xref: /freebsd-src/share/i18n/csmapper/APPLE/HEBREW%UCS.src (revision d0b2dbfa0ecf2bbc9709efc5e20baf8e4b44bbbf)
1
2TYPE		ROWCOL
3NAME		HEBREW/UCS
4SRC_ZONE	0x00-0xFF
5OOB_MODE	ILSEQ
6DST_ILSEQ	0xFFFE
7DST_UNIT_BITS	16
8
9BEGIN_MAP
10#=======================================================================
11#   File name:  HEBREW.TXT
12#
13#   Contents:   Map (external version) from Mac OS Hebrew
14#               character set to Unicode 2.1 and later.
15#
16#   Copyright:  (c) 1995-2002, 2005 by Apple Computer, Inc., all rights
17#               reserved.
18#
19#   Contact:    charsets@apple.com
20#
21#   Changes:
22#
23#       c02  2005-Apr-05    Update header comments; add section on
24#                           roundtrip considerations. Matches internal
25#                           xml <c1.4> and Text Encoding Converter 2.0.
26#      b3,c1 2002-Dec-19    Don't require left-right context for digits
27#                           0x30-0x39. Change mapping of 0x81 to use
28#                           decomposition. Reverse the mappings of 0xA8,
29#                           0xA9. Update URLs, notes. Matches internal
30#                           utom<b7>.
31#       b02  1999-Sep-22    Update contact e-mail address. Matches
32#                           internal utom<b1>, ufrm<b1>, and Text
33#                           Encoding Converter version 1.5.
34#       n03  1998-Feb-05    Show required Unicode character
35#                           directionality in a different way. Update
36#                           mappings for 0xC0 and 0xDE to use
37#                           transcoding hints; matches internal utom<n6>,
38#                           ufrm<n20>, and Text Encoding Converter
39#                           version 1.3. Rewrite header comments.
40#       n01  1995-Nov-15    First version. Matches internal ufrm<n8>.
41#
42# Standard header:
43# ----------------
44#
45#   Apple, the Apple logo, and Macintosh are trademarks of Apple
46#   Computer, Inc., registered in the United States and other countries.
47#   Unicode is a trademark of Unicode Inc. For the sake of brevity,
48#   throughout this document, "Macintosh" can be used to refer to
49#   Macintosh computers and "Unicode" can be used to refer to the
50#   Unicode standard.
51#
52#   Apple Computer, Inc. ("Apple") makes no warranty or representation,
53#   either express or implied, with respect to this document and the
54#   included data, its quality, accuracy, or fitness for a particular
55#   purpose. In no event will Apple be liable for direct, indirect,
56#   special, incidental, or consequential damages resulting from any
57#   defect or inaccuracy in this document or the included data.
58#
59#   These mapping tables and character lists are subject to change.
60#   The latest tables should be available from the following:
61#
62#   <http://www.unicode.org/Public/MAPPINGS/VENDORS/APPLE/>
63#
64#   For general information about Mac OS encodings and these mapping
65#   tables, see the file "README.TXT".
66#
67# Format:
68# -------
69#
70#   Three tab-separated columns;
71#   '#' begins a comment which continues to the end of the line.
72#     Column #1 is the Mac OS Hebrew code (in hex as 0xNN).
73#     Column #2 is the corresponding Unicode or Unicode sequence (in
74#       hex as 0xNNNN, 0xNNNN+0xNNNN, etc.). Sequences of up to 3
75#       Unicode characters are used here. A single Unicode character
76#       may be preceded by a tag indicating required directionality
77#       (i.e. 0xNNNN or 0xNNNN).
78#     Column #3 is a comment containing the Unicode name.
79#
80#   The entries are in Mac OS Hebrew code order.
81#
82#   Some of these mappings require the use of corporate characters.
83#   See the file "CORPCHAR.TXT" and notes below.
84#
85#   Control character mappings are not shown in this table, following
86#   the conventions of the standard UTC mapping tables. However, the
87#   Mac OS Hebrew character set uses the standard control characters at
88#   0x00-0x1F and 0x7F.
89#
90# Notes on Mac OS Hebrew:
91# -----------------------
92#
93#   This is a legacy Mac OS encoding; in the Mac OS X Carbon and Cocoa
94#   environments, it is only supported via transcoding to and from
95#   Unicode.
