lib/LaTeXML/Util/Unicode.pm untested

filter: Statementstmt 0.0% 0 / 57 Branchbran 0.0% 0 / 16 Conditioncond