower(struct unit_test_state *uts)
+{
+	ut_asserteq('@', utf_to_lower('@'));
+	ut_asserteq('a', utf_to_lower('A'));
+	ut_asserteq('z', utf_to_lower('Z'));
+	ut_asserteq('[', utf_to_lower('['));
+	ut_asserteq('m', utf_to_lower('m'));
+	/* Latin letter O with diaresis (umlaut) */
+	ut_asserteq(0x00f6, utf_to_lower(0x00d6));
+#ifdef CONFIG_EFI_UNICODE_CAPITALIZATION
+	/* Cyrillic letter I*/
+	ut_asserteq(0x0438, utf_to_lower(0x0418));
+#endif
+	return 0;
+}
+UNICODE_TEST(ut_utf_to_lower);
+
+int ut_utf_to_upper(struct unit_test_state *uts)
+{
+	ut_asserteq('`', utf_to_upper('`'));
+	ut_asserteq('A', utf_to_upper('a'));
+	ut_asserteq('Z', utf_to_upper('z'));
+	ut_asserteq('{', utf_to_upper('{'));
+	ut_asserteq('M', utf_to_upper('M'));
+	/* Latin letter O with diaresis (umlaut) */
+	ut_asserteq(0x00d6, utf_to_upper(0x00f6));
+#ifdef CONFIG_EFI_UNICODE_CAPITALIZATION
+	/* Cyrillic letter I */
+	ut_asserteq(0x0418, utf_to_upper(0x0438));
+#endif
+	return 0;
+}
+UNICODE_TEST(ut_utf_to_upper);
+
 int do_ut_unicode(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
 {
 	struct unit_test *tests = ll_entry_start(struct unit_test, unicode_test);
-- 
2.18.0

[U-Boot] [PATCH v3 4/5] efi_loader: EFI_UNICODE_COLLATION_PROTOCOL

[Heinrich Schuchardt]

The patch implements the EFI_UNICODE_COLLATION_PROTOCOL.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
---
v3:
	no change
v2:
	add more comments
---
 MAINTAINERS                            |   2 +
 include/cp1250.h                       |  40 +++
 include/cp437.h                        |  40 +++
 include/efi_api.h                      |  21 ++
 include/efi_loader.h                   |   5 +
 lib/efi_loader/Makefile                |  18 +-
 lib/efi_loader/efi_boottime.c          |   6 +
 lib/efi_loader/efi_unicode_collation.c | 329 +++++++++++++++++++++++++
 8 files changed, 457 insertions(+), 4 deletions(-)
 create mode 100644 include/cp1250.h
 create mode 100644 include/cp437.h
 create mode 100644 lib/efi_loader/efi_unicode_collation.c

diff --git a/MAINTAINERS b/MAINTAINERS
index 8c9cd83347..67b015205c 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -382,6 +382,8 @@ F:	doc/README.uefi
 F:	doc/README.iscsi
 F:	Documentation/efi.rst
 F:	include/capitalization.h
+F:	include/cp1250.h
+F:	include/cp437.h
 F:	include/efi*
 F:	include/pe.h
 F:	include/asm-generic/pe.h
diff --git a/include/cp1250.h b/include/cp1250.h
new file mode 100644
index 0000000000..adacf8a958
--- /dev/null
+++ b/include/cp1250.h
@@ -0,0 +1,40 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+
+/*
+ * Constant CP1250 contains the Unicode code points for characters 0x80 - 0xff
+ * of the code page 1250.
+ */
+#define CP1250 { \
+	0x20ac, 0x0000, 0x201a, 0x0000, \
+	0x201e, 0x2026, 0x2020, 0x2021, \
+	0x0000, 0x2030, 0x0160, 0x2039, \
+	0x015a, 0x0164, 0x017d, 0x0179, \
+	0x0000, 0x2018, 0x2019, 0x201c, \
+	0x201d, 0x2022, 0x2013, 0x2014, \
+	0x0000, 0x2122, 0x0161, 0x203a, \
+	0x015b, 0x0165, 0x017e, 0x017a, \
+	0x00a0, 0x02c7, 0x02d8, 0x0141, \
+	0x00a4, 0x0104, 0x00a6, 0x00a7, \
+	0x00a8, 0x00a9, 0x015e, 0x00ab, \
+	0x00ac, 0x00ad, 0x00ae, 0x017b, \
+	0x00b0, 0x00b1, 0x02db, 0x0142, \
+	0x00b4, 0x00b5, 0x00b6, 0x00b7, \
+	0x00b8, 0x0105, 0x015f, 0x00bb, \
+	0x013d, 0x02dd, 0x013e, 0x017c, \
+	0x0154, 0x00c1, 0x00c2, 0x0102, \
+	0x00c4, 0x0139, 0x0106, 0x00c7, \
+	0x010c, 0x00c9, 0x0118, 0x00cb, \
+	0x011a, 0x00cd, 0x00ce, 0x010e, \
+	0x0110, 0x0143, 0x0147, 0x00d3, \
+	0x00d4, 0x0150, 0x00d6, 0x00d7, \
+	0x0158, 0x016e, 0x00da, 0x0170, \
+	0x00dc, 0x00dd, 0x0162, 0x00df, \
+	0x0155, 0x00e1, 0x00e2, 0x0103, \
+	0x00e4, 0x013a, 0x0107, 0x00e7, \
+	0x010d, 0x00e9, 0x0119, 0x00eb, \
+	0x011b, 0x00ed, 0x00ee, 0x010f, \
+	0x0111, 0x0144, 0x0148, 0x00f3, \
+	0x00f4, 0x0151, 0x00f6, 0x00f7, \
+	0x0159, 0x016f, 0x00fa, 0x0171, \
+	0x00fc, 0x00fd, 0x0163, 0x02d9, \
+}
diff --git a/include/cp437.h b/include/cp437.h
new file mode 100644
index 0000000000..0b2b97132e
--- /dev/null
+++ b/include/cp437.h
@@ -0,0 +1,40 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+
+/*
+ * Constant CP437 contains the Unicode code points for characters 0x80 - 0xff
+ * of the code page 437.
+ */
+#define CP437 { \
+	0x00c7, 0x00fc, 0x00e9, 0x00e2, \
+	0x00e4, 0x00e0, 0x00e5, 0x00e7, \
+	0x00ea, 0x00eb, 0x00e8, 0x00ef, \
+	0x00ee, 0x00ec, 0x00c4, 0x00c5, \
+	0x00c9, 0x00e6, 0x00c6, 0x00f4, \
+	0x00f6, 0x00f2, 0x00fb, 0x00f9, \
+	0x00ff, 0x00d6, 0x00dc, 0x00a2, \
+	0x00a3, 0x00a5, 0x20a7, 0x0192, \
+	0x00e1, 0x00ed, 0x00f3, 0x00fa, \
+	0x00f1, 0x00d1, 0x00aa, 0x00ba, \
+	0x00bf, 0x2310, 0x00ac, 0x00bd, \
+	0x00bc, 0x00a1, 0x00ab, 0x00bb, \
+	0x2591, 0x2592, 0x2593, 0x2502, \
+	0x2524, 0x2561, 0x2562, 0x2556, \
+	0x2555, 0x2563, 0x2551, 0x2557, \
+	0x255d, 0x255c, 0x255b, 0x2510, \
+	0x2514, 0x2534, 0x252c, 0x251c, \
+	0x2500, 0x253c, 0x255e, 0x255f, \
+	0x255a, 0x2554, 0x2569, 0x2566, \
+	0x2560, 0x2550, 0x256c, 0x2567, \
+	0x2568, 0x2564, 0x2565, 0x2559, \
+	0x2558, 0x2552, 0x2553, 0x256b, \
+	0x256a, 0x2518, 0x250c, 0x2588, \
+	0x2584, 0x258c, 0x2590, 0x2580, \
+	0x03b1, 0x00df, 0x0393, 0x03c0, \
+	0x03a3, 0x03c3, 0x00b5, 0x03c4, \
+	0x03a6, 0x0398, 0x03a9, 0x03b4, \
+	0x221e, 0x03c6, 0x03b5, 0x2229, \
+	0x2261, 0x00b1, 0x2265, 0x2264, \
+	0x2320, 0x2321, 0x00f7, 0x2248, \
+	0x00b0, 0x2219, 0x00b7, 0x221a, \
+	0x207f, 0x00b2, 0x25a0, 0x00a0, \
+}
diff --git a/include/efi_api.h b/include/efi_api.h
index ebf2a3bc18..1efc448184 100644
--- a/include/efi_api.h
+++ b/include/efi_api.h
@@ -31,6 +31,7 @@ enum efi_timer_delay {
 	EFI_TIMER_RELATIVE = 2
 };
 
+#define efi_intn_t ssize_t
 #define efi_uintn_t size_t
 typedef uint16_t *efi_string_t;
 
@@ -958,4 +959,24 @@ struct efi_driver_binding_protocol {
 	efi_handle_t driver_binding_handle;
 };
 
+#define EFI_UNICODE_COLLATION_PROTOCOL2_GUID \
+	EFI_GUID(0xa4c751fc, 0x23ae, 0x4c3e, \
+		 0x92, 0xe9, 0x49, 0x64, 0xcf, 0x63, 0xf3, 0x49)
+struct efi_unicode_collation_protocol {
+	efi_intn_t (EFIAPI *stri_coll)(
+		struct efi_unicode_collation_protocol *this, u16 *s1, u16 *s2);
+	bool (EFIAPI *metai_match)(struct efi_unicode_collation_protocol *this,
+				   const u16 *string, const u16 *patter);
+	void (EFIAPI *str_lwr)(struct efi_unicode_collation_protocol
+			       *this, u16 *string);
+	void (EFIAPI *str_upr)(struct efi_unicode_collation_protocol *this,
+			       u16 *string);
+	void (EFIAPI *fat_to_str)(struct efi_unicode_collation_protocol *this,
+				  efi_uintn_t fat_size, char *fat, u16 *string);
+	bool (EFIAPI *str_to_fat)(struct efi_unicode_collation_protocol *this,
+				  const u16 *string, efi_uintn_t fat_size,
+				  char *fat);
+	char *supported_languages;
+};
+
 #endif
diff --git a/include/efi_loader.h b/include/efi_loader.h
index b46babf931..736e82c506 100644
--- a/include/efi_loader.h
+++ b/include/efi_loader.h
@@ -98,6 +98,9 @@ extern const struct efi_device_path_to_text_protocol efi_device_path_to_text;
 /* implementation of the EFI_DEVICE_PATH_UTILITIES_PROTOCOL */
 extern const struct efi_device_path_utilities_protocol
 					efi_device_path_utilities;
+/* Implementation of the EFI_UNICODE_COLLATION_PROTOCOL */
+extern const struct efi_unicode_collation_protocol
+					efi_unicode_collation_protocol;
 
 uint16_t *efi_dp_str(struct efi_device_path *dp);
 
@@ -127,6 +130,8 @@ extern const efi_guid_t efi_file_info_guid;
 /* GUID for file system information */
 extern const efi_guid_t efi_file_system_info_guid;
 extern const efi_guid_t efi_guid_device_path_utilities_protocol;
+/* GUID of the Unicode collation protocol */
+extern const efi_guid_t efi_guid_unicode_collation_protocol;
 
 extern unsigned int __efi_runtime_start, __efi_runtime_stop;
 extern unsigned int __efi_runtime_rel_start, __efi_runtime_rel_stop;
diff --git a/lib/efi_loader/Makefile b/lib/efi_loader/Makefile
index 1ffbf52a89..0769df20f7 100644
--- a/lib/efi_loader/Makefile
+++ b/lib/efi_loader/Makefile
@@ -17,10 +17,20 @@ always += helloworld.efi
 endif
 
 obj-$(CONFIG_CMD_BOOTEFI_HELLO) += helloworld_efi.o
-obj-y += efi_image_loader.o efi_boottime.o efi_runtime.o efi_console.o
-obj-y += efi_memory.o efi_device_path_to_text.o efi_device_path.o
-obj-y += efi_device_path_utilities.o efi_file.o efi_variable.o efi_bootmgr.o
-obj-y += efi_watchdog.o
+obj-y += \
+efi_bootmgr.o \
+efi_boottime.o \
+efi_console.o \
+efi_device_path.o \
+efi_device_path_to_text.o \
+efi_device_path_utilities.o \
+efi_file.o \
+efi_image_loader.o \
+efi_memory.o \
+efi_runtime.o \
+efi_unicode_collation.o \
+efi_variable.o \
+efi_watchdog.o
 obj-$(CONFIG_LCD) += efi_gop.o
 obj-$(CONFIG_DM_VIDEO) += efi_gop.o
 obj-$(CONFIG_PARTITIONS) += efi_disk.o
diff --git a/lib/efi_loader/efi_boottime.c b/lib/efi_loader/efi_boottime.c
index cd5fdb7aa8..f9f59b93f9 100644
--- a/lib/efi_loader/efi_boottime.c
+++ b/lib/efi_loader/efi_boottime.c
@@ -1527,6 +1527,12 @@ efi_status_t efi_setup_loaded_image(
 	if (ret != EFI_SUCCESS)
 		goto failure;
 