96#
97#   1. General
98#
99#   The Mac OS Hebrew character set supports the Hebrew and Yiddish
100#   languages. It incorporates the Hebrew letter repertoire of
101#   ISO 8859-8, and uses the same code points for them, 0xE0-0xFA.
102#   It also incorporates the ASCII character set. In addition, the
103#   Mac OS Hebrew character set includes the following:
104#
105#   - Hebrew points (nikud marks) at 0xC6, 0xCB-0xCF and 0xD8-0xDF.
106#     These are non-spacing combining marks. Note that the RAFE point
107#     at 0xD8 is not displayed correctly in some fonts, and cannot be
108#     typed using the keyboard layouts in the current Hebrew localized
109#     systems. Also note: The character given in Unicode as QAMATS
110#     (U+05B8) actually refers to two different sounds, depending on
111#     context. For example, when ALEF is followed by QAMATS, the QAMATS
112#     can actually refer to two different sounds depending on the
113#     following letters. The Mac OS Hebrew character set separately
114#     encodes these two sounds for the same graphic shape, as "qamats"
115#     (0xCB) and "qamats qatan" (0xDE). The "qamats" character is more
116#     common, so it is mapped to the Unicode QAMATS; "qamats qatan" can
117#     only be used with a limited number of characters, and it is
118#     mapped using a corporate-zone variant tag (see below).
119#
120#   - Various Hebrew ligatures at 0x81, 0xC0, 0xC7, 0xC8, 0xD6, and
121#     0xD7. Also note that the Yiddish YOD YOD PATAH ligature at 0x81
122#     is missing in some fonts.
123#
124#   - The NEW SHEQEL SIGN at 0xA6.
125#
126#   - Latin characters with diacritics at 0x80 and 0x82-0x9F. However,
127#     most of these cannot be typed using the keyboard layouts in the
128#     Hebrew localized systems.
129#
130#   - Right-left versions of certain ASCII punctuation, symbols and
131#     digits: 0xA0-0xA5, 0xA7-0xBF, 0xFB-0xFF. See below.
132#
133#   - Miscellaneous additional punctuation at 0xC1, 0xC9, 0xCA, and
134#     0xD0-0xD5. There is a variant of the Hebrew encoding in which
135#     the LEFT SINGLE QUOTATION MARK at 0xD4 is replaced by FIGURE
136#     SPACE. The glyphs for some of the other punctuation characters
137#     are missing in some fonts.
138#
139#   - Four obsolete characters at 0xC2-0xC5 known as canorals (not to
140#     be confused with cantillation marks!). These were used for
141#     manual positioning of nikud marks before System 7.1 (at which
142#     point nikud positioning became automatic with WorldScript.).
143#
144#   2. Directional characters and roundtrip fidelity
145#
146#   The Mac OS Hebrew character set was developed around 1987. At that
147#   time the bidirectional line line layout algorithm used in the Mac OS
148#   Hebrew system was fairly simple; it used only a few direction
149#   classes (instead of the 19 now used in the Unicode bidirectional
150#   algorithm). In order to permit users to handle some tricky layou
151#   problems, certain punctuation, symbol, and digit characters have
152#   duplicate code points, one with a left-right direction attribute and
153#   the other with a right-left direction attribute.
154#
155#   For example, plus sign is encoded at 0x2B with a left-right
156#   attribute, and at 0xAB with a right-left attribute. However, there
157#   is only one PLUS SIGN character in Unicode. This leads to some
158#   interesting problems when mapping between Mac OS Hebrew and Unicode;
159#   see below.
160#
161#   A related problem is that even when a particular character is
162#   encoded only once in Mac OS Hebrew, it may have a different
163#   direction attribute than the corresponding Unicode character.
164#
165#   For example, the Mac OS Hebrew character at 0xC9 is HORIZONTAL
166#   ELLIPSIS with strong right-left direction. However, the Unicode
167#   character HORIZONTAL ELLIPSIS has direction class neutral.
168#
169#   3. Font variants
170#
171#   The table in this file gives the Unicode mappings for the standard
172#   Mac OS Hebrew encoding. This encoding is supported by many of the
173#   Apple fonts (including all of the fonts in the Hebrew Language Kit),
174#   and is the encoding supported by the text processing utilities.
