11 #=======================================================================
12 # File name: FARSI.TXT
14 # Contents: Map (external version) from Mac OS Farsi
15 # character set to Unicode 2.1 and later.
17 # Copyright: (c) 1997-2002, 2005 by Apple Computer, Inc., all rights
20 # Contact: charsets@apple.com
24 # c02 2005-Apr-05 Update header comments. Matches internal xml
25 # <c1.1> and Text Encoding Converter 2.0.
26 # b3,c1 2002-Dec-19 Add comments about character display and
27 # direction overrides. Update URLs, notes.
28 # Matches internal utom<b3>.
29 # b02 1999-Sep-22 Update contact e-mail address. Matches
30 # internal utom<b1>, ufrm<b1>, and Text
31 # Encoding Converter version 1.5.
32 # n04 1998-Feb-05 Show required Unicode character
33 # directionality in a different way. Matches
34 # internal utom<n3>, ufrm<n9>, and Text
35 # Encoding Converter version 1.3. Update
36 # header comments; include information on
37 # loose mapping of digits, and changes to
38 # mapping for the TrueType variant.
39 # n01 1997-Jul-17 First version. Matches internal utom<n1>,
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
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.
59 # These mapping tables and character lists are subject to change.
60 # The latest tables should be available from the following:
62 # <http://www.unicode.org/Public/MAPPINGS/VENDORS/APPLE/>
64 # For general information about Mac OS encodings and these mapping
65 # tables, see the file "README.TXT".
70 # Three tab-separated columns;
71 # '#' begins a comment which continues to the end of the line.
72 # Column #1 is the Mac OS Farsi code (in hex as 0xNN)
73 # Column #2 is the corresponding Unicode (in hex as 0xNNNN),
74 # possibly preceded by a tag indicating required directionality
75 # (i.e. <LR>+0xNNNN or <RL>+0xNNNN).
76 # Column #3 is a comment containing the Unicode name.
78 # The entries are in Mac OS Farsi code order.
80 # Control character mappings are not shown in this table, following
81 # the conventions of the standard UTC mapping tables. However, the
82 # Mac OS Farsi character set uses the standard control characters at
85 # Notes on Mac OS Farsi:
86 # ----------------------
88 # This is a legacy Mac OS encoding; in the Mac OS X Carbon and Cocoa
89 # environments, it is only supported via transcoding to and from
94 # The Mac OS Farsi character set is based on the Mac OS Arabic
95 # character set. The main difference is in the right-to-left digits
96 # 0xB0-0xB9: For Mac OS Arabic these correspond to right-left
97 # versions of the Unicode ARABIC-INDIC DIGITs 0660-0669; for
98 # Mac OS Farsi these correspond to right-left versions of the
99 # Unicode EXTENDED ARABIC-INDIC DIGITs 06F0-06F9. The other
100 # difference is in the nature of the font variants.
102 # For more information, see the comments in the mapping table for
105 # Mac OS Farsi characters 0xEB-0xF2 are non-spacing/combining marks.
107 # 2. Directional characters and roundtrip fidelity
109 # The Mac OS Arabic character set (on which Mac OS Farsi is based)
110 # was developed in 1986-1987. At that time the bidirectional line
111 # layout algorithm used in the Mac OS Arabic system was fairly simple;
112 # it used only a few direction classes (instead of the 19 now used in
113 # the Unicode bidirectional algorithm). In order to permit users to
114 # handle some tricky layout problems, certain punctuation and symbol
115 # characters were encoded twice, one with a left-right direction
116 # attribute and the other with a right-left direction attribute. This
117 # is the case in Mac OS Farsi too.
119 # For example, plus sign is encoded at 0x2B with a left-right
120 # attribute, and at 0xAB with a right-left attribute. However, there
121 # is only one PLUS SIGN character in Unicode. This leads to some
122 # interesting problems when mapping between Mac OS Farsi and Unicode;
125 # A related problem is that even when a particular character is
126 # encoded only once in Mac OS Farsi, it may have a different
127 # direction attribute than the corresponding Unicode character.
129 # For example, the Mac OS Farsi character at 0x93 is HORIZONTAL
130 # ELLIPSIS with strong right-left direction. However, the Unicode
131 # character HORIZONTAL ELLIPSIS has direction class neutral.
133 # 3. Behavior of ASCII-range numbers in WorldScript
135 # Mac OS Farsi also has two sets of digit codes.