+	ret = efi_add_protocol(obj->handle,
+			       &efi_guid_unicode_collation_protocol,
+			       (void *)&efi_unicode_collation_protocol);
+	if (ret != EFI_SUCCESS)
+		goto failure;
+
 	return ret;
 failure:
 	printf("ERROR: Failure to install protocols for loaded image\n");
diff --git a/lib/efi_loader/efi_unicode_collation.c b/lib/efi_loader/efi_unicode_collation.c
new file mode 100644
index 0000000000..2c1e819174
--- /dev/null
+++ b/lib/efi_loader/efi_unicode_collation.c
@@ -0,0 +1,329 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * EFI Unicode collation protocol
+ *
+ * Copyright (c) 2018 Heinrich Schuchardt <xypron.glpk@gmx.de>
+ */
+
+#include <common.h>
+#include <charset.h>
+#include <cp1250.h>
+#include <cp437.h>
+#include <efi_loader.h>
+
+/* Characters that may not be used in file names */
+static const char illegal[] = "<>:\"/\\|?*";
+
+/*
+ * EDK2 assumes codepage 1250 when creating FAT 8.3 file names.
+ * Linux defaults to codepage 437 for FAT 8.3 file names.
+ */
+#if CONFIG_FAT_DEFAULT_CODEPAGE == 1250
+/* Unicode code points for code page 1250 characters 0x80 - 0xff */
+static const u16 codepage[] = CP1250;
+#else
+/* Unicode code points for code page 437 characters 0x80 - 0xff */
+static const u16 codepage[] = CP437;
+#endif
+
+/* GUID of the EFI_UNICODE_COLLATION_PROTOCOL */
+const efi_guid_t efi_guid_unicode_collation_protocol =
+	EFI_UNICODE_COLLATION_PROTOCOL2_GUID;
+
+/**
+ * efi_stri_coll() - compare utf-16 strings case-insenitively
+ *
+ * @this:	unicode collation protocol instance
+ * @s1:		first string
+ * @s2:		second string
+ *
+ * This function implements the StriColl() service of the
+ * EFI_UNICODE_COLLATION_PROTOCOL.
+ *
+ * See the Unified Extensible Firmware Interface (UEFI) specification for
+ * details.
+ *
+ * TODO:
+ * The implementation does not follow the Unicode collation algorithm.
+ * For ASCII characters it results in the same sort order as EDK2.
+ * We could use table UNICODE_CAPITALIZATION_TABLE for better results.
+ *
+ * Return:	0: s1 == s2, > 0: s1 > s2, < 0: s1 < s2
+ */
+static efi_intn_t EFIAPI efi_stri_coll(
+		struct efi_unicode_collation_protocol *this, u16 *s1, u16 *s2)
+{
+	s32 c1, c2;
+	efi_intn_t ret = 0;
+
+	EFI_ENTRY("%p, %ls, %ls", this, s1, s2);
+	for (; *s1 | *s2; ++s1, ++s2) {
+		c1 = utf_to_upper(*s1);
+		c2 = utf_to_upper(*s2);
+		if (c1 < c2) {
+			ret = -1;
+			goto out;
+		} else if (c1 > c2) {
+			ret = 1;
+			goto out;
+		}
+	}
+out:
+	EFI_EXIT(EFI_SUCCESS);
+	return ret;
+}
+
+/**
+ * metai_match() - compare utf-16 string with a pattern string case-insenitively
+ *
+ * @s:		string to compare
+ * @p:		pattern string
+ *
+ * The pattern string may use these:
+ *	- * matches >= 0 characters
+ *	- ? matches 1 character
+ *	- [<char1><char2>...<charN>] match any character in the set
+ *	- [<char1>-<char2>] matches any character in the range
+ *
+ * This function is called my efi_metai_match().
+ *
+ * For '*' pattern searches this function calls itself recursively.
+ * Performance-wise this is suboptimal, especially for multiple '*' wildcards.
+ * But it results in simple code.
+ *
+ * Return:	true if the string is matched.
+ */
+static bool metai_match(const u16 *s, const u16 *p)
+{
+	u16 first;
+
+	for (; *s && *p; ++s, ++p) {
+		switch (*p) {
+		case '*':
+			/* Match 0 or more characters */
+			++p;
+			for (;; ++s) {
+				if (metai_match(s, p))
+					return true;
+				if (!*s)
+					return false;
+			}
+		case '?':
+			/* Match any one character */
+			break;
+		case '[':
+			/* Match any character in the set */
+			++p;
+			first = *p;
+			if (first == ']')
+				/* Empty set */
+				return false;
+			++p;
+			if (*p == '-') {
+				/* Range */
+				++p;
+				if (*s < first || *s > *p)
+					return false;
+				++p;
+				if (*p != ']')
+					return false;
+			} else {
+				/* Set */
+				bool hit = false;
+
+				if (*s == first)
+					hit = true;
+				for (; *p && *p != ']'; ++p) {
+					if (*p == *s)
+						hit = true;
+				}
+				if (!hit || *p != ']')
+					return false;
+			}
+			break;
+		default:
+			/* Match one character */
+			if (*p != *s)
+				return false;
+		}
+	}
+	if (!*p && !*s)
+		return true;
+	return false;
+}
+
+/**
+ * efi_metai_match() - compare utf-16 string with a pattern string
+ *		       case-insenitively
+ *
+ * @this:	unicode collation protocol instance
+ * @s:		string to compare
+ * @p:		pattern string
+ *
+ * The pattern string may use these:
+ *	- * matches >= 0 characters
+ *	- ? matches 1 character
+ *	- [<char1><char2>...<charN>] match any character in the set
+ *	- [<char1>-<char2>] matches any character in the range
+ *
+ * This function implements the MetaMatch() service of the
+ * EFI_UNICODE_COLLATION_PROTOCOL.
+ *
+ * Return:	true if the string is matched.
+ */
+static bool EFIAPI efi_metai_match(struct efi_unicode_collation_protocol *this,
+				   const u16 *string, const u16 *pattern)
+{
+	bool ret;
+
+	EFI_ENTRY("%p, %ls, %ls", this, string, pattern);
+	ret =  metai_match(string, pattern);
+	EFI_EXIT(EFI_SUCCESS);
+	return ret;
+}
+
+/**
+ * efi_str_lwr() - convert to lower case
+ *
+ * @this:	unicode collation protocol instance
+ * @string:	string to convert
+ * @p:		pattern string
+ *
+ * The conversion is done in place. As long as upper and lower letters use the
+ * same number of words this does not pose a problem.
+ *
+ * This function implements the StrLwr() service of the
+ * EFI_UNICODE_COLLATION_PROTOCOL.
+ */
+static void EFIAPI efi_str_lwr(struct efi_unicode_collation_protocol *this,
+			       u16 *string)
+{
+	EFI_ENTRY("%p, %ls", this, string);
+	for (; *string; ++string)
+		*string = utf_to_lower(*string);
+	EFI_EXIT(EFI_SUCCESS);
+}
+
+/**
+ * efi_str_upr() - convert to upper case
+ *
+ * @this:	unicode collation protocol instance
+ * @string:	string to convert
+ * @p:		pattern string
+ *
+ * The conversion is done in place. As long as upper and lower letters use the
+ * same number of words this does not pose a problem.
+ *
+ * This function implements the StrUpr() service of the
+ * EFI_UNICODE_COLLATION_PROTOCOL.
+ */
+static void EFIAPI efi_str_upr(struct efi_unicode_collation_protocol *this,
+			       u16 *string)
+{
+	EFI_ENTRY("%p, %ls", this, string);
+	for (; *string; ++string)
+		*string = utf_to_upper(*string);
+	EFI_EXIT(EFI_SUCCESS);
+}
+
+/**
+ * efi_fat_to_str() - convert an 8.3 file name from an OEM codepage to Unicode
+ *
+ * @this:	unicode collation protocol instance
+ * @fat_size:	size of the string to convert
+ * @fat:	string to convert
+ * @string:	converted string
+ *
+ * This function implements the FatToStr() service of the
+ * EFI_UNICODE_COLLATION_PROTOCOL.
+ */
+static void EFIAPI efi_fat_to_str(struct efi_unicode_collation_protocol *this,
+				  efi_uintn_t fat_size, char *fat, u16 *string)
+{
+	efi_uintn_t i;
+	u16 c;
+
+	EFI_ENTRY("%p, %zu, %s, %p", this, fat_size, fat, string);
+	for (i = 0; i < fat_size; ++i) {
+		c = (unsigned char)fat[i];
+		if (c > 0x80)
+			c = codepage[i - 0x80];
+		string[i] = c;
+		if (!c)
+			break;
+	}
+	string[i] = 0;
+	EFI_EXIT(EFI_SUCCESS);
+}
+
+/**
+ * efi_fat_to_str() - convert a utf-16 string to legal characters for a FAT
+ *                    file name in an OEM code page
+ *
+ * @this:	unicode collation protocol instance
+ * @string:	Unicode string to convert
+ * @fat_size:	size of the target buffer
+ * @fat:	converted string
+ *
+ * This function implements the StrToFat() service of the
+ * EFI_UNICODE_COLLATION_PROTOCOL.
+ *
+ * Return:	true if an illegal character was substituted by '_'.
+ */
+static bool EFIAPI efi_str_to_fat(struct efi_unicode_collation_protocol *this,
+				  const u16 *string, efi_uintn_t fat_size,
+				  char *fat)
+{
+	efi_uintn_t i;
+	s32 c;
+	bool ret = false;
+
+	EFI_ENTRY("%p, %ls, %zu, %p", this, string, fat_size, fat);
+	for (i = 0; i < fat_size;) {
+		c = utf16_get(&string);
+		switch (c) {
+		/* Ignore period and space */
+		case '.':
+		case ' ':
+			continue;
+		case 0:
+			break;
+		}
+		c = utf_to_upper(c);
+		if (c >= 0x80) {
+			int j;
+
+			/* Look for codepage translation */
+			for (j = 0; j < 0x80; ++j) {
+				if (c == codepage[j]) {
+					c = j + 0x80;
+					break;
+				}
+			}
+			if (j >= 0x80) {
+				c = '_';
+				ret = true;
+			}
+		} else if (c && (c < 0x20 || strchr(illegal, c))) {
+			c = '_';
+			ret = true;
+		}
+
+		fat[i] = c;
+		if (!c)
+			break;
+		++i;
+	}
+	EFI_EXIT(EFI_SUCCESS);
+	return ret;
+}
+
+const struct efi_unicode_collation_protocol efi_unicode_collation_protocol = {
+	.stri_coll = efi_stri_coll,
+	.metai_match = efi_metai_match,
+	.str_lwr = efi_str_lwr,
+	.str_upr = efi_str_upr,
+	.fat_to_str = efi_fat_to_str,
+	.str_to_fat = efi_str_to_fat,
+	.supported_languages = "en-US",
+};
-- 
2.18.0

[U-Boot] [PATCH v3 4/5] efi_loader: EFI_UNICODE_COLLATION_PROTOCOL

[Alexander Graf]