175#   However, some TrueType fonts provided with the localized Hebrew
176#   system implement a slightly different encoding; the difference is
177#   only in one code point, 0xD4. For the standard variant, this is:
178#     0xD4 -> 0x2018  LEFT SINGLE QUOTATION MARK, right-left
179#
180#   The TrueType variant is used by the following TrueType fonts from
181#   the localized system: Caesarea, Carmel Book, Gilboa, Ramat Sharon,
182#   and Sinai Book. For these, 0xD4 is as follows:
183#     0xD4 -> 0x2007  FIGURE SPACE, right-left
184#
185# Unicode mapping issues and notes:
186# ---------------------------------
187#
188#   1. Matching the direction of Mac OS Hebrew characters
189#
190#   When Mac OS Hebrew encodes a character twice but with different
191#   direction attributes for the two code points - as in the case of
192#   plus sign mentioned above - we need a way to map both Mac OS Hebrew
193#   code points to Unicode and back again without loss of information.
194#   With the plus sign, for example, mapping one of the Mac OS Hebrew
195#   characters to a code in the Unicode corporate use zone is
196#   undesirable, since both of the plus sign characters are likely to
197#   be used in text that is interchanged.
198#
199#   The problem is solved with the use of direction override characters
200#   and direction-dependent mappings. When mapping from Mac OS Hebrew
201#   to Unicode, we use direction overrides as necessary to force the
202#   direction of the resulting Unicode characters.
203#
204#   The required direction is indicated by a direction tag in the
205#   mappings. A tag of <LR> means the corresponding Unicode character
206#   must have a strong left-right context, and a tag of <RL> indicates
207#   a right-left context.
208#
209#   For example, the mapping of 0x2B is given as 0x002B; the
210#   mapping of 0xAB is given as 0x002B. If we map an isolated
211#   instance of 0x2B to Unicode, it should be mapped as follows (LRO
212#   indicates LEFT-RIGHT OVERRIDE, PDF indicates POP DIRECTION
213#   FORMATTING):
214#
215#     0x2B ->  0x202D (LRO) + 0x002B (PLUS SIGN) + 0x202C (PDF)
216#
217#   When mapping several characters in a row that require direction
218#   forcing, the overrides need only be used at the beginning and end.
219#   For example:
220#
221#     0x24 0x20 0x28 0x29 -> 0x202D 0x0024 0x0020 0x0028 0x0029 0x202C
222#
223#   If neutral characters that require direction forcing are already
224#   between strong-direction characters with matching directionality,
225#   then direction overrides need not be used. Direction overrides are
226#   always needed to map the right-left digits at 0xB0-0xB9.
227#
228#   When mapping from Unicode to Mac OS Hebrew, the Unicode
229#   bidirectional algorithm should be used to determine resolved
230#   direction of the Unicode characters. The mapping from Unicode to
231#   Mac OS Hebrew can then be disambiguated by the use of the resolved
232#   direction:
233#
234#     Unicode 0x002B -> Mac OS Hebrew 0x2B (if L) or 0xAB (if R)
235#
236#   However, this also means the direction override characters should
237#   be discarded when mapping from Unicode to Mac OS Hebrew (after
238#   they have been used to determine resolved direction), since the
239#   direction override information is carried by the code point itself.
240#
241#   Even when direction overrides are not needed for roundtrip
242#   fidelity, they are sometimes used when mapping Mac OS Hebrew
243#   characters to Unicode in order to achieve similar text layout with
244#   the resulting Unicode text. For example, the single Mac OS Hebrew
245#   ellipsis character has direction class right-left,and there is no
246#   left-right version. However, the Unicode HORIZONTAL ELLIPSIS
247#   character has direction class neutral (which means it may end up
248#   with a resolved direction of left-right if surrounded by left-right
249#   characters). When mapping the Mac OS Hebrew ellipsis to Unicode, it
250#   is surrounded with a direction override to help preserve proper
251#   text layout. The resolved direction is not needed or used when
252#   mapping the Unicode HORIZONTAL ELLIPSIS back to Mac OS Hebrew.