137 # The digits at 0x30-0x39 may be displayed using either European
138 # digit forms or Persian digit forms, depending on context. If there
139 # is a "strong European" character such as a Latin letter on either
140 # side of a sequence consisting of digits 0x30-0x39 and possibly comma
141 # 0x2C or period 0x2E, then the characters will be displayed using
142 # European forms (This will happen even if there are neutral characters
143 # between the digits and the strong European character). Otherwise, the
144 # digits will be displayed using Persian forms, the comma will be
145 # displayed as Arabic thousands separator, and the period as Arabic
146 # decimal separator. In any case, 0x2C, 0x2E, and 0x30-0x39 are always
149 # The digits at 0xB0-0xB9 are always displayed using Persian digit
150 # shapes, and moreover, these digits always have strong right-left
151 # directionality. These are mainly intended for special layout
152 # purposes such as part numbers, etc.
156 # The table in this file gives the Unicode mappings for the standard
157 # Mac OS Farsi encoding. This encoding is supported by the Tehran font
158 # (the system font for Farsi), and is the encoding supported by the
159 # text processing utilities. However, the other Farsi fonts actually
160 # implement a somewhat different encoding; this affects nine code
161 # points including 0xAA and 0xC0 (which are also affected by font
162 # variants in Mac OS Arabic). For these nine code points the standard
163 # Mac OS Farsi encoding has the following mappings:
164 # 0x8B -> 0x06BA ARABIC LETTER NOON GHUNNA (Urdu)
165 # 0xA4 -> <RL>+0x0024 DOLLAR SIGN, right-left
166 # 0xAA -> <RL>+0x002A ASTERISK, right-left
167 # 0xC0 -> <RL>+0x274A EIGHT TEARDROP-SPOKED PROPELLER ASTERISK,
169 # 0xF4 -> 0x0679 ARABIC LETTER TTEH (Urdu)
170 # 0xF7 -> 0x06A4 ARABIC LETTER VEH (for transliteration)
171 # 0xF9 -> 0x0688 ARABIC LETTER DDAL (Urdu)
172 # 0xFA -> 0x0691 ARABIC LETTER RREH (Urdu)
173 # 0xFF -> 0x06D2 ARABIC LETTER YEH BARREE (Urdu)
175 # The TrueType variant is used for the Farsi TrueType fonts: Ashfahan,
176 # Amir, Kamran, Mashad, NadeemFarsi. It differs from the standard
177 # variant in the following ways:
178 # 0x8B -> 0xF882 Arabic ligature "peace on him" (corporate char.)
179 # 0xA4 -> 0xFDFC RIAL SIGN (added in Unicode 3.2)
180 # 0xAA -> <RL>+0x00D7 MULTIPLICATION SIGN, right-left
181 # 0xC0 -> <RL>+0x002A ASTERISK, right-left
182 # 0xF4 -> <RL>+0x00B0 DEGREE SIGN, right-left
183 # 0xF7 -> 0xFDFA ARABIC LIGATURE SALLALLAHOU ALAYHE WASALLAM
184 # 0xF9 -> <RL>+0x25CF BLACK CIRCLE, right-left
185 # 0xFA -> <RL>+0x25A0 BLACK SQUARE, right-left
186 # 0xFF -> <RL>+0x25B2 BLACK UP-POINTING TRIANGLE, right-left
188 # Unicode mapping issues and notes:
189 # ---------------------------------
191 # 1. Matching the direction of Mac OS Farsi characters
193 # When Mac OS Farsi encodes a character twice but with different
194 # direction attributes for the two code points - as in the case of
195 # plus sign mentioned above - we need a way to map both Mac OS Farsi
196 # code points to Unicode and back again without loss of information.
197 # With the plus sign, for example, mapping one of the Mac OS Farsi
198 # characters to a code in the Unicode corporate use zone is
199 # undesirable, since both of the plus sign characters are likely to
200 # be used in text that is interchanged.
202 # The problem is solved with the use of direction override characters
203 # and direction-dependent mappings. When mapping from Mac OS Farsi
204 # to Unicode, we use direction overrides as necessary to force the
205 # direction of the resulting Unicode characters.
207 # The required direction is indicated by a direction tag in the
208 # mappings. A tag of <LR> means the corresponding Unicode character
209 # must have a strong left-right context, and a tag of <RL> indicates
210 # a right-left context.
212 # For example, the mapping of 0x2B is given as <LR>+0x002B; the
213 # mapping of 0xAB is given as <RL>+0x002B. If we map an isolated
214 # instance of 0x2B to Unicode, it should be mapped as follows (LRO
215 # indicates LEFT-RIGHT OVERRIDE, PDF indicates POP DIRECTION
218 # 0x2B -> 0x202D (LRO) + 0x002B (PLUS SIGN) + 0x202C (PDF)
220 # When mapping several characters in a row that require direction
221 # forcing, the overrides need only be used at the beginning and end.