On 03.09.18 18:35, Heinrich Schuchardt wrote:
> The patch implements the EFI_UNICODE_COLLATION_PROTOCOL.
> 
> Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
> ---
> v3:
> 	no change
> v2:
> 	add more comments
> ---
>  MAINTAINERS                            |   2 +
>  include/cp1250.h                       |  40 +++
>  include/cp437.h                        |  40 +++
>  include/efi_api.h                      |  21 ++
>  include/efi_loader.h                   |   5 +
>  lib/efi_loader/Makefile                |  18 +-
>  lib/efi_loader/efi_boottime.c          |   6 +
>  lib/efi_loader/efi_unicode_collation.c | 329 +++++++++++++++++++++++++
>  8 files changed, 457 insertions(+), 4 deletions(-)
>  create mode 100644 include/cp1250.h
>  create mode 100644 include/cp437.h
>  create mode 100644 lib/efi_loader/efi_unicode_collation.c
> 
> diff --git a/MAINTAINERS b/MAINTAINERS
> index 8c9cd83347..67b015205c 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -382,6 +382,8 @@ F:	doc/README.uefi
>  F:	doc/README.iscsi
>  F:	Documentation/efi.rst
>  F:	include/capitalization.h
> +F:	include/cp1250.h
> +F:	include/cp437.h
>  F:	include/efi*
>  F:	include/pe.h
>  F:	include/asm-generic/pe.h
> diff --git a/include/cp1250.h b/include/cp1250.h
> new file mode 100644
> index 0000000000..adacf8a958
> --- /dev/null
> +++ b/include/cp1250.h
> @@ -0,0 +1,40 @@
> +/* SPDX-License-Identifier: GPL-2.0+ */
> +
> +/*
> + * Constant CP1250 contains the Unicode code points for characters 0x80 - 0xff
> + * of the code page 1250.
> + */
> +#define CP1250 { \
> +	0x20ac, 0x0000, 0x201a, 0x0000, \
> +	0x201e, 0x2026, 0x2020, 0x2021, \
> +	0x0000, 0x2030, 0x0160, 0x2039, \
> +	0x015a, 0x0164, 0x017d, 0x0179, \
> +	0x0000, 0x2018, 0x2019, 0x201c, \
> +	0x201d, 0x2022, 0x2013, 0x2014, \
> +	0x0000, 0x2122, 0x0161, 0x203a, \
> +	0x015b, 0x0165, 0x017e, 0x017a, \
> +	0x00a0, 0x02c7, 0x02d8, 0x0141, \
> +	0x00a4, 0x0104, 0x00a6, 0x00a7, \
> +	0x00a8, 0x00a9, 0x015e, 0x00ab, \
> +	0x00ac, 0x00ad, 0x00ae, 0x017b, \
> +	0x00b0, 0x00b1, 0x02db, 0x0142, \
> +	0x00b4, 0x00b5, 0x00b6, 0x00b7, \
> +	0x00b8, 0x0105, 0x015f, 0x00bb, \
> +	0x013d, 0x02dd, 0x013e, 0x017c, \
> +	0x0154, 0x00c1, 0x00c2, 0x0102, \
> +	0x00c4, 0x0139, 0x0106, 0x00c7, \
> +	0x010c, 0x00c9, 0x0118, 0x00cb, \
> +	0x011a, 0x00cd, 0x00ce, 0x010e, \
> +	0x0110, 0x0143, 0x0147, 0x00d3, \
> +	0x00d4, 0x0150, 0x00d6, 0x00d7, \
> +	0x0158, 0x016e, 0x00da, 0x0170, \
> +	0x00dc, 0x00dd, 0x0162, 0x00df, \
> +	0x0155, 0x00e1, 0x00e2, 0x0103, \
> +	0x00e4, 0x013a, 0x0107, 0x00e7, \
> +	0x010d, 0x00e9, 0x0119, 0x00eb, \
> +	0x011b, 0x00ed, 0x00ee, 0x010f, \
> +	0x0111, 0x0144, 0x0148, 0x00f3, \
> +	0x00f4, 0x0151, 0x00f6, 0x00f7, \
> +	0x0159, 0x016f, 0x00fa, 0x0171, \
> +	0x00fc, 0x00fd, 0x0163, 0x02d9, \
> +}
> diff --git a/include/cp437.h b/include/cp437.h
> new file mode 100644
> index 0000000000..0b2b97132e
> --- /dev/null
> +++ b/include/cp437.h
> @@ -0,0 +1,40 @@
> +/* SPDX-License-Identifier: GPL-2.0+ */
> +
> +/*
> + * Constant CP437 contains the Unicode code points for characters 0x80 - 0xff
> + * of the code page 437.
> + */
> +#define CP437 { \
> +	0x00c7, 0x00fc, 0x00e9, 0x00e2, \
> +	0x00e4, 0x00e0, 0x00e5, 0x00e7, \
> +	0x00ea, 0x00eb, 0x00e8, 0x00ef, \
> +	0x00ee, 0x00ec, 0x00c4, 0x00c5, \
> +	0x00c9, 0x00e6, 0x00c6, 0x00f4, \
> +	0x00f6, 0x00f2, 0x00fb, 0x00f9, \
> +	0x00ff, 0x00d6, 0x00dc, 0x00a2, \
> +	0x00a3, 0x00a5, 0x20a7, 0x0192, \
> +	0x00e1, 0x00ed, 0x00f3, 0x00fa, \
> +	0x00f1, 0x00d1, 0x00aa, 0x00ba, \
> +	0x00bf, 0x2310, 0x00ac, 0x00bd, \
> +	0x00bc, 0x00a1, 0x00ab, 0x00bb, \
> +	0x2591, 0x2592, 0x2593, 0x2502, \
> +	0x2524, 0x2561, 0x2562, 0x2556, \
> +	0x2555, 0x2563, 0x2551, 0x2557, \
> +	0x255d, 0x255c, 0x255b, 0x2510, \
> +	0x2514, 0x2534, 0x252c, 0x251c, \
> +	0x2500, 0x253c, 0x255e, 0x255f, \
> +	0x255a, 0x2554, 0x2569, 0x2566, \
> +	0x2560, 0x2550, 0x256c, 0x2567, \
> +	0x2568, 0x2564, 0x2565, 0x2559, \
> +	0x2558, 0x2552, 0x2553, 0x256b, \
> +	0x256a, 0x2518, 0x250c, 0x2588, \
> +	0x2584, 0x258c, 0x2590, 0x2580, \
> +	0x03b1, 0x00df, 0x0393, 0x03c0, \
> +	0x03a3, 0x03c3, 0x00b5, 0x03c4, \
> +	0x03a6, 0x0398, 0x03a9, 0x03b4, \
> +	0x221e, 0x03c6, 0x03b5, 0x2229, \
> +	0x2261, 0x00b1, 0x2265, 0x2264, \
> +	0x2320, 0x2321, 0x00f7, 0x2248, \
> +	0x00b0, 0x2219, 0x00b7, 0x221a, \
> +	0x207f, 0x00b2, 0x25a0, 0x00a0, \
> +}
> diff --git a/include/efi_api.h b/include/efi_api.h
> index ebf2a3bc18..1efc448184 100644
> --- a/include/efi_api.h
> +++ b/include/efi_api.h
> @@ -31,6 +31,7 @@ enum efi_timer_delay {
>  	EFI_TIMER_RELATIVE = 2
>  };
>  
> +#define efi_intn_t ssize_t
>  #define efi_uintn_t size_t
>  typedef uint16_t *efi_string_t;
>  
> @@ -958,4 +959,24 @@ struct efi_driver_binding_protocol {
>  	efi_handle_t driver_binding_handle;
>  };
>  
> +#define EFI_UNICODE_COLLATION_PROTOCOL2_GUID \
> +	EFI_GUID(0xa4c751fc, 0x23ae, 0x4c3e, \
> +		 0x92, 0xe9, 0x49, 0x64, 0xcf, 0x63, 0xf3, 0x49)
> +struct efi_unicode_collation_protocol {
> +	efi_intn_t (EFIAPI *stri_coll)(
> +		struct efi_unicode_collation_protocol *this, u16 *s1, u16 *s2);
> +	bool (EFIAPI *metai_match)(struct efi_unicode_collation_protocol *this,
> +				   const u16 *string, const u16 *patter);
> +	void (EFIAPI *str_lwr)(struct efi_unicode_collation_protocol
> +			       *this, u16 *string);
> +	void (EFIAPI *str_upr)(struct efi_unicode_collation_protocol *this,
> +			       u16 *string);
> +	void (EFIAPI *fat_to_str)(struct efi_unicode_collation_protocol *this,
> +				  efi_uintn_t fat_size, char *fat, u16 *string);
> +	bool (EFIAPI *str_to_fat)(struct efi_unicode_collation_protocol *this,
> +				  const u16 *string, efi_uintn_t fat_size,
> +				  char *fat);
> +	char *supported_languages;
> +};
> +
>  #endif
> diff --git a/include/efi_loader.h b/include/efi_loader.h
> index b46babf931..736e82c506 100644
> --- a/include/efi_loader.h
> +++ b/include/efi_loader.h
> @@ -98,6 +98,9 @@ extern const struct efi_device_path_to_text_protocol efi_device_path_to_text;
>  /* implementation of the EFI_DEVICE_PATH_UTILITIES_PROTOCOL */
>  extern const struct efi_device_path_utilities_protocol
>  					efi_device_path_utilities;
> +/* Implementation of the EFI_UNICODE_COLLATION_PROTOCOL */
> +extern const struct efi_unicode_collation_protocol
> +					efi_unicode_collation_protocol;
>  
>  uint16_t *efi_dp_str(struct efi_device_path *dp);
>  
> @@ -127,6 +130,8 @@ extern const efi_guid_t efi_file_info_guid;
>  /* GUID for file system information */
>  extern const efi_guid_t efi_file_system_info_guid;
>  extern const efi_guid_t efi_guid_device_path_utilities_protocol;
> +/* GUID of the Unicode collation protocol */
> +extern const efi_guid_t efi_guid_unicode_collation_protocol;
>  
>  extern unsigned int __efi_runtime_start, __efi_runtime_stop;
>  extern unsigned int __efi_runtime_rel_start, __efi_runtime_rel_stop;
> diff --git a/lib/efi_loader/Makefile b/lib/efi_loader/Makefile
> index 1ffbf52a89..0769df20f7 100644
> --- a/lib/efi_loader/Makefile
> +++ b/lib/efi_loader/Makefile
> @@ -17,10 +17,20 @@ always += helloworld.efi
>  endif
>  
>  obj-$(CONFIG_CMD_BOOTEFI_HELLO) += helloworld_efi.o
> -obj-y += efi_image_loader.o efi_boottime.o efi_runtime.o efi_console.o
> -obj-y += efi_memory.o efi_device_path_to_text.o efi_device_path.o
> -obj-y += efi_device_path_utilities.o efi_file.o efi_variable.o efi_bootmgr.o
> -obj-y += efi_watchdog.o
> +obj-y += \
> +efi_bootmgr.o \
> +efi_boottime.o \
> +efi_console.o \
> +efi_device_path.o \
> +efi_device_path_to_text.o \
> +efi_device_path_utilities.o \
> +efi_file.o \
> +efi_image_loader.o \
> +efi_memory.o \
> +efi_runtime.o \
> +efi_unicode_collation.o \
> +efi_variable.o \
> +efi_watchdog.o

Instead of line breaking them, just make each line an obj-y += line.

>  obj-$(CONFIG_LCD) += efi_gop.o
>  obj-$(CONFIG_DM_VIDEO) += efi_gop.o
>  obj-$(CONFIG_PARTITIONS) += efi_disk.o
> diff --git a/lib/efi_loader/efi_boottime.c b/lib/efi_loader/efi_boottime.c
> index cd5fdb7aa8..f9f59b93f9 100644
> --- a/lib/efi_loader/efi_boottime.c
> +++ b/lib/efi_loader/efi_boottime.c
> @@ -1527,6 +1527,12 @@ efi_status_t efi_setup_loaded_image(
>  	if (ret != EFI_SUCCESS)
>  		goto failure;
>  
> +	ret = efi_add_protocol(obj->handle,
> +			       &efi_guid_unicode_collation_protocol,
> +			       (void *)&efi_unicode_collation_protocol);
> +	if (ret != EFI_SUCCESS)
> +		goto failure;
> +
>  	return ret;
>  failure:
>  	printf("ERROR: Failure to install protocols for loaded image\n");
> diff --git a/lib/efi_loader/efi_unicode_collation.c b/lib/efi_loader/efi_unicode_collation.c
> new file mode 100644
> index 0000000000..2c1e819174
> --- /dev/null
> +++ b/lib/efi_loader/efi_unicode_collation.c
> @@ -0,0 +1,329 @@
> +// SPDX-License-Identifier: GPL-2.0+
> +/*
> + * EFI Unicode collation protocol
> + *
> + * Copyright (c) 2018 Heinrich Schuchardt <xypron.glpk@gmx.de>
> + */
> +
> +#include <common.h>
> +#include <charset.h>
> +#include <cp1250.h>
> +#include <cp437.h>
> +#include <efi_loader.h>
> +
> +/* Characters that may not be used in file names */
> +static const char illegal[] = "<>:\"/\\|?*";
> +
> +/*
> + * EDK2 assumes codepage 1250 when creating FAT 8.3 file names.
> + * Linux defaults to codepage 437 for FAT 8.3 file names.
> + */
> +#if CONFIG_FAT_DEFAULT_CODEPAGE == 1250
> +/* Unicode code points for code page 1250 characters 0x80 - 0xff */
> +static const u16 codepage[] = CP1250;
> +#else
> +/* Unicode code points for code page 437 characters 0x80 - 0xff */
> +static const u16 codepage[] = CP437;
> +#endif
> +
> +/* GUID of the EFI_UNICODE_COLLATION_PROTOCOL */
> +const efi_guid_t efi_guid_unicode_collation_protocol =
> +	EFI_UNICODE_COLLATION_PROTOCOL2_GUID;
> +
> +/**
> + * efi_stri_coll() - compare utf-16 strings case-insenitively
> + *
> + * @this:	unicode collation protocol instance
> + * @s1:		first string
> + * @s2:		second string
> + *
> + * This function implements the StriColl() service of the
> + * EFI_UNICODE_COLLATION_PROTOCOL.
> + *
> + * See the Unified Extensible Firmware Interface (UEFI) specification for
> + * details.
> + *
> + * TODO:
> + * The implementation does not follow the Unicode collation algorithm.
> + * For ASCII characters it results in the same sort order as EDK2.
> + * We could use table UNICODE_CAPITALIZATION_TABLE for better results.
> + *
> + * Return:	0: s1 == s2, > 0: s1 > s2, < 0: s1 < s2
> + */
> +static efi_intn_t EFIAPI efi_stri_coll(
> +		struct efi_unicode_collation_protocol *this, u16 *s1, u16 *s2)
> +{
> +	s32 c1, c2;
> +	efi_intn_t ret = 0;
> +
> +	EFI_ENTRY("%p, %ls, %ls", this, s1, s2);
> +	for (; *s1 | *s2; ++s1, ++s2) {
> +		c1 = utf_to_upper(*s1);
> +		c2 = utf_to_upper(*s2);
> +		if (c1 < c2) {
> +			ret = -1;
> +			goto out;
> +		} else if (c1 > c2) {
> +			ret = 1;
> +			goto out;
> +		}
> +	}
> +out:
> +	EFI_EXIT(EFI_SUCCESS);
> +	return ret;
> +}
> +
> +/**
> + * metai_match() - compare utf-16 string with a pattern string case-insenitively
> + *
> + * @s:		string to compare
> + * @p:		pattern string
> + *
> + * The pattern string may use these:
> + *	- * matches >= 0 characters
> + *	- ? matches 1 character
> + *	- [<char1><char2>...<charN>] match any character in the set
> + *	- [<char1>-<char2>] matches any character in the range
> + *
> + * This function is called my efi_metai_match().
> + *
> + * For '*' pattern searches this function calls itself recursively.
> + * Performance-wise this is suboptimal, especially for multiple '*' wildcards.
> + * But it results in simple code.
> + *
> + * Return:	true if the string is matched.
> + */
> +static bool metai_match(const u16 *s, const u16 *p)
> +{
> +	u16 first;
> +
> +	for (; *s && *p; ++s, ++p) {
> +		switch (*p) {
> +		case '*':
> +			/* Match 0 or more characters */
> +			++p;
> +			for (;; ++s) {
> +				if (metai_match(s, p))
> +					return true;
> +				if (!*s)
> +					return false;
> +			}
> +		case '?':
> +			/* Match any one character */
> +			break;
> +		case '[':
> +			/* Match any character in the set */
> +			++p;
> +			first = *p;
> +			if (first == ']')
> +				/* Empty set */
> +				return false;
> +			++p;
> +			if (*p == '-') {
> +				/* Range */
> +				++p;
> +				if (*s < first || *s > *p)
> +					return false;
> +				++p;
> +				if (*p != ']')
> +					return false;
> +			} else {
> +				/* Set */
> +				bool hit = false;
> +
> +				if (*s == first)
> +					hit = true;
> +				for (; *p && *p != ']'; ++p) {
> +					if (*p == *s)
> +						hit = true;
> +				}
> +				if (!hit || *p != ']')
> +					return false;
> +			}
> +			break;
> +		default:
> +			/* Match one character */
> +			if (*p != *s)
> +				return false;
> +		}
> +	}
> +	if (!*p && !*s)
> +		return true;
> +	return false;
> +}
> +
> +/**
> + * efi_metai_match() - compare utf-16 string with a pattern string
> + *		       case-insenitively
> + *
> + * @this:	unicode collation protocol instance
> + * @s:		string to compare
> + * @p:		pattern string
> + *
> + * The pattern string may use these:
> + *	- * matches >= 0 characters
> + *	- ? matches 1 character
> + *	- [<char1><char2>...<charN>] match any character in the set
> + *	- [<char1>-<char2>] matches any character in the range
> + *
> + * This function implements the MetaMatch() service of the
> + * EFI_UNICODE_COLLATION_PROTOCOL.
> + *
> + * Return:	true if the string is matched.
> + */
> +static bool EFIAPI efi_metai_match(struct efi_unicode_collation_protocol *this,
> +				   const u16 *string, const u16 *pattern)
> +{
> +	bool ret;
> +
> +	EFI_ENTRY("%p, %ls, %ls", this, string, pattern);
> +	ret =  metai_match(string, pattern);
> +	EFI_EXIT(EFI_SUCCESS);
> +	return ret;
> +}
> +
> +/**
> + * efi_str_lwr() - convert to lower case
> + *
> + * @this:	unicode collation protocol instance
> + * @string:	string to convert
> + * @p:		pattern string
> + *
> + * The conversion is done in place. As long as upper and lower letters use the
> + * same number of words this does not pose a problem.
> + *
> + * This function implements the StrLwr() service of the
> + * EFI_UNICODE_COLLATION_PROTOCOL.
> + */
> +static void EFIAPI efi_str_lwr(struct efi_unicode_collation_protocol *this,
> +			       u16 *string)
> +{
> +	EFI_ENTRY("%p, %ls", this, string);
> +	for (; *string; ++string)
> +		*string = utf_to_lower(*string);
> +	EFI_EXIT(EFI_SUCCESS);
> +}
> +
> +/**
> + * efi_str_upr() - convert to upper case
> + *
> + * @this:	unicode collation protocol instance
> + * @string:	string to convert
> + * @p:		pattern string
> + *
> + * The conversion is done in place. As long as upper and lower letters use the
> + * same number of words this does not pose a problem.
> + *
> + * This function implements the StrUpr() service of the
> + * EFI_UNICODE_COLLATION_PROTOCOL.
> + */
> +static void EFIAPI efi_str_upr(struct efi_unicode_collation_protocol *this,
> +			       u16 *string)
> +{
> +	EFI_ENTRY("%p, %ls", this, string);
> +	for (; *string; ++string)
> +		*string = utf_to_upper(*string);
> +	EFI_EXIT(EFI_SUCCESS);
> +}
> +
> +/**
> + * efi_fat_to_str() - convert an 8.3 file name from an OEM codepage to Unicode
> + *
> + * @this:	unicode collation protocol instance
> + * @fat_size:	size of the string to convert
> + * @fat:	string to convert
> + * @string:	converted string
> + *
> + * This function implements the FatToStr() service of the
> + * EFI_UNICODE_COLLATION_PROTOCOL.
> + */
> +static void EFIAPI efi_fat_to_str(struct efi_unicode_collation_protocol *this,
> +				  efi_uintn_t fat_size, char *fat, u16 *string)
> +{
> +	efi_uintn_t i;
> +	u16 c;
> +
> +	EFI_ENTRY("%p, %zu, %s, %p", this, fat_size, fat, string);
> +	for (i = 0; i < fat_size; ++i) {
> +		c = (unsigned char)fat[i];
> +		if (c > 0x80)
> +			c = codepage[i - 0x80];
> +		string[i] = c;
> +		if (!c)
> +			break;
> +	}
> +	string[i] = 0;
> +	EFI_EXIT(EFI_SUCCESS);
> +}
> +
> +/**
> + * efi_fat_to_str() - convert a utf-16 string to legal characters for a FAT
> + *                    file name in an OEM code page
> + *
> + * @this:	unicode collation protocol instance
> + * @string:	Unicode string to convert
> + * @fat_size:	size of the target buffer
> + * @fat:	converted string
> + *
> + * This function implements the StrToFat() service of the
> + * EFI_UNICODE_COLLATION_PROTOCOL.
> + *
> + * Return:	true if an illegal character was substituted by '_'.
> + */
> +static bool EFIAPI efi_str_to_fat(struct efi_unicode_collation_protocol *this,
> +				  const u16 *string, efi_uintn_t fat_size,
> +				  char *fat)
> +{
> +	efi_uintn_t i;
> +	s32 c;
> +	bool ret = false;
> +
> +	EFI_ENTRY("%p, %ls, %zu, %p", this, string, fat_size, fat);
> +	for (i = 0; i < fat_size;) {
> +		c = utf16_get(&string);
> +		switch (c) {
> +		/* Ignore period and space */
> +		case '.':
> +		case ' ':
> +			continue;
> +		case 0:
> +			break;
> +		}
> +		c = utf_to_upper(c);
> +		if (c >= 0x80) {
> +			int j;
> +
> +			/* Look for codepage translation */
> +			for (j = 0; j < 0x80; ++j) {
> +				if (c == codepage[j]) {
> +					c = j + 0x80;
> +					break;
> +				}
> +			}
> +			if (j >= 0x80) {
> +				c = '_';
> +				ret = true;
> +			}
> +		} else if (c && (c < 0x20 || strchr(illegal, c))) {
> +			c = '_';
> +			ret = true;
> +		}
> +
> +		fat[i] = c;
> +		if (!c)
> +			break;
> +		++i;
> +	}
> +	EFI_EXIT(EFI_SUCCESS);
> +	return ret;
> +}
> +
> +const struct efi_unicode_collation_protocol efi_unicode_collation_protocol = {
> +	.stri_coll = efi_stri_coll,
> +	.metai_match = efi_metai_match,
> +	.str_lwr = efi_str_lwr,
> +	.str_upr = efi_str_upr,
> +	.fat_to_str = efi_fat_to_str,
> +	.str_to_fat = efi_str_to_fat,
> +	.supported_languages = "en-US",