253#
254#   2. Use of corporate-zone Unicodes
255#
256#   The goals in the mappings provided here are:
257#   - Ensure roundtrip mapping from every character in the Mac OS
258#     Hebrew character set to Unicode and back
259#   - Use standard Unicode characters as much as possible, to
260#     maximize interchangeability of the resulting Unicode text.
261#     Whenever possible, avoid having content carried by private-use
262#     characters.
263#
264#   Some of the characters in the Mac OS Hebrew character set do not
265#   correspond to distinct, single Unicode characters. To map these
266#   and satisfy both goals above, we employ various strategies.
267#
268#   a) If possible, use private use characters in combination with
269#   standard Unicode characters to mark variants of the standard
270#   Unicode character.
271#
272#   Apple has defined a block of 32 corporate characters as "transcoding
273#   hints." These are used in combination with standard Unicode characters
274#   to force them to be treated in a special way for mapping to other
275#   encodings; they have no other effect. Sixteen of these transcoding
276#   hints are "grouping hints" - they indicate that the next 2-4 Unicode
277#   characters should be treated as a single entity for transcoding. The
278#   other sixteen transcoding hints are "variant tags" - they are like
279#   combining characters, and can follow a standard Unicode (or a sequence
280#   consisting of a base character and other combining characters) to
281#   cause it to be treated in a special way for transcoding. These always
282#   terminate a combining-character sequence.
283#
284#   Two transcoding hints are used in this mapping table: a grouping hint
285#   and a variant tag:
286#   hint:
287#     0xF86A  group next 2 characters, right-left directionality
288#     0xF87F  variant tag
289#
290#   In Mac OS Hebrew, 0xC0 is a ligature for lamed holam. This can also
291#   be represented in Mac OS Hebrew as 0xEC+0xDD, using separate
292#   characters for lamed and holam. The latter sequence is mapped to
293#   Unicode as 0x05DC+0x05B9, i.e. as the sequence HEBREW LETTER LAMED +
294#   HEBREW POINT HOLAM. We want to map the ligature 0xC0 using the same
295#   standard Unicode characters, but for round-trip fidelity we need to
296#   distinguish it from the mapping of the sequence 0xEC+0xDD. Thus for
297#   0xC0 we use a grouping hint, and map as follows:
298#
299#     0xC0 -> 0xF86A+0x05DC+0x05B9
300#
301#   The variant tag is used for "qamats qatan" to mark it as an alternate
302#   for HEBREW POINT QAMATS, as follows:
303#
304#     0xDE -> 0x05B8+0xF87F
305#
306#   b) Otherwise, use private use characters by themselves to map Mac OS
307#   Hebrew characters which  have no relationship to any standard Unicode
308#   character.
309#
310#   The following additional corporate zone Unicode characters are used
311#   for this purpose here (to map the obsolete "canorals", see above):
312#
313#     0xF89B  Hebrew canoral 1
314#     0xF89C  Hebrew canoral 2
315#     0xF89D  Hebrew canoral 3
316#     0xF89E  Hebrew canoral 4
317#
318#   3. Roundtrip considerations when mapping to decomposed Unicode
319#
320#   Both Mac OS Hebrew and Unicode provide multiple ways of representing
321#   certain letter-and-point combinations. For example, HEBREW LETTER
322#   VAV WITH HOLAM can be represented in Unicode as the single character
323#   0xFB4B or as the sequence 0x05D5 0x05B9; similarly, it can be
324#   represented in Mac OS Hebrew as 0xC7 or as the sequence 0xE5 0xDD.
325#   This leads to some roundtrip problems. First note that we have the
326#   following mappings without such problems:
327#
328#   Mac   standard                            decomp. of     reverse map
329#   OS    Unicode mapping                     std. mapping   of decomp.
330#   ----  ----------------------------------  -------------  -----------
331#   0xC6  0x05BC  ... POINT DAGESH OR MAPIQ   0x05BC (same)  0xC6
332#   0xE5  0x05D5  ... LETTER VAV              0x05D5 (same)  0xE5
333#   0xDD  0x05B9  ... POINT HOLAM             0x05B9 (same)  0xDD
334#
335#   However, those mappings above cause roundtrip problems for the
336#   the following mappings if they are decomposed:
337#
338#   Mac   standard                            decomp. of     reverse map
339#   OS    Unicode mapping                     std. mapping   of decomp.