224 # 0x24 0x20 0x28 0x29 -> 0x202D 0x0024 0x0020 0x0028 0x0029 0x202C
226 # If neutral characters that require direction forcing are already
227 # between strong-direction characters with matching directionality,
228 # then direction overrides need not be used. Direction overrides are
229 # always needed to map the right-left digits at 0xB0-0xB9.
231 # When mapping from Unicode to Mac OS Farsi, the Unicode
232 # bidirectional algorithm should be used to determine resolved
233 # direction of the Unicode characters. The mapping from Unicode to
234 # Mac OS Farsi can then be disambiguated by the use of the resolved
237 # Unicode 0x002B -> Mac OS Farsi 0x2B (if L) or 0xAB (if R)
239 # However, this also means the direction override characters should
240 # be discarded when mapping from Unicode to Mac OS Farsi (after
241 # they have been used to determine resolved direction), since the
242 # direction override information is carried by the code point itself.
244 # Even when direction overrides are not needed for roundtrip
245 # fidelity, they are sometimes used when mapping Mac OS Farsi
246 # characters to Unicode in order to achieve similar text layout with
247 # the resulting Unicode text. For example, the single Mac OS Farsi
248 # ellipsis character has direction class right-left,and there is no
249 # left-right version. However, the Unicode HORIZONTAL ELLIPSIS
250 # character has direction class neutral (which means it may end up
251 # with a resolved direction of left-right if surrounded by left-right
252 # characters). When mapping the Mac OS Farsi ellipsis to Unicode, it
253 # is surrounded with a direction override to help preserve proper
254 # text layout. The resolved direction is not needed or used when
255 # mapping the Unicode HORIZONTAL ELLIPSIS back to Mac OS Farsi.
257 # 2. Mapping the Mac OS Farsi digits
259 # The main table below contains mappings that should be used when
260 # strict round-trip fidelity is required. However, for numeric
261 # values, the mappings in that table will produce Unicode characters
262 # that may appear different than the Mac OS Farsi text displayed on
263 # a Mac OS system using WorldScript. This is because WorldScript
264 # uses context-dependent display for the 0x30-0x39 digits.
266 # If roundtrip fidelity is not required, then the following
267 # alternate mappings should be used when a sequence of 0x30-0x39
268 # digits - possibly including 0x2C and 0x2E - occurs in an Arabic
269 # context (that is, when the first "strong" character on either side
270 # of the digit sequence is Arabic, or there is no strong character):
272 # 0x2C 0x066C # ARABIC THOUSANDS SEPARATOR
273 # 0x2E 0x066B # ARABIC DECIMAL SEPARATOR
274 # 0x30 0x06F0 # EXTENDED ARABIC-INDIC DIGIT ZERO
275 # 0x31 0x06F1 # EXTENDED ARABIC-INDIC DIGIT ONE
276 # 0x32 0x06F2 # EXTENDED ARABIC-INDIC DIGIT TWO
277 # 0x33 0x06F3 # EXTENDED ARABIC-INDIC DIGIT THREE
278 # 0x34 0x06F4 # EXTENDED ARABIC-INDIC DIGIT FOUR
279 # 0x35 0x06F5 # EXTENDED ARABIC-INDIC DIGIT FIVE
280 # 0x36 0x06F6 # EXTENDED ARABIC-INDIC DIGIT SIX
281 # 0x37 0x06F7 # EXTENDED ARABIC-INDIC DIGIT SEVEN
282 # 0x38 0x06F8 # EXTENDED ARABIC-INDIC DIGIT EIGHT
283 # 0x39 0x06F9 # EXTENDED ARABIC-INDIC DIGIT NINE
285 # 3. Use of corporate-zone Unicodes (mapping the TrueType variant)
287 # The following corporate zone Unicode character is used in this
290 # 0xF882 Arabic ligature "peace on him"
292 # Details of mapping changes in each version:
293 # -------------------------------------------
295 # Changes from version b02 to version b03/c01:
297 # - Update mapping of 0xA4 in TrueType variant to use new Unicode
298 # character U+FDFC RIAL SIGN addded for Unicode 3.2
300 # Changes from version n01 to version n04:
302 # - Change mapping of 0xA4 in TrueType variant (just described in
303 # header comment) from single corporate character to use
308 0x0000 - 0x007F = 0x00 -