I have to admit I'm slightly puzzled on the supported language field.
The spec says quite clearly in appendix M what to do, but everyone seems
to interpret the specs differently:

  edk2 provides "en"
  appendix M example is "en-us"
  your patch provides "en-US"

I'm not sure which one of them is correct :).


Alex

[U-Boot] [PATCH v3 4/5] efi_loader: EFI_UNICODE_COLLATION_PROTOCOL

[Heinrich Schuchardt]

On 09/03/2018 11:09 PM, Alexander Graf wrote:
> 
> 
> On 03.09.18 18:35, Heinrich Schuchardt wrote:
>> The patch implements the EFI_UNICODE_COLLATION_PROTOCOL.
>>
>> Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
>> ---
>> v3:
>> 	no change
>> v2:
>> 	add more comments
>> ---
>>  MAINTAINERS                            |   2 +
>>  include/cp1250.h                       |  40 +++
>>  include/cp437.h                        |  40 +++
>>  include/efi_api.h                      |  21 ++
>>  include/efi_loader.h                   |   5 +
>>  lib/efi_loader/Makefile                |  18 +-
>>  lib/efi_loader/efi_boottime.c          |   6 +
>>  lib/efi_loader/efi_unicode_collation.c | 329 +++++++++++++++++++++++++
>>  8 files changed, 457 insertions(+), 4 deletions(-)
>>  create mode 100644 include/cp1250.h
>>  create mode 100644 include/cp437.h
>>  create mode 100644 lib/efi_loader/efi_unicode_collation.c
>>
>> diff --git a/MAINTAINERS b/MAINTAINERS
>> index 8c9cd83347..67b015205c 100644
>> --- a/MAINTAINERS
>> +++ b/MAINTAINERS
>> @@ -382,6 +382,8 @@ F:	doc/README.uefi
>>  F:	doc/README.iscsi
>>  F:	Documentation/efi.rst
>>  F:	include/capitalization.h
>> +F:	include/cp1250.h
>> +F:	include/cp437.h
>>  F:	include/efi*
>>  F:	include/pe.h
>>  F:	include/asm-generic/pe.h
>> diff --git a/include/cp1250.h b/include/cp1250.h
>> new file mode 100644
>> index 0000000000..adacf8a958
>> --- /dev/null
>> +++ b/include/cp1250.h
>> @@ -0,0 +1,40 @@
>> +/* SPDX-License-Identifier: GPL-2.0+ */
>> +
>> +/*
>> + * Constant CP1250 contains the Unicode code points for characters 0x80 - 0xff
>> + * of the code page 1250.
>> + */
>> +#define CP1250 { \
>> +	0x20ac, 0x0000, 0x201a, 0x0000, \
>> +	0x201e, 0x2026, 0x2020, 0x2021, \
>> +	0x0000, 0x2030, 0x0160, 0x2039, \
>> +	0x015a, 0x0164, 0x017d, 0x0179, \
>> +	0x0000, 0x2018, 0x2019, 0x201c, \
>> +	0x201d, 0x2022, 0x2013, 0x2014, \
>> +	0x0000, 0x2122, 0x0161, 0x203a, \
>> +	0x015b, 0x0165, 0x017e, 0x017a, \
>> +	0x00a0, 0x02c7, 0x02d8, 0x0141, \
>> +	0x00a4, 0x0104, 0x00a6, 0x00a7, \
>> +	0x00a8, 0x00a9, 0x015e, 0x00ab, \
>> +	0x00ac, 0x00ad, 0x00ae, 0x017b, \
>> +	0x00b0, 0x00b1, 0x02db, 0x0142, \
>> +	0x00b4, 0x00b5, 0x00b6, 0x00b7, \
>> +	0x00b8, 0x0105, 0x015f, 0x00bb, \
>> +	0x013d, 0x02dd, 0x013e, 0x017c, \
>> +	0x0154, 0x00c1, 0x00c2, 0x0102, \
>> +	0x00c4, 0x0139, 0x0106, 0x00c7, \
>> +	0x010c, 0x00c9, 0x0118, 0x00cb, \
>> +	0x011a, 0x00cd, 0x00ce, 0x010e, \
>> +	0x0110, 0x0143, 0x0147, 0x00d3, \
>> +	0x00d4, 0x0150, 0x00d6, 0x00d7, \
>> +	0x0158, 0x016e, 0x00da, 0x0170, \
>> +	0x00dc, 0x00dd, 0x0162, 0x00df, \
>> +	0x0155, 0x00e1, 0x00e2, 0x0103, \
>> +	0x00e4, 0x013a, 0x0107, 0x00e7, \
>> +	0x010d, 0x00e9, 0x0119, 0x00eb, \
>> +	0x011b, 0x00ed, 0x00ee, 0x010f, \
>> +	0x0111, 0x0144, 0x0148, 0x00f3, \
>> +	0x00f4, 0x0151, 0x00f6, 0x00f7, \
>> +	0x0159, 0x016f, 0x00fa, 0x0171, \
>> +	0x00fc, 0x00fd, 0x0163, 0x02d9, \
>> +}
>> diff --git a/include/cp437.h b/include/cp437.h
>> new file mode 100644
>> index 0000000000..0b2b97132e
>> --- /dev/null
>> +++ b/include/cp437.h
>> @@ -0,0 +1,40 @@
>> +/* SPDX-License-Identifier: GPL-2.0+ */
>> +
>> +/*
>> + * Constant CP437 contains the Unicode code points for characters 0x80 - 0xff
>> + * of the code page 437.
>> + */
>> +#define CP437 { \
>> +	0x00c7, 0x00fc, 0x00e9, 0x00e2, \
>> +	0x00e4, 0x00e0, 0x00e5, 0x00e7, \
>> +	0x00ea, 0x00eb, 0x00e8, 0x00ef, \
>> +	0x00ee, 0x00ec, 0x00c4, 0x00c5, \
>> +	0x00c9, 0x00e6, 0x00c6, 0x00f4, \
>> +	0x00f6, 0x00f2, 0x00fb, 0x00f9, \
>> +	0x00ff, 0x00d6, 0x00dc, 0x00a2, \
>> +	0x00a3, 0x00a5, 0x20a7, 0x0192, \
>> +	0x00e1, 0x00ed, 0x00f3, 0x00fa, \
>> +	0x00f1, 0x00d1, 0x00aa, 0x00ba, \
>> +	0x00bf, 0x2310, 0x00ac, 0x00bd, \
>> +	0x00bc, 0x00a1, 0x00ab, 0x00bb, \
>> +	0x2591, 0x2592, 0x2593, 0x2502, \
>> +	0x2524, 0x2561, 0x2562, 0x2556, \
>> +	0x2555, 0x2563, 0x2551, 0x2557, \
>> +	0x255d, 0x255c, 0x255b, 0x2510, \
>> +	0x2514, 0x2534, 0x252c, 0x251c, \
>> +	0x2500, 0x253c, 0x255e, 0x255f, \
>> +	0x255a, 0x2554, 0x2569, 0x2566, \
>> +	0x2560, 0x2550, 0x256c, 0x2567, \
>> +	0x2568, 0x2564, 0x2565, 0x2559, \
>> +	0x2558, 0x2552, 0x2553, 0x256b, \
>> +	0x256a, 0x2518, 0x250c, 0x2588, \
>> +	0x2584, 0x258c, 0x2590, 0x2580, \
>> +	0x03b1, 0x00df, 0x0393, 0x03c0, \
>> +	0x03a3, 0x03c3, 0x00b5, 0x03c4, \
>> +	0x03a6, 0x0398, 0x03a9, 0x03b4, \
>> +	0x221e, 0x03c6, 0x03b5, 0x2229, \
>> +	0x2261, 0x00b1, 0x2265, 0x2264, \
>> +	0x2320, 0x2321, 0x00f7, 0x2248, \
>> +	0x00b0, 0x2219, 0x00b7, 0x221a, \
>> +	0x207f, 0x00b2, 0x25a0, 0x00a0, \
>> +}
>> diff --git a/include/efi_api.h b/include/efi_api.h
>> index ebf2a3bc18..1efc448184 100644
>> --- a/include/efi_api.h
>> +++ b/include/efi_api.h
>> @@ -31,6 +31,7 @@ enum efi_timer_delay {
>>  	EFI_TIMER_RELATIVE = 2
>>  };
>>  
>> +#define efi_intn_t ssize_t
>>  #define efi_uintn_t size_t
>>  typedef uint16_t *efi_string_t;
>>  
>> @@ -958,4 +959,24 @@ struct efi_driver_binding_protocol {
>>  	efi_handle_t driver_binding_handle;
>>  };
>>  
>> +#define EFI_UNICODE_COLLATION_PROTOCOL2_GUID \
>> +	EFI_GUID(0xa4c751fc, 0x23ae, 0x4c3e, \
>> +		 0x92, 0xe9, 0x49, 0x64, 0xcf, 0x63, 0xf3, 0x49)
>> +struct efi_unicode_collation_protocol {
>> +	efi_intn_t (EFIAPI *stri_coll)(
>> +		struct efi_unicode_collation_protocol *this, u16 *s1, u16 *s2);
>> +	bool (EFIAPI *metai_match)(struct efi_unicode_collation_protocol *this,
>> +				   const u16 *string, const u16 *patter);
>> +	void (EFIAPI *str_lwr)(struct efi_unicode_collation_protocol
>> +			       *this, u16 *string);
>> +	void (EFIAPI *str_upr)(struct efi_unicode_collation_protocol *this,
>> +			       u16 *string);
>> +	void (EFIAPI *fat_to_str)(struct efi_unicode_collation_protocol *this,
>> +				  efi_uintn_t fat_size, char *fat, u16 *string);
>> +	bool (EFIAPI *str_to_fat)(struct efi_unicode_collation_protocol *this,
>> +				  const u16 *string, efi_uintn_t fat_size,
>> +				  char *fat);
>> +	char *supported_languages;
>> +};
>> +
>>  #endif
>> diff --git a/include/efi_loader.h b/include/efi_loader.h
>> index b46babf931..736e82c506 100644
>> --- a/include/efi_loader.h
>> +++ b/include/efi_loader.h
>> @@ -98,6 +98,9 @@ extern const struct efi_device_path_to_text_protocol efi_device_path_to_text;
>>  /* implementation of the EFI_DEVICE_PATH_UTILITIES_PROTOCOL */
>>  extern const struct efi_device_path_utilities_protocol
>>  					efi_device_path_utilities;
>> +/* Implementation of the EFI_UNICODE_COLLATION_PROTOCOL */
>> +extern const struct efi_unicode_collation_protocol
>> +					efi_unicode_collation_protocol;
>>  
>>  uint16_t *efi_dp_str(struct efi_device_path *dp);
>>  
>> @@ -127,6 +130,8 @@ extern const efi_guid_t efi_file_info_guid;
>>  /* GUID for file system information */
>>  extern const efi_guid_t efi_file_system_info_guid;
>>  extern const efi_guid_t efi_guid_device_path_utilities_protocol;
>> +/* GUID of the Unicode collation protocol */
>> +extern const efi_guid_t efi_guid_unicode_collation_protocol;
>>  
>>  extern unsigned int __efi_runtime_start, __efi_runtime_stop;
>>  extern unsigned int __efi_runtime_rel_start, __efi_runtime_rel_stop;
>> diff --git a/lib/efi_loader/Makefile b/lib/efi_loader/Makefile
>> index 1ffbf52a89..0769df20f7 100644
>> --- a/lib/efi_loader/Makefile
>> +++ b/lib/efi_loader/Makefile
>> @@ -17,10 +17,20 @@ always += helloworld.efi
>>  endif
>>  
>>  obj-$(CONFIG_CMD_BOOTEFI_HELLO) += helloworld_efi.o
>> -obj-y += efi_image_loader.o efi_boottime.o efi_runtime.o efi_console.o
>> -obj-y += efi_memory.o efi_device_path_to_text.o efi_device_path.o
>> -obj-y += efi_device_path_utilities.o efi_file.o efi_variable.o efi_bootmgr.o
>> -obj-y += efi_watchdog.o
>> +obj-y += \
>> +efi_bootmgr.o \
>> +efi_boottime.o \
>> +efi_console.o \
>> +efi_device_path.o \
>> +efi_device_path_to_text.o \
>> +efi_device_path_utilities.o \
>> +efi_file.o \
>> +efi_image_loader.o \
>> +efi_memory.o \
>> +efi_runtime.o \
>> +efi_unicode_collation.o \
>> +efi_variable.o \
>> +efi_watchdog.o
> 
> Instead of line breaking them, just make each line an obj-y += line.
> 
>>  obj-$(CONFIG_LCD) += efi_gop.o
>>  obj-$(CONFIG_DM_VIDEO) += efi_gop.o
>>  obj-$(CONFIG_PARTITIONS) += efi_disk.o
>> diff --git a/lib/efi_loader/efi_boottime.c b/lib/efi_loader/efi_boottime.c
>> index cd5fdb7aa8..f9f59b93f9 100644
>> --- a/lib/efi_loader/efi_boottime.c
>> +++ b/lib/efi_loader/efi_boottime.c
>> @@ -1527,6 +1527,12 @@ efi_status_t efi_setup_loaded_image(
>>  	if (ret != EFI_SUCCESS)
>>  		goto failure;
>>  
>> +	ret = efi_add_protocol(obj->handle,
>> +			       &efi_guid_unicode_collation_protocol,
>> +			       (void *)&efi_unicode_collation_protocol);
>> +	if (ret != EFI_SUCCESS)
>> +		goto failure;
>> +
>>  	return ret;
>>  failure:
>>  	printf("ERROR: Failure to install protocols for loaded image\n");
>> diff --git a/lib/efi_loader/efi_unicode_collation.c b/lib/efi_loader/efi_unicode_collation.c
>> new file mode 100644
>> index 0000000000..2c1e819174
>> --- /dev/null
>> +++ b/lib/efi_loader/efi_unicode_collation.c
>> @@ -0,0 +1,329 @@
>> +// SPDX-License-Identifier: GPL-2.0+
>> +/*
>> + * EFI Unicode collation protocol
>> + *
>> + * Copyright (c) 2018 Heinrich Schuchardt <xypron.glpk@gmx.de>
>> + */
>> +
>> +#include <common.h>
>> +#include <charset.h>
>> +#include <cp1250.h>