340#   ----  ----------------------------------  -------------  -----------
341#   0xC7  0xFB4B  ... LETTER VAV WITH HOLAM   0x05D5 0x05B9  0xE5 0xDD
342#   0xC8  0xFB35  ... LETTER VAV WITH DAGESH  0x05D5 0x05BC  0xE5 0xC6
343#
344#   One solution is to use a grouping transcoding hint with the two
345#   decompositions above to mark the decomposed sequence for special
346#   treatment in transcoding. This yields the following mappings to
347#   decomposed Unicode:
348#
349#   Mac                                decomposed
350#   OS                                 Unicode mapping
351#   ----                               --------------------
352#   0xC7                               0xF86A 0x05D5 0x05B9
353#   0xC8                               0xF86A 0x05D5 0x05BC
354#
355# Details of mapping changes in each version:
356# -------------------------------------------
357#
358#   Changes from version b02 to version b03/c01:
359#
360#   - Stop specifying left-right context for digits 0x30-0x39, since the
361#     corresponding Unicodes 0x0030-0x0039 already have left-right
362#     directionality.
363#
364#   - Change mapping of 0x81 from 0xFB1F HEBREW LIGATURE YIDDISH YOD YOD
365#     PATAH to its canonical decomposition 0x05F2+0x05B7 to improve
366#     cross-platform compatibility (Windows doesn't handle 0xFB1F)
367#
368#   - Interchange the mappings of 0xA8 and 0xA9 to obtain the correct
369#     open/close behavior; they work differently than in Mac Arabic.
370#     The old mapping was
371#         0xA8 0x0028 # LEFT PARENTHESIS, right-left
372#         0xA9 0x0029 # RIGHT PARENTHESIS, right-left
373#     and the new mapping is
374#         0xA8 0x0029 # RIGHT PARENTHESIS, right-left
375#         0xA9 0x0028 # LEFT PARENTHESIS, right-left
376#
377#   Changes from version n01 to version n03:
378#
379#   - Change mapping for 0xC0 from single corporate character to
380#     grouping hint plus standard Unicodes
381#
382#   - Change mapping for 0xDE from single corporate character to
383#     standard Unicode plus variant tag
384#
385##################
386
3870x00 - 0x7F = 0x0000 -
3880x80 = 0x00C4	# LATIN CAPITAL LETTER A WITH DIAERESIS
3890x81 = 0xFB1F	# 0x05F2+0x05B7	# HEBREW LIGATURE YIDDISH YOD YOD PATAH
3900x82 = 0x00C7	# LATIN CAPITAL LETTER C WITH CEDILLA
3910x83 = 0x00C9	# LATIN CAPITAL LETTER E WITH ACUTE
3920x84 = 0x00D1	# LATIN CAPITAL LETTER N WITH TILDE
3930x85 = 0x00D6	# LATIN CAPITAL LETTER O WITH DIAERESIS
3940x86 = 0x00DC	# LATIN CAPITAL LETTER U WITH DIAERESIS
3950x87 = 0x00E1	# LATIN SMALL LETTER A WITH ACUTE
3960x88 = 0x00E0	# LATIN SMALL LETTER A WITH GRAVE
3970x89 = 0x00E2	# LATIN SMALL LETTER A WITH CIRCUMFLEX
3980x8A = 0x00E4	# LATIN SMALL LETTER A WITH DIAERESIS