>> +#include <cp437.h>
>> +#include <efi_loader.h>
>> +
>> +/* Characters that may not be used in file names */
>> +static const char illegal[] = "<>:\"/\\|?*";
>> +
>> +/*
>> + * EDK2 assumes codepage 1250 when creating FAT 8.3 file names.
>> + * Linux defaults to codepage 437 for FAT 8.3 file names.
>> + */
>> +#if CONFIG_FAT_DEFAULT_CODEPAGE == 1250
>> +/* Unicode code points for code page 1250 characters 0x80 - 0xff */
>> +static const u16 codepage[] = CP1250;
>> +#else
>> +/* Unicode code points for code page 437 characters 0x80 - 0xff */
>> +static const u16 codepage[] = CP437;
>> +#endif
>> +
>> +/* GUID of the EFI_UNICODE_COLLATION_PROTOCOL */
>> +const efi_guid_t efi_guid_unicode_collation_protocol =
>> +	EFI_UNICODE_COLLATION_PROTOCOL2_GUID;
>> +
>> +/**
>> + * efi_stri_coll() - compare utf-16 strings case-insenitively
>> + *
>> + * @this:	unicode collation protocol instance
>> + * @s1:		first string
>> + * @s2:		second string
>> + *
>> + * This function implements the StriColl() service of the
>> + * EFI_UNICODE_COLLATION_PROTOCOL.
>> + *
>> + * See the Unified Extensible Firmware Interface (UEFI) specification for
>> + * details.
>> + *
>> + * TODO:
>> + * The implementation does not follow the Unicode collation algorithm.
>> + * For ASCII characters it results in the same sort order as EDK2.
>> + * We could use table UNICODE_CAPITALIZATION_TABLE for better results.
>> + *
>> + * Return:	0: s1 == s2, > 0: s1 > s2, < 0: s1 < s2
>> + */
>> +static efi_intn_t EFIAPI efi_stri_coll(
>> +		struct efi_unicode_collation_protocol *this, u16 *s1, u16 *s2)
>> +{
>> +	s32 c1, c2;
>> +	efi_intn_t ret = 0;
>> +
>> +	EFI_ENTRY("%p, %ls, %ls", this, s1, s2);
>> +	for (; *s1 | *s2; ++s1, ++s2) {
>> +		c1 = utf_to_upper(*s1);
>> +		c2 = utf_to_upper(*s2);
>> +		if (c1 < c2) {
>> +			ret = -1;
>> +			goto out;
>> +		} else if (c1 > c2) {
>> +			ret = 1;
>> +			goto out;
>> +		}
>> +	}
>> +out:
>> +	EFI_EXIT(EFI_SUCCESS);
>> +	return ret;
>> +}
>> +
>> +/**
>> + * metai_match() - compare utf-16 string with a pattern string case-insenitively
>> + *
>> + * @s:		string to compare
>> + * @p:		pattern string
>> + *
>> + * The pattern string may use these:
>> + *	- * matches >= 0 characters
>> + *	- ? matches 1 character
>> + *	- [<char1><char2>...<charN>] match any character in the set
>> + *	- [<char1>-<char2>] matches any character in the range
>> + *
>> + * This function is called my efi_metai_match().
>> + *
>> + * For '*' pattern searches this function calls itself recursively.
>> + * Performance-wise this is suboptimal, especially for multiple '*' wildcards.
>> + * But it results in simple code.
>> + *
>> + * Return:	true if the string is matched.
>> + */
>> +static bool metai_match(const u16 *s, const u16 *p)
>> +{
>> +	u16 first;
>> +
>> +	for (; *s && *p; ++s, ++p) {
>> +		switch (*p) {
>> +		case '*':
>> +			/* Match 0 or more characters */
>> +			++p;
>> +			for (;; ++s) {
>> +				if (metai_match(s, p))
>> +					return true;
>> +				if (!*s)
>> +					return false;
>> +			}
>> +		case '?':
>> +			/* Match any one character */
>> +			break;
>> +		case '[':
>> +			/* Match any character in the set */
>> +			++p;
>> +			first = *p;
>> +			if (first == ']')
>> +				/* Empty set */
>> +				return false;
>> +			++p;
>> +			if (*p == '-') {
>> +				/* Range */
>> +				++p;
>> +				if (*s < first || *s > *p)
>> +					return false;
>> +				++p;
>> +				if (*p != ']')
>> +					return false;
>> +			} else {
>> +				/* Set */
>> +				bool hit = false;
>> +
>> +				if (*s == first)
>> +					hit = true;
>> +				for (; *p && *p != ']'; ++p) {
>> +					if (*p == *s)
>> +						hit = true;
>> +				}
>> +				if (!hit || *p != ']')
>> +					return false;
>> +			}
>> +			break;
>> +		default:
>> +			/* Match one character */
>> +			if (*p != *s)
>> +				return false;
>> +		}
>> +	}
>> +	if (!*p && !*s)
>> +		return true;
>> +	return false;
>> +}
>> +
>> +/**
>> + * efi_metai_match() - compare utf-16 string with a pattern string
>> + *		       case-insenitively
>> + *
>> + * @this:	unicode collation protocol instance
>> + * @s:		string to compare
>> + * @p:		pattern string
>> + *
>> + * The pattern string may use these:
>> + *	- * matches >= 0 characters
>> + *	- ? matches 1 character
>> + *	- [<char1><char2>...<charN>] match any character in the set
>> + *	- [<char1>-<char2>] matches any character in the range
>> + *
>> + * This function implements the MetaMatch() service of the
>> + * EFI_UNICODE_COLLATION_PROTOCOL.
>> + *
>> + * Return:	true if the string is matched.
>> + */
>> +static bool EFIAPI efi_metai_match(struct efi_unicode_collation_protocol *this,
>> +				   const u16 *string, const u16 *pattern)
>> +{
>> +	bool ret;
>> +
>> +	EFI_ENTRY("%p, %ls, %ls", this, string, pattern);
>> +	ret =  metai_match(string, pattern);
>> +	EFI_EXIT(EFI_SUCCESS);
>> +	return ret;
>> +}
>> +
>> +/**
>> + * efi_str_lwr() - convert to lower case
>> + *
>> + * @this:	unicode collation protocol instance
>> + * @string:	string to convert
>> + * @p:		pattern string
>> + *
>> + * The conversion is done in place. As long as upper and lower letters use the
>> + * same number of words this does not pose a problem.
>> + *
>> + * This function implements the StrLwr() service of the
>> + * EFI_UNICODE_COLLATION_PROTOCOL.
>> + */
>> +static void EFIAPI efi_str_lwr(struct efi_unicode_collation_protocol *this,
>> +			       u16 *string)
>> +{
>> +	EFI_ENTRY("%p, %ls", this, string);
>> +	for (; *string; ++string)
>> +		*string = utf_to_lower(*string);
>> +	EFI_EXIT(EFI_SUCCESS);
>> +}
>> +
>> +/**
>> + * efi_str_upr() - convert to upper case
>> + *
>> + * @this:	unicode collation protocol instance
>> + * @string:	string to convert
>> + * @p:		pattern string
>> + *
>> + * The conversion is done in place. As long as upper and lower letters use the
>> + * same number of words this does not pose a problem.
>> + *
>> + * This function implements the StrUpr() service of the
>> + * EFI_UNICODE_COLLATION_PROTOCOL.
>> + */
>> +static void EFIAPI efi_str_upr(struct efi_unicode_collation_protocol *this,
>> +			       u16 *string)
>> +{
>> +	EFI_ENTRY("%p, %ls", this, string);
>> +	for (; *string; ++string)
>> +		*string = utf_to_upper(*string);
>> +	EFI_EXIT(EFI_SUCCESS);
>> +}
>> +
>> +/**
>> + * efi_fat_to_str() - convert an 8.3 file name from an OEM codepage to Unicode
>> + *
>> + * @this:	unicode collation protocol instance
>> + * @fat_size:	size of the string to convert
>> + * @fat:	string to convert
>> + * @string:	converted string
>> + *
>> + * This function implements the FatToStr() service of the
>> + * EFI_UNICODE_COLLATION_PROTOCOL.
>> + */
>> +static void EFIAPI efi_fat_to_str(struct efi_unicode_collation_protocol *this,
>> +				  efi_uintn_t fat_size, char *fat, u16 *string)
>> +{
>> +	efi_uintn_t i;
>> +	u16 c;
>> +
>> +	EFI_ENTRY("%p, %zu, %s, %p", this, fat_size, fat, string);
>> +	for (i = 0; i < fat_size; ++i) {
>> +		c = (unsigned char)fat[i];
>> +		if (c > 0x80)
>> +			c = codepage[i - 0x80];
>> +		string[i] = c;
>> +		if (!c)
>> +			break;
>> +	}
>> +	string[i] = 0;
>> +	EFI_EXIT(EFI_SUCCESS);
>> +}
>> +
>> +/**
>> + * efi_fat_to_str() - convert a utf-16 string to legal characters for a FAT
>> + *                    file name in an OEM code page
>> + *
>> + * @this:	unicode collation protocol instance
>> + * @string:	Unicode string to convert
>> + * @fat_size:	size of the target buffer
>> + * @fat:	converted string
>> + *
>> + * This function implements the StrToFat() service of the
>> + * EFI_UNICODE_COLLATION_PROTOCOL.
>> + *
>> + * Return:	true if an illegal character was substituted by '_'.
>> + */
>> +static bool EFIAPI efi_str_to_fat(struct efi_unicode_collation_protocol *this,
>> +				  const u16 *string, efi_uintn_t fat_size,
>> +				  char *fat)
>> +{
>> +	efi_uintn_t i;
>> +	s32 c;
>> +	bool ret = false;
>> +
>> +	EFI_ENTRY("%p, %ls, %zu, %p", this, string, fat_size, fat);
>> +	for (i = 0; i < fat_size;) {
>> +		c = utf16_get(&string);
>> +		switch (c) {
>> +		/* Ignore period and space */
>> +		case '.':
>> +		case ' ':
>> +			continue;
>> +		case 0:
>> +			break;
>> +		}
>> +		c = utf_to_upper(c);
>> +		if (c >= 0x80) {
>> +			int j;
>> +
>> +			/* Look for codepage translation */
>> +			for (j = 0; j < 0x80; ++j) {
>> +				if (c == codepage[j]) {
>> +					c = j + 0x80;
>> +					break;
>> +				}
>> +			}
>> +			if (j >= 0x80) {
>> +				c = '_';
>> +				ret = true;
>> +			}
>> +		} else if (c && (c < 0x20 || strchr(illegal, c))) {
>> +			c = '_';
>> +			ret = true;
>> +		}
>> +
>> +		fat[i] = c;
>> +		if (!c)
>> +			break;
>> +		++i;
>> +	}
>> +	EFI_EXIT(EFI_SUCCESS);
>> +	return ret;
>> +}
>> +
>> +const struct efi_unicode_collation_protocol efi_unicode_collation_protocol = {
>> +	.stri_coll = efi_stri_coll,
>> +	.metai_match = efi_metai_match,
>> +	.str_lwr = efi_str_lwr,
>> +	.str_upr = efi_str_upr,
>> +	.fat_to_str = efi_fat_to_str,
>> +	.str_to_fat = efi_str_to_fat,
>> +	.supported_languages = "en-US",
> 
> I have to admit I'm slightly puzzled on the supported language field.
> The spec says quite clearly in appendix M what to do, but everyone seems
> to interpret the specs differently:
> 
>   edk2 provides "en"
>   appendix M example is "en-us"
>   your patch provides "en-US"
> 
> I'm not sure which one of them is correct :).
> 

The spec refers to RFC 4646. RFC 4646 refers to ISO 3166 which uses
capital letters for the two letter region codes.

In RFC 4646 you can find the following example:

en-US (English as used in the United States)

The value from edk2 complies with RFC 4646. It simply does not specify a
region.

So should I change the value to match edk2?

Best regards

Heinrich

> 
> Alex
> 

[U-Boot] [PATCH v3 4/5] efi_loader: EFI_UNICODE_COLLATION_PROTOCOL

[Alexander Graf]

On 03.09.18 23:24, Heinrich Schuchardt wrote:
> On 09/03/2018 11:09 PM, Alexander Graf wrote:
>>
>>
>> On 03.09.18 18:35, Heinrich Schuchardt wrote:
>>> The patch implements the EFI_UNICODE_COLLATION_PROTOCOL.
>>>
>>> Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
>>> ---
>>> v3:
>>> 	no change
>>> v2:
>>> 	add more comments
>>> ---
>>>  MAINTAINERS                            |   2 +
>>>  include/cp1250.h                       |  40 +++
>>>  include/cp437.h                        |  40 +++
>>>  include/efi_api.h                      |  21 ++
>>>  include/efi_loader.h                   |   5 +
>>>  lib/efi_loader/Makefile                |  18 +-
>>>  lib/efi_loader/efi_boottime.c          |   6 +
>>>  lib/efi_loader/efi_unicode_collation.c | 329 +++++++++++++++++++++++++
>>>  8 files changed, 457 insertions(+), 4 deletions(-)
>>>  create mode 100644 include/cp1250.h
>>>  create mode 100644 include/cp437.h
>>>  create mode 100644 lib/efi_loader/efi_unicode_collation.c
>>>

[...]

>>> @@ -127,6 +130,8 @@ extern const efi_guid_t efi_file_info_guid;
>>>  /* GUID for file system information */
>>>  extern const efi_guid_t efi_file_system_info_guid;
>>>  extern const efi_guid_t efi_guid_device_path_utilities_protocol;
>>> +/* GUID of the Unicode collation protocol */
>>> +extern const efi_guid_t efi_guid_unicode_collation_protocol;
>>>  
>>>  extern unsigned int __efi_runtime_start, __efi_runtime_stop;
>>>  extern unsigned int __efi_runtime_rel_start, __efi_runtime_rel_stop;
>>> diff --git a/lib/efi_loader/Makefile b/lib/efi_loader/Makefile
>>> index 1ffbf52a89..0769df20f7 100644
>>> --- a/lib/efi_loader/Makefile
>>> +++ b/lib/efi_loader/Makefile
>>> @@ -17,10 +17,20 @@ always += helloworld.efi
>>>  endif
>>>  
>>>  obj-$(CONFIG_CMD_BOOTEFI_HELLO) += helloworld_efi.o
>>> -obj-y += efi_image_loader.o efi_boottime.o efi_runtime.o efi_console.o
>>> -obj-y += efi_memory.o efi_device_path_to_text.o efi_device_path.o
>>> -obj-y += efi_device_path_utilities.o efi_file.o efi_variable.o efi_bootmgr.o
>>> -obj-y += efi_watchdog.o
>>> +obj-y += \
>>> +efi_bootmgr.o \
>>> +efi_boottime.o \
>>> +efi_console.o \
>>> +efi_device_path.o \
>>> +efi_device_path_to_text.o \
>>> +efi_device_path_utilities.o \
>>> +efi_file.o \
>>> +efi_image_loader.o \
>>> +efi_memory.o \
>>> +efi_runtime.o \
>>> +efi_unicode_collation.o \
>>> +efi_variable.o \
>>> +efi_watchdog.o
>>
>> Instead of line breaking them, just make each line an obj-y += line.
>>

[...]

>>> +const struct efi_unicode_collation_protocol efi_unicode_collation_protocol = {
>>> +	.stri_coll = efi_stri_coll,
>>> +	.metai_match = efi_metai_match,
>>> +	.str_lwr = efi_str_lwr,
>>> +	.str_upr = efi_str_upr,
>>> +	.fat_to_str = efi_fat_to_str,
>>> +	.str_to_fat = efi_str_to_fat,
>>> +	.supported_languages = "en-US",
>>
>> I have to admit I'm slightly puzzled on the supported language field.
>> The spec says quite clearly in appendix M what to do, but everyone seems
>> to interpret the specs differently:
>>
>>   edk2 provides "en"
>>   appendix M example is "en-us"
>>   your patch provides "en-US"
>>
>> I'm not sure which one of them is correct :).
>>
> 
> The spec refers to RFC 4646. RFC 4646 refers to ISO 3166 which uses
> capital letters for the two letter region codes.
> 
> In RFC 4646 you can find the following example:
> 
> en-US (English as used in the United States)
> 
> The value from edk2 complies with RFC 4646. It simply does not specify a
> region.
> 
> So should I change the value to match edk2?

I think that would remove a potential pitfall, yes. And please change
the Makefile according to my comment above :).


Alex

[U-Boot] [PATCH v3 4/5] efi_loader: EFI_UNICODE_COLLATION_PROTOCOL

[Alexander Graf]

On 03.09.18 18:35, Heinrich Schuchardt wrote:
> The patch implements the EFI_UNICODE_COLLATION_PROTOCOL.
> 
> Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
> ---
> v3:
> 	no change
> v2:
> 	add more comments
> ---
>  MAINTAINERS                            |   2 +
>  include/cp1250.h                       |  40 +++
>  include/cp437.h                        |  40 +++
>  include/efi_api.h                      |  21 ++
>  include/efi_loader.h                   |   5 +
>  lib/efi_loader/Makefile                |  18 +-
>  lib/efi_loader/efi_boottime.c          |   6 +
>  lib/efi_loader/efi_unicode_collation.c | 329 +++++++++++++++++++++++++
>  8 files changed, 457 insertions(+), 4 deletions(-)
>  create mode 100644 include/cp1250.h
>  create mode 100644 include/cp437.h
>  create mode 100644 lib/efi_loader/efi_unicode_collation.c
> 
> diff --git a/MAINTAINERS b/MAINTAINERS
> index 8c9cd83347..67b015205c 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -382,6 +382,8 @@ F:	doc/README.uefi
>  F:	doc/README.iscsi
>  F:	Documentation/efi.rst
>  F:	include/capitalization.h
> +F:	include/cp1250.h
> +F:	include/cp437.h
>  F:	include/efi*
>  F:	include/pe.h
>  F:	include/asm-generic/pe.h
> diff --git a/include/cp1250.h b/include/cp1250.h
> new file mode 100644
> index 0000000000..adacf8a958
> --- /dev/null
> +++ b/include/cp1250.h
> @@ -0,0 +1,40 @@
> +/* SPDX-License-Identifier: GPL-2.0+ */
> +
> +/*
> + * Constant CP1250 contains the Unicode code points for characters 0x80 - 0xff
> + * of the code page 1250.
> + */
> +#define CP1250 { \
> +	0x20ac, 0x0000, 0x201a, 0x0000, \
> +	0x201e, 0x2026, 0x2020, 0x2021, \
> +	0x0000, 0x2030, 0x0160, 0x2039, \
> +	0x015a, 0x0164, 0x017d, 0x0179, \
> +	0x0000, 0x2018, 0x2019, 0x201c, \
> +	0x201d, 0x2022, 0x2013, 0x2014, \
> +	0x0000, 0x2122, 0x0161, 0x203a, \
> +	0x015b, 0x0165, 0x017e, 0x017a, \
> +	0x00a0, 0x02c7, 0x02d8, 0x0141, \
> +	0x00a4, 0x0104, 0x00a6, 0x00a7, \
> +	0x00a8, 0x00a9, 0x015e, 0x00ab, \
> +	0x00ac, 0x00ad, 0x00ae, 0x017b, \
> +	0x00b0, 0x00b1, 0x02db, 0x0142, \
> +	0x00b4, 0x00b5, 0x00b6, 0x00b7, \
> +	0x00b8, 0x0105, 0x015f, 0x00bb, \
> +	0x013d, 0x02dd, 0x013e, 0x017c, \
> +	0x0154, 0x00c1, 0x00c2, 0x0102, \
> +	0x00c4, 0x0139, 0x0106, 0x00c7, \
> +	0x010c, 0x00c9, 0x0118, 0x00cb, \
> +	0x011a, 0x00cd, 0x00ce, 0x010e, \
> +	0x0110, 0x0143, 0x0147, 0x00d3, \
> +	0x00d4, 0x0150, 0x00d6, 0x00d7, \
> +	0x0158, 0x016e, 0x00da, 0x0170, \
> +	0x00dc, 0x00dd, 0x0162, 0x00df, \
> +	0x0155, 0x00e1, 0x00e2, 0x0103, \
> +	0x00e4, 0x013a, 0x0107, 0x00e7, \
> +	0x010d, 0x00e9, 0x0119, 0x00eb, \
> +	0x011b, 0x00ed, 0x00ee, 0x010f, \
> +	0x0111, 0x0144, 0x0148, 0x00f3, \
> +	0x00f4, 0x0151, 0x00f6, 0x00f7, \
> +	0x0159, 0x016f, 0x00fa, 0x0171, \
> +	0x00fc, 0x00fd, 0x0163, 0x02d9, \
> +}
> diff --git a/include/cp437.h b/include/cp437.h
> new file mode 100644
> index 0000000000..0b2b97132e
> --- /dev/null
> +++ b/include/cp437.h
> @@ -0,0 +1,40 @@
> +/* SPDX-License-Identifier: GPL-2.0+ */
> +
> +/*
> + * Constant CP437 contains the Unicode code points for characters 0x80 - 0xff
> + * of the code page 437.
> + */
> +#define CP437 { \
> +	0x00c7, 0x00fc, 0x00e9, 0x00e2, \
> +	0x00e4, 0x00e0, 0x00e5, 0x00e7, \
> +	0x00ea, 0x00eb, 0x00e8, 0x00ef, \
> +	0x00ee, 0x00ec, 0x00c4, 0x00c5, \
> +	0x00c9, 0x00e6, 0x00c6, 0x00f4, \
> +	0x00f6, 0x00f2, 0x00fb, 0x00f9, \
> +	0x00ff, 0x00d6, 0x00dc, 0x00a2, \
> +	0x00a3, 0x00a5, 0x20a7, 0x0192, \
> +	0x00e1, 0x00ed, 0x00f3, 0x00fa, \
> +	0x00f1, 0x00d1, 0x00aa, 0x00ba, \
> +	0x00bf, 0x2310, 0x00ac, 0x00bd, \
> +	0x00bc, 0x00a1, 0x00ab, 0x00bb, \
> +	0x2591, 0x2592, 0x2593, 0x2502, \
> +	0x2524, 0x2561, 0x2562, 0x2556, \
> +	0x2555, 0x2563, 0x2551, 0x2557, \
> +	0x255d, 0x255c, 0x255b, 0x2510, \
> +	0x2514, 0x2534, 0x252c, 0x251c, \
> +	0x2500, 0x253c, 0x255e, 0x255f, \
> +	0x255a, 0x2554, 0x2569, 0x2566, \
> +	0x2560, 0x2550, 0x256c, 0x2567, \
> +	0x2568, 0x2564, 0x2565, 0x2559, \
> +	0x2558, 0x2552, 0x2553, 0x256b, \
> +	0x256a, 0x2518, 0x250c, 0x2588, \
> +	0x2584, 0x258c, 0x2590, 0x2580, \
> +	0x03b1, 0x00df, 0x0393, 0x03c0, \
> +	0x03a3, 0x03c3, 0x00b5, 0x03c4, \
> +	0x03a6, 0x0398, 0x03a9, 0x03b4, \
> +	0x221e, 0x03c6, 0x03b5, 0x2229, \
> +	0x2261, 0x00b1, 0x2265, 0x2264, \
> +	0x2320, 0x2321, 0x00f7, 0x2248, \
> +	0x00b0, 0x2219, 0x00b7, 0x221a, \
> +	0x207f, 0x00b2, 0x25a0, 0x00a0, \
> +}
> diff --git a/include/efi_api.h b/include/efi_api.h
> index ebf2a3bc18..1efc448184 100644
> --- a/include/efi_api.h
> +++ b/include/efi_api.h
> @@ -31,6 +31,7 @@ enum efi_timer_delay {
>  	EFI_TIMER_RELATIVE = 2
>  };
>  
> +#define efi_intn_t ssize_t
>  #define efi_uintn_t size_t
>  typedef uint16_t *efi_string_t;
>  
> @@ -958,4 +959,24 @@ struct efi_driver_binding_protocol {
>  	efi_handle_t driver_binding_handle;
>  };
>  
> +#define EFI_UNICODE_COLLATION_PROTOCOL2_GUID \
> +	EFI_GUID(0xa4c751fc, 0x23ae, 0x4c3e, \
> +		 0x92, 0xe9, 0x49, 0x64, 0xcf, 0x63, 0xf3, 0x49)

Looks like the UEFI Shell requires the v1 GUID of the collation
protocol. Please add and advertise that too.


Alex

[U-Boot] [Fwd] [PATCH v3 4/5] efi_loader: EFI_UNICODE_COLLATION_PROTOCOL

[Heinrich Schuchardt]

Somehow delivery failed:
u-boot at lists.denx.de: no valid MX hosts found


-------- Forwarded Message --------
Subject: Re: [PATCH v3 4/5] efi_loader: EFI_UNICODE_COLLATION_PROTOCOL
Date: Tue, 4 Sep 2018 19:12:53 +0200
From: Heinrich Schuchardt <xypron.glpk@gmx.de>
To: Alexander Graf <agraf@suse.de>
CC: Simon Glass <sjg@chromium.org>, Rob Clark <robdclark@gmail.com>,
AKASHI, Takahiro <takahiro.akashi@linaro.org>, u-boot at lists.denx.de,
Leif Lindholm <leif.lindholm@linaro.org>