3990x8B = 0x00E3	# LATIN SMALL LETTER A WITH TILDE
4000x8C = 0x00E5	# LATIN SMALL LETTER A WITH RING ABOVE
4010x8D = 0x00E7	# LATIN SMALL LETTER C WITH CEDILLA
4020x8E = 0x00E9	# LATIN SMALL LETTER E WITH ACUTE
4030x8F = 0x00E8	# LATIN SMALL LETTER E WITH GRAVE
4040x90 = 0x00EA	# LATIN SMALL LETTER E WITH CIRCUMFLEX
4050x91 = 0x00EB	# LATIN SMALL LETTER E WITH DIAERESIS
4060x92 = 0x00ED	# LATIN SMALL LETTER I WITH ACUTE
4070x93 = 0x00EC	# LATIN SMALL LETTER I WITH GRAVE
4080x94 = 0x00EE	# LATIN SMALL LETTER I WITH CIRCUMFLEX
4090x95 = 0x00EF	# LATIN SMALL LETTER I WITH DIAERESIS
4100x96 = 0x00F1	# LATIN SMALL LETTER N WITH TILDE
4110x97 = 0x00F3	# LATIN SMALL LETTER O WITH ACUTE
4120x98 = 0x00F2	# LATIN SMALL LETTER O WITH GRAVE
4130x99 = 0x00F4	# LATIN SMALL LETTER O WITH CIRCUMFLEX
4140x9A = 0x00F6	# LATIN SMALL LETTER O WITH DIAERESIS
4150x9B = 0x00F5	# LATIN SMALL LETTER O WITH TILDE
4160x9C = 0x00FA	# LATIN SMALL LETTER U WITH ACUTE
4170x9D = 0x00F9	# LATIN SMALL LETTER U WITH GRAVE
4180x9E = 0x00FB	# LATIN SMALL LETTER U WITH CIRCUMFLEX
4190x9F = 0x00FC	# LATIN SMALL LETTER U WITH DIAERESIS
4200xA0 = 0x0020	# SPACE, right-left
4210xA1 = 0x0021	# EXCLAMATION MARK, right-left
4220xA2 = 0x0022	# QUOTATION MARK, right-left
4230xA3 = 0x0023	# NUMBER SIGN, right-left
4240xA4 = 0x0024	# DOLLAR SIGN, right-left
4250xA5 = 0x0025	# PERCENT SIGN, right-left
4260xA6 = 0x20AA	# NEW SHEQEL SIGN
4270xA7 = 0x0027	# APOSTROPHE, right-left
4280xA8 = 0x0029	# RIGHT PARENTHESIS, right-left # close parenthesis
4290xA9 = 0x0028	# LEFT PARENTHESIS, right-left # open parenthesis
4300xAA = 0x002A	# ASTERISK, right-left
4310xAB = 0x002B	# PLUS SIGN, right-left
4320xAC = 0x002C	# COMMA, right-left
4330xAD = 0x002D	# HYPHEN-MINUS, right-left
4340xAE = 0x002E	# FULL STOP, right-left
4350xAF = 0x002F	# SOLIDUS, right-left
4360xB0 = 0x0030	# DIGIT ZERO, right-left (need override)
4370xB1 = 0x0031	# DIGIT ONE, right-left (need override)
4380xB2 = 0x0032	# DIGIT TWO, right-left (need override)
4390xB3 = 0x0033	# DIGIT THREE, right-left (need override)
4400xB4 = 0x0034	# DIGIT FOUR, right-left (need override)
4410xB5 = 0x0035	# DIGIT FIVE, right-left (need override)
4420xB6 = 0x0036	# DIGIT SIX, right-left (need override)
4430xB7 = 0x0037	# DIGIT SEVEN, right-left (need override)
4440xB8 = 0x0038	# DIGIT EIGHT, right-left (need override)
4450xB9 = 0x0039	# DIGIT NINE, right-left (need override)
4460xBA = 0x003A	# COLON, right-left
4470xBB = 0x003B	# SEMICOLON, right-left
4480xBC = 0x003C	# LESS-THAN SIGN, right-left
4490xBD = 0x003D	# EQUALS SIGN, right-left
4500xBE = 0x003E	# GREATER-THAN SIGN, right-left
4510xBF = 0x003F	# QUESTION MARK, right-left
4520xC0 = 0x05B9	# 0xF86A+0x05DC+0x05B9	# Hebrew ligature lamed holam