On 09/04/2018 03:46 PM, Alexander Graf wrote:
> 
> 
> On 03.09.18 18:35, Heinrich Schuchardt wrote:
>> The patch implements the EFI_UNICODE_COLLATION_PROTOCOL.
>>
>> Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
>> ---
>> v3:
>> 	no change
>> v2:
>> 	add more comments
>> ---
>>  MAINTAINERS                            |   2 +
>>  include/cp1250.h                       |  40 +++
>>  include/cp437.h                        |  40 +++
>>  include/efi_api.h                      |  21 ++
>>  include/efi_loader.h                   |   5 +
>>  lib/efi_loader/Makefile                |  18 +-
>>  lib/efi_loader/efi_boottime.c          |   6 +
>>  lib/efi_loader/efi_unicode_collation.c | 329 +++++++++++++++++++++++++
>>  8 files changed, 457 insertions(+), 4 deletions(-)
>>  create mode 100644 include/cp1250.h
>>  create mode 100644 include/cp437.h
>>  create mode 100644 lib/efi_loader/efi_unicode_collation.c
>>
>> diff --git a/MAINTAINERS b/MAINTAINERS
>> index 8c9cd83347..67b015205c 100644
>> --- a/MAINTAINERS
>> +++ b/MAINTAINERS
>> @@ -382,6 +382,8 @@ F:	doc/README.uefi
>>  F:	doc/README.iscsi
>>  F:	Documentation/efi.rst
>>  F:	include/capitalization.h
>> +F:	include/cp1250.h
>> +F:	include/cp437.h
>>  F:	include/efi*
>>  F:	include/pe.h
>>  F:	include/asm-generic/pe.h
>> diff --git a/include/cp1250.h b/include/cp1250.h
>> new file mode 100644
>> index 0000000000..adacf8a958
>> --- /dev/null
>> +++ b/include/cp1250.h
>> @@ -0,0 +1,40 @@
>> +/* SPDX-License-Identifier: GPL-2.0+ */
>> +
>> +/*
>> + * Constant CP1250 contains the Unicode code points for characters 0x80 - 0xff
>> + * of the code page 1250.
>> + */
>> +#define CP1250 { \
>> +	0x20ac, 0x0000, 0x201a, 0x0000, \
>> +	0x201e, 0x2026, 0x2020, 0x2021, \
>> +	0x0000, 0x2030, 0x0160, 0x2039, \
>> +	0x015a, 0x0164, 0x017d, 0x0179, \
>> +	0x0000, 0x2018, 0x2019, 0x201c, \
>> +	0x201d, 0x2022, 0x2013, 0x2014, \
>> +	0x0000, 0x2122, 0x0161, 0x203a, \
>> +	0x015b, 0x0165, 0x017e, 0x017a, \
>> +	0x00a0, 0x02c7, 0x02d8, 0x0141, \
>> +	0x00a4, 0x0104, 0x00a6, 0x00a7, \
>> +	0x00a8, 0x00a9, 0x015e, 0x00ab, \
>> +	0x00ac, 0x00ad, 0x00ae, 0x017b, \
>> +	0x00b0, 0x00b1, 0x02db, 0x0142, \
>> +	0x00b4, 0x00b5, 0x00b6, 0x00b7, \
>> +	0x00b8, 0x0105, 0x015f, 0x00bb, \
>> +	0x013d, 0x02dd, 0x013e, 0x017c, \
>> +	0x0154, 0x00c1, 0x00c2, 0x0102, \
>> +	0x00c4, 0x0139, 0x0106, 0x00c7, \
>> +	0x010c, 0x00c9, 0x0118, 0x00cb, \
>> +	0x011a, 0x00cd, 0x00ce, 0x010e, \
>> +	0x0110, 0x0143, 0x0147, 0x00d3, \
>> +	0x00d4, 0x0150, 0x00d6, 0x00d7, \
>> +	0x0158, 0x016e, 0x00da, 0x0170, \
>> +	0x00dc, 0x00dd, 0x0162, 0x00df, \
>> +	0x0155, 0x00e1, 0x00e2, 0x0103, \
>> +	0x00e4, 0x013a, 0x0107, 0x00e7, \
>> +	0x010d, 0x00e9, 0x0119, 0x00eb, \
>> +	0x011b, 0x00ed, 0x00ee, 0x010f, \
>> +	0x0111, 0x0144, 0x0148, 0x00f3, \
>> +	0x00f4, 0x0151, 0x00f6, 0x00f7, \
>> +	0x0159, 0x016f, 0x00fa, 0x0171, \
>> +	0x00fc, 0x00fd, 0x0163, 0x02d9, \
>> +}
>> diff --git a/include/cp437.h b/include/cp437.h
>> new file mode 100644
>> index 0000000000..0b2b97132e
>> --- /dev/null
>> +++ b/include/cp437.h
>> @@ -0,0 +1,40 @@
>> +/* SPDX-License-Identifier: GPL-2.0+ */
>> +
>> +/*
>> + * Constant CP437 contains the Unicode code points for characters 0x80 - 0xff
>> + * of the code page 437.
>> + */
>> +#define CP437 { \
>> +	0x00c7, 0x00fc, 0x00e9, 0x00e2, \
>> +	0x00e4, 0x00e0, 0x00e5, 0x00e7, \
>> +	0x00ea, 0x00eb, 0x00e8, 0x00ef, \
>> +	0x00ee, 0x00ec, 0x00c4, 0x00c5, \
>> +	0x00c9, 0x00e6, 0x00c6, 0x00f4, \
>> +	0x00f6, 0x00f2, 0x00fb, 0x00f9, \
>> +	0x00ff, 0x00d6, 0x00dc, 0x00a2, \
>> +	0x00a3, 0x00a5, 0x20a7, 0x0192, \
>> +	0x00e1, 0x00ed, 0x00f3, 0x00fa, \
>> +	0x00f1, 0x00d1, 0x00aa, 0x00ba, \
>> +	0x00bf, 0x2310, 0x00ac, 0x00bd, \
>> +	0x00bc, 0x00a1, 0x00ab, 0x00bb, \
>> +	0x2591, 0x2592, 0x2593, 0x2502, \
>> +	0x2524, 0x2561, 0x2562, 0x2556, \
>> +	0x2555, 0x2563, 0x2551, 0x2557, \
>> +	0x255d, 0x255c, 0x255b, 0x2510, \
>> +	0x2514, 0x2534, 0x252c, 0x251c, \
>> +	0x2500, 0x253c, 0x255e, 0x255f, \
>> +	0x255a, 0x2554, 0x2569, 0x2566, \
>> +	0x2560, 0x2550, 0x256c, 0x2567, \
>> +	0x2568, 0x2564, 0x2565, 0x2559, \
>> +	0x2558, 0x2552, 0x2553, 0x256b, \
>> +	0x256a, 0x2518, 0x250c, 0x2588, \
>> +	0x2584, 0x258c, 0x2590, 0x2580, \
>> +	0x03b1, 0x00df, 0x0393, 0x03c0, \
>> +	0x03a3, 0x03c3, 0x00b5, 0x03c4, \
>> +	0x03a6, 0x0398, 0x03a9, 0x03b4, \
>> +	0x221e, 0x03c6, 0x03b5, 0x2229, \
>> +	0x2261, 0x00b1, 0x2265, 0x2264, \
>> +	0x2320, 0x2321, 0x00f7, 0x2248, \
>> +	0x00b0, 0x2219, 0x00b7, 0x221a, \
>> +	0x207f, 0x00b2, 0x25a0, 0x00a0, \
>> +}
>> diff --git a/include/efi_api.h b/include/efi_api.h
>> index ebf2a3bc18..1efc448184 100644
>> --- a/include/efi_api.h
>> +++ b/include/efi_api.h
>> @@ -31,6 +31,7 @@ enum efi_timer_delay {
>>  	EFI_TIMER_RELATIVE = 2
>>  };
>>  
>> +#define efi_intn_t ssize_t
>>  #define efi_uintn_t size_t
>>  typedef uint16_t *efi_string_t;
>>  
>> @@ -958,4 +959,24 @@ struct efi_driver_binding_protocol {
>>  	efi_handle_t driver_binding_handle;
>>  };
>>  
>> +#define EFI_UNICODE_COLLATION_PROTOCOL2_GUID \
>> +	EFI_GUID(0xa4c751fc, 0x23ae, 0x4c3e, \
>> +		 0x92, 0xe9, 0x49, 0x64, 0xcf, 0x63, 0xf3, 0x49)
> 
> Looks like the UEFI Shell requires the v1 GUID of the collation
> protocol. Please add and advertise that too.

Hello Alex,

that protocol does not exist in the UEFI 2.7 spec and it is not required
to run EDK2 ./ShellBinPkg/UefiShell/AArch64/Shell.efi on
qemu_arm64_defconfig.

I just removed your patch
XXX efi_loader collation: Expose v1 GUID as well
from https://github.com/agraf/u-boot/commits/ebbr-demo
and I am still able to run Shell.efi.

Best regards

Heinrich

> 
> 
> Alex
> 

[U-Boot] [PATCH v3 4/5] efi_loader: EFI_UNICODE_COLLATION_PROTOCOL

[Alexander Graf]

On 04.09.18 19:12, Heinrich Schuchardt wrote:
> On 09/04/2018 03:46 PM, Alexander Graf wrote:
>>
>>
>> On 03.09.18 18:35, Heinrich Schuchardt wrote:
>>> The patch implements the EFI_UNICODE_COLLATION_PROTOCOL.
>>>
>>> Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
>>> ---
>>> v3:
>>> 	no change
>>> v2:
>>> 	add more comments
>>> ---
>>>  MAINTAINERS                            |   2 +
>>>  include/cp1250.h                       |  40 +++
>>>  include/cp437.h                        |  40 +++
>>>  include/efi_api.h                      |  21 ++
>>>  include/efi_loader.h                   |   5 +
>>>  lib/efi_loader/Makefile                |  18 +-
>>>  lib/efi_loader/efi_boottime.c          |   6 +
>>>  lib/efi_loader/efi_unicode_collation.c | 329 +++++++++++++++++++++++++
>>>  8 files changed, 457 insertions(+), 4 deletions(-)
>>>  create mode 100644 include/cp1250.h
>>>  create mode 100644 include/cp437.h
>>>  create mode 100644 lib/efi_loader/efi_unicode_collation.c
>>>
>>> diff --git a/MAINTAINERS b/MAINTAINERS
>>> index 8c9cd83347..67b015205c 100644
>>> --- a/MAINTAINERS
>>> +++ b/MAINTAINERS
>>> @@ -382,6 +382,8 @@ F:	doc/README.uefi
>>>  F:	doc/README.iscsi
>>>  F:	Documentation/efi.rst
>>>  F:	include/capitalization.h
>>> +F:	include/cp1250.h
>>> +F:	include/cp437.h
>>>  F:	include/efi*
>>>  F:	include/pe.h
>>>  F:	include/asm-generic/pe.h
>>> diff --git a/include/cp1250.h b/include/cp1250.h
>>> new file mode 100644
>>> index 0000000000..adacf8a958
>>> --- /dev/null
>>> +++ b/include/cp1250.h
>>> @@ -0,0 +1,40 @@
>>> +/* SPDX-License-Identifier: GPL-2.0+ */
>>> +
>>> +/*
>>> + * Constant CP1250 contains the Unicode code points for characters 0x80 - 0xff
>>> + * of the code page 1250.
>>> + */
>>> +#define CP1250 { \
>>> +	0x20ac, 0x0000, 0x201a, 0x0000, \
>>> +	0x201e, 0x2026, 0x2020, 0x2021, \
>>> +	0x0000, 0x2030, 0x0160, 0x2039, \
>>> +	0x015a, 0x0164, 0x017d, 0x0179, \
>>> +	0x0000, 0x2018, 0x2019, 0x201c, \
>>> +	0x201d, 0x2022, 0x2013, 0x2014, \
>>> +	0x0000, 0x2122, 0x0161, 0x203a, \
>>> +	0x015b, 0x0165, 0x017e, 0x017a, \
>>> +	0x00a0, 0x02c7, 0x02d8, 0x0141, \
>>> +	0x00a4, 0x0104, 0x00a6, 0x00a7, \
>>> +	0x00a8, 0x00a9, 0x015e, 0x00ab, \
>>> +	0x00ac, 0x00ad, 0x00ae, 0x017b, \
>>> +	0x00b0, 0x00b1, 0x02db, 0x0142, \
>>> +	0x00b4, 0x00b5, 0x00b6, 0x00b7, \
>>> +	0x00b8, 0x0105, 0x015f, 0x00bb, \
>>> +	0x013d, 0x02dd, 0x013e, 0x017c, \
>>> +	0x0154, 0x00c1, 0x00c2, 0x0102, \
>>> +	0x00c4, 0x0139, 0x0106, 0x00c7, \
>>> +	0x010c, 0x00c9, 0x0118, 0x00cb, \
>>> +	0x011a, 0x00cd, 0x00ce, 0x010e, \
>>> +	0x0110, 0x0143, 0x0147, 0x00d3, \
>>> +	0x00d4, 0x0150, 0x00d6, 0x00d7, \
>>> +	0x0158, 0x016e, 0x00da, 0x0170, \
>>> +	0x00dc, 0x00dd, 0x0162, 0x00df, \
>>> +	0x0155, 0x00e1, 0x00e2, 0x0103, \
>>> +	0x00e4, 0x013a, 0x0107, 0x00e7, \
>>> +	0x010d, 0x00e9, 0x0119, 0x00eb, \
>>> +	0x011b, 0x00ed, 0x00ee, 0x010f, \
>>> +	0x0111, 0x0144, 0x0148, 0x00f3, \
>>> +	0x00f4, 0x0151, 0x00f6, 0x00f7, \
>>> +	0x0159, 0x016f, 0x00fa, 0x0171, \
>>> +	0x00fc, 0x00fd, 0x0163, 0x02d9, \
>>> +}
>>> diff --git a/include/cp437.h b/include/cp437.h
>>> new file mode 100644
>>> index 0000000000..0b2b97132e
>>> --- /dev/null
>>> +++ b/include/cp437.h
>>> @@ -0,0 +1,40 @@
>>> +/* SPDX-License-Identifier: GPL-2.0+ */
>>> +
>>> +/*
>>> + * Constant CP437 contains the Unicode code points for characters 0x80 - 0xff
>>> + * of the code page 437.
>>> + */
>>> +#define CP437 { \
>>> +	0x00c7, 0x00fc, 0x00e9, 0x00e2, \
>>> +	0x00e4, 0x00e0, 0x00e5, 0x00e7, \
>>> +	0x00ea, 0x00eb, 0x00e8, 0x00ef, \
>>> +	0x00ee, 0x00ec, 0x00c4, 0x00c5, \
>>> +	0x00c9, 0x00e6, 0x00c6, 0x00f4, \
>>> +	0x00f6, 0x00f2, 0x00fb, 0x00f9, \
>>> +	0x00ff, 0x00d6, 0x00dc, 0x00a2, \
>>> +	0x00a3, 0x00a5, 0x20a7, 0x0192, \
>>> +	0x00e1, 0x00ed, 0x00f3, 0x00fa, \
>>> +	0x00f1, 0x00d1, 0x00aa, 0x00ba, \
>>> +	0x00bf, 0x2310, 0x00ac, 0x00bd, \
>>> +	0x00bc, 0x00a1, 0x00ab, 0x00bb, \
>>> +	0x2591, 0x2592, 0x2593, 0x2502, \
>>> +	0x2524, 0x2561, 0x2562, 0x2556, \
>>> +	0x2555, 0x2563, 0x2551, 0x2557, \
>>> +	0x255d, 0x255c, 0x255b, 0x2510, \
>>> +	0x2514, 0x2534, 0x252c, 0x251c, \
>>> +	0x2500, 0x253c, 0x255e, 0x255f, \
>>> +	0x255a, 0x2554, 0x2569, 0x2566, \
>>> +	0x2560, 0x2550, 0x256c, 0x2567, \
>>> +	0x2568, 0x2564, 0x2565, 0x2559, \
>>> +	0x2558, 0x2552, 0x2553, 0x256b, \
>>> +	0x256a, 0x2518, 0x250c, 0x2588, \
>>> +	0x2584, 0x258c, 0x2590, 0x2580, \
>>> +	0x03b1, 0x00df, 0x0393, 0x03c0, \
>>> +	0x03a3, 0x03c3, 0x00b5, 0x03c4, \
>>> +	0x03a6, 0x0398, 0x03a9, 0x03b4, \
>>> +	0x221e, 0x03c6, 0x03b5, 0x2229, \
>>> +	0x2261, 0x00b1, 0x2265, 0x2264, \
>>> +	0x2320, 0x2321, 0x00f7, 0x2248, \
>>> +	0x00b0, 0x2219, 0x00b7, 0x221a, \
>>> +	0x207f, 0x00b2, 0x25a0, 0x00a0, \
>>> +}
>>> diff --git a/include/efi_api.h b/include/efi_api.h
>>> index ebf2a3bc18..1efc448184 100644
>>> --- a/include/efi_api.h
>>> +++ b/include/efi_api.h
>>> @@ -31,6 +31,7 @@ enum efi_timer_delay {
>>>  	EFI_TIMER_RELATIVE = 2
>>>  };
>>>  
>>> +#define efi_intn_t ssize_t
>>>  #define efi_uintn_t size_t
>>>  typedef uint16_t *efi_string_t;
>>>  
>>> @@ -958,4 +959,24 @@ struct efi_driver_binding_protocol {
>>>  	efi_handle_t driver_binding_handle;
>>>  };
>>>  
>>> +#define EFI_UNICODE_COLLATION_PROTOCOL2_GUID \
>>> +	EFI_GUID(0xa4c751fc, 0x23ae, 0x4c3e, \
>>> +		 0x92, 0xe9, 0x49, 0x64, 0xcf, 0x63, 0xf3, 0x49)
>>
>> Looks like the UEFI Shell requires the v1 GUID of the collation
>> protocol. Please add and advertise that too.
> 
> Hello Alex,
> 
> that protocol does not exist in the UEFI 2.7 spec and it is not required
> to run EDK2 ./ShellBinPkg/UefiShell/AArch64/Shell.efi on
> qemu_arm64_defconfig.
> 
> I just removed your patch
> XXX efi_loader collation: Expose v1 GUID as well
> from https://github.com/agraf/u-boot/commits/ebbr-demo
> and I am still able to run Shell.efi.

I just double checked and Shell.efi definitely fails for me without the
v1 protocol patch:

  $ wget
'https://github.com/ARM-software/edk2/raw/master/ShellBinPkg/UefiShell/AArch64/Shell.efi'
  $ qemu-system-aarch64 -M virt -cpu cortex-a57 -kernel u-boot
-nographic -m 2G -net nic,model=e1000 -net user,tftp=.,bootfile=Shell.efi


Alex

[U-Boot] [PATCH v3 4/5] efi_loader: EFI_UNICODE_COLLATION_PROTOCOL

[Leif Lindholm]

On Wed, Sep 05, 2018 at 12:04:04AM +0200, Alexander Graf wrote:
> > I just removed your patch
> > XXX efi_loader collation: Expose v1 GUID as well
> > from https://github.com/agraf/u-boot/commits/ebbr-demo
> > and I am still able to run Shell.efi.
> 
> I just double checked and Shell.efi definitely fails for me without the
> v1 protocol patch:
> 
>   $ wget
> 'https://github.com/ARM-software/edk2/raw/master/ShellBinPkg/UefiShell/AArch64/Shell.efi'

I'm not even going to ask why you are using that repository.
I want to know how you _found_ that repository!?

Latest commit 16 July 2015.

Please use only https://github.com/tianocore/edk2.

/
    Leif

[U-Boot] [PATCH v3 4/5] efi_loader: EFI_UNICODE_COLLATION_PROTOCOL

[Alexander Graf]

On 05.09.18 00:28, Leif Lindholm wrote:
> On Wed, Sep 05, 2018 at 12:04:04AM +0200, Alexander Graf wrote:
>>> I just removed your patch
>>> XXX efi_loader collation: Expose v1 GUID as well
>>> from https://github.com/agraf/u-boot/commits/ebbr-demo
>>> and I am still able to run Shell.efi.
>>
>> I just double checked and Shell.efi definitely fails for me without the
>> v1 protocol patch:
>>
>>   $ wget
>> 'https://github.com/ARM-software/edk2/raw/master/ShellBinPkg/UefiShell/AArch64/Shell.efi'
> 
> I'm not even going to ask why you are using that repository.
> I want to know how you _found_ that repository!?
> 
> Latest commit 16 July 2015.

Yeah, I just stumbled over that one with google I guess.

> Please use only https://github.com/tianocore/edk2.

That binary does indeed work without collation v1.


Alex

[U-Boot] [PATCH v3 5/5] efi_selftest: EFI_UNICODE_COLLATION_PROTOCOL

[Heinrich Schuchardt]

Provide a unit test for the EFI_UNICODE_COLLATION_PROTOCOL.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
---
v3:
	no change
v2:
	avoid lines over 80 characters
---
 lib/efi_selftest/Makefile                     |   1 +
 .../efi_selftest_unicode_collation.c          | 260 ++++++++++++++++++
 2 files changed, 261 insertions(+)
 create mode 100644 lib/efi_selftest/efi_selftest_unicode_collation.c

diff --git a/lib/efi_selftest/Makefile b/lib/efi_selftest/Makefile
index 86aa72b0e0..80bc5d8a64 100644
--- a/lib/efi_selftest/Makefile
+++ b/lib/efi_selftest/Makefile
@@ -31,6 +31,7 @@ efi_selftest_snp.o \
 efi_selftest_textinput.o \
 efi_selftest_textoutput.o \
 efi_selftest_tpl.o \
+efi_selftest_unicode_collation.o \
 efi_selftest_util.o \
 efi_selftest_variables.o \
 efi_selftest_watchdog.o
diff --git a/lib/efi_selftest/efi_selftest_unicode_collation.c b/lib/efi_selftest/efi_selftest_unicode_collation.c
new file mode 100644
index 0000000000..9765bd3e44
--- /dev/null
+++ b/lib/efi_selftest/efi_selftest_unicode_collation.c
@@ -0,0 +1,260 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * efi_selftest_unicode_collation
+ *
+ * Copyright (c) 2018 Heinrich Schuchardt <xypron.glpk@gmx.de>
+ *
+ * Test unicode collation protocol.
+ */
+
+#include <efi_selftest.h>
+
+static const efi_guid_t unicode_collation_protocol_guid =
+	EFI_UNICODE_COLLATION_PROTOCOL2_GUID;
+
+static struct efi_boot_services *boottime;
+
+static struct efi_unicode_collation_protocol *unicode_collation_protocol;
+
+/**
+ * setup() - setup unit test.
+ *
+ * @handle:	handle of the loaded image
+ * @systable:	system table
+ * ReturnValue:	EFI_ST_SUCCESS for success
+ */
+static int setup(const efi_handle_t handle,
+		 const struct efi_system_table *systable)
+{
+	efi_status_t ret;
+
+	boottime = systable->boottime;
+
+	ret = boottime->locate_protocol(&unicode_collation_protocol_guid, NULL,
+					(void **)&unicode_collation_protocol);
+	if (ret != EFI_SUCCESS) {
+		unicode_collation_protocol = NULL;
+		efi_st_error("Unicode collation protocol is not available.\n");
+		return EFI_ST_FAILURE;
+	}
+
+	return EFI_ST_SUCCESS;
+}
+
+static int test_stri_coll(void)
+{
+	efi_intn_t ret;
+	u16 c1[] = L"first";
+	u16 c2[] = L"FIRST";
+	u16 c3[] = L"second";
+
+	ret = unicode_collation_protocol->stri_coll(unicode_collation_protocol,
+						    c1, c2);
+	if (ret) {
+		efi_st_error(
+			"stri_coll(\"%ps\", \"%ps\") = %zu\n", c1, c2, ret);
+		return EFI_ST_FAILURE;
+	}
+
+	ret = unicode_collation_protocol->stri_coll(unicode_collation_protocol,
+						    c1, c3);
+	if (ret >= 0) {
+		efi_st_error(
+			"stri_coll(\"%ps\", \"%ps\") = %zu\n", c1, c3, ret);
+		return EFI_ST_FAILURE;
+	}
+
+	ret = unicode_collation_protocol->stri_coll(unicode_collation_protocol,
+						    c3, c1);
+	if (ret <= 0) {
+		efi_st_error(
+			"stri_coll(\"%ps\", \"%ps\") = %zu\n", c3, c1, ret);
+		return EFI_ST_FAILURE;
+	}
+
+	return EFI_ST_SUCCESS;
+}
+
+static int test_metai_match(void)
+{
+	bool ret;
+	const u16 c[] = L"Das U-Boot";
+
+	ret = unicode_collation_protocol->metai_match(
+		unicode_collation_protocol, c, L"*");
+	if (!ret) {
+		efi_st_error("metai_match returned %u\n", ret);
+		return EFI_ST_FAILURE;
+	}
+
+	ret = unicode_collation_protocol->metai_match(
+		unicode_collation_protocol, c, L"Da[rstu] U-Boot");
+	if (!ret) {
+		efi_st_error("metai_match returned %u\n", ret);
+		return EFI_ST_FAILURE;
+	}
+
+	ret = unicode_collation_protocol->metai_match(
+		unicode_collation_protocol, c, L"Da[q-v] U-Boot");
+	if (!ret) {
+		efi_st_error("metai_match returned %u\n", ret);
+		return EFI_ST_FAILURE;
+	}
+
+	ret = unicode_collation_protocol->metai_match(
+		unicode_collation_protocol, c, L"Da? U-Boot");
+	if (!ret) {
+		efi_st_error("metai_match returned %u\n", ret);
+		return EFI_ST_FAILURE;
+	}
+
+	ret = unicode_collation_protocol->metai_match(
+		unicode_collation_protocol, c, L"D*Bo*t");
+	if (!ret) {
+		efi_st_error("metai_match returned %u\n", ret);
+		return EFI_ST_FAILURE;
+	}
+
+	ret = unicode_collation_protocol->metai_match(
+		unicode_collation_protocol, c, L"Da[xyz] U-Boot");
+	if (ret) {
+		efi_st_error("metai_match returned %u\n", ret);
+		return EFI_ST_FAILURE;
+	}
+
+	ret = unicode_collation_protocol->metai_match(
+		unicode_collation_protocol, c, L"Da[a-d] U-Boot");
+	if (ret) {
+		efi_st_error("metai_match returned %u\n", ret);
+		return EFI_ST_FAILURE;
+	}
+
+	ret = unicode_collation_protocol->metai_match(
+		unicode_collation_protocol, c, L"Da?? U-Boot");
+	if (ret) {
+		efi_st_error("metai_match returned %u\n", ret);
+		return EFI_ST_FAILURE;
+	}
+
+	ret = unicode_collation_protocol->metai_match(
+		unicode_collation_protocol, c, L"D*Bo*tt");
+	if (ret) {
+		efi_st_error("metai_match returned %u\n", ret);
+		return EFI_ST_FAILURE;
+	}
+
+	return EFI_ST_SUCCESS;
+}
+
+static int test_str_lwr(void)
+{
+	u16 c[] = L"U-Boot";
+
+	unicode_collation_protocol->str_lwr(unicode_collation_protocol, c);
+	if (efi_st_strcmp_16_8(c, "u-boot")) {
+		efi_st_error("str_lwr returned \"%ps\"\n", c);
+		return EFI_ST_FAILURE;
+	}
+
+	return EFI_ST_SUCCESS;
+}
+
+static int test_str_upr(void)
+{
+	u16 c[] = L"U-Boot";
+
+	unicode_collation_protocol->str_upr(unicode_collation_protocol, c);
+	if (efi_st_strcmp_16_8(c, "U-BOOT")) {
+		efi_st_error("str_lwr returned \"%ps\"\n", c);
+		return EFI_ST_FAILURE;
+	}
+
+	return EFI_ST_SUCCESS;
+}
+
+static int test_fat_to_str(void)
+{
+	u16 str[16];
+
+	boottime->set_mem(str, sizeof(str), 0);
+	unicode_collation_protocol->fat_to_str(unicode_collation_protocol, 6,
+					       "U-BOOT", str);
+	if (efi_st_strcmp_16_8(str, "U-BOOT")) {
+		efi_st_error("fat_to_str returned \"%ps\"\n", str);
+		return EFI_ST_FAILURE;
+	}
+
+	return EFI_ST_SUCCESS;
+}
+
+static int test_str_to_fat(void)
+{
+	char fat[16];
+	bool ret;
+
+	boottime->set_mem(fat, sizeof(fat), 0);
+	ret = unicode_collation_protocol->str_to_fat(unicode_collation_protocol,
+						     L"U -Boo.t", 6, fat);
+	if (ret || efi_st_strcmp_16_8(L"U-BOOT", fat)) {
+		efi_st_error("str_to_fat returned %u, \"%s\"\n", ret, fat);
+		return EFI_ST_FAILURE;
+	}
+
+	boottime->set_mem(fat, 16, 0);
+	ret = unicode_collation_protocol->str_to_fat(unicode_collation_protocol,
+						     L"U\\Boot", 6, fat);
+	if (!ret || efi_st_strcmp_16_8(L"U_BOOT", fat)) {
+		efi_st_error("str_to_fat returned %u, \"%s\"\n", ret, fat);
+		return EFI_ST_FAILURE;
+	}
+
+	return EFI_ST_SUCCESS;
+}
+
+/**
+ * execute() - Execute unit test.
+ *
+ * ReturnValue:	EFI_ST_SUCCESS for success
+ */
+static int execute(void)
+{
+	int ret;
+
+	if (!unicode_collation_protocol) {
+		efi_st_printf("Unicode collation protocol missing\n");
+		return EFI_ST_FAILURE;
+	}
+
+	ret = test_stri_coll();
+	if (ret != EFI_ST_SUCCESS)
+		return ret;
+
+	ret = test_metai_match();
+	if (ret != EFI_ST_SUCCESS)
+		return ret;
+
+	ret = test_str_lwr();
+	if (ret != EFI_ST_SUCCESS)
+		return ret;
+
+	ret = test_str_upr();
+	if (ret != EFI_ST_SUCCESS)
+		return ret;
+
+	ret = test_fat_to_str();
+	if (ret != EFI_ST_SUCCESS)
+		return ret;
+
+	ret = test_str_to_fat();
+	if (ret != EFI_ST_SUCCESS)
+		return ret;
+
+	return EFI_ST_SUCCESS;
+}
+
+EFI_UNIT_TEST(unicoll) = {
+	.name = "unicode collation",
+	.phase = EFI_EXECUTE_BEFORE_BOOTTIME_EXIT,
+	.execute = execute,
+	.setup = setup,
+};
-- 
2.18.0