4530xC1 = 0x201E	# DOUBLE LOW-9 QUOTATION MARK, right-left
4540xC2 = 0xF89B	# Hebrew canoral 1
4550xC3 = 0xF89C	# Hebrew canoral 2
4560xC4 = 0xF89D	# Hebrew canoral 3
4570xC5 = 0xF89E	# Hebrew canoral 4
4580xC6 = 0x05BC	# HEBREW POINT DAGESH OR MAPIQ
4590xC7 = 0xFB4B	# HEBREW LETTER VAV WITH HOLAM
4600xC8 = 0xFB35	# HEBREW LETTER VAV WITH DAGESH
4610xC9 = 0x2026	# HORIZONTAL ELLIPSIS, right-left
4620xCA = 0x00A0	# NO-BREAK SPACE, right-left
4630xCB = 0x05B8	# HEBREW POINT QAMATS
4640xCC = 0x05B7	# HEBREW POINT PATAH
4650xCD = 0x05B5	# HEBREW POINT TSERE
4660xCE = 0x05B6	# HEBREW POINT SEGOL
4670xCF = 0x05B4	# HEBREW POINT HIRIQ
4680xD0 = 0x2013	# EN DASH, right-left
4690xD1 = 0x2014	# EM DASH, right-left
4700xD2 = 0x201C	# LEFT DOUBLE QUOTATION MARK, right-left
4710xD3 = 0x201D	# RIGHT DOUBLE QUOTATION MARK, right-left
4720xD4 = 0x2018	# LEFT SINGLE QUOTATION MARK, right-left
4730xD5 = 0x2019	# RIGHT SINGLE QUOTATION MARK, right-left
4740xD6 = 0xFB2A	# HEBREW LETTER SHIN WITH SHIN DOT
4750xD7 = 0xFB2B	# HEBREW LETTER SHIN WITH SIN DOT
4760xD8 = 0x05BF	# HEBREW POINT RAFE
4770xD9 = 0x05B0	# HEBREW POINT SHEVA
4780xDA = 0x05B2	# HEBREW POINT HATAF PATAH
4790xDB = 0x05B1	# HEBREW POINT HATAF SEGOL
4800xDC = 0x05BB	# HEBREW POINT QUBUTS
4810xDD = 0x05B9	# HEBREW POINT HOLAM
4820xDE = 0xF87F	# 0x05B8+0xF87F	# HEBREW POINT QAMATS, alternate form "qamats qatan"
4830xDF = 0x05B3	# HEBREW POINT HATAF QAMATS
4840xE0 = 0x05D0	# HEBREW LETTER ALEF
4850xE1 = 0x05D1	# HEBREW LETTER BET
4860xE2 = 0x05D2	# HEBREW LETTER GIMEL
4870xE3 = 0x05D3	# HEBREW LETTER DALET
4880xE4 = 0x05D4	# HEBREW LETTER HE
4890xE5 = 0x05D5	# HEBREW LETTER VAV
4900xE6 = 0x05D6	# HEBREW LETTER ZAYIN
4910xE7 = 0x05D7	# HEBREW LETTER HET
4920xE8 = 0x05D8	# HEBREW LETTER TET
4930xE9 = 0x05D9	# HEBREW LETTER YOD
4940xEA = 0x05DA	# HEBREW LETTER FINAL KAF
4950xEB = 0x05DB	# HEBREW LETTER KAF
4960xEC = 0x05DC	# HEBREW LETTER LAMED
4970xED = 0x05DD	# HEBREW LETTER FINAL MEM
4980xEE = 0x05DE	# HEBREW LETTER MEM
4990xEF = 0x05DF	# HEBREW LETTER FINAL NUN
5000xF0 = 0x05E0	# HEBREW LETTER NUN
5010xF1 = 0x05E1	# HEBREW LETTER SAMEKH
5020xF2 = 0x05E2	# HEBREW LETTER AYIN
5030xF3 = 0x05E3	# HEBREW LETTER FINAL PE
5040xF4 = 0x05E4	# HEBREW LETTER PE
5050xF5 = 0x05E5	# HEBREW LETTER FINAL TSADI
5060xF6 = 0x05E6	# HEBREW LETTER TSADI
5070xF7 = 0x05E7	# HEBREW LETTER QOF
5080xF8 = 0x05E8	# HEBREW LETTER RESH
5090xF9 = 0x05E9	# HEBREW LETTER SHIN
5100xFA = 0x05EA	# HEBREW LETTER TAV
5110xFB = 0x007D	# RIGHT CURLY BRACKET, right-left
5120xFC = 0x005D	# RIGHT SQUARE BRACKET, right-left
5130xFD = 0x007B	# LEFT CURLY BRACKET, right-left
5140xFE = 0x005B	# LEFT SQUARE BRACKET, right-left
5150xFF = 0x007C	# VERTICAL LINE, right-left
516END_MAP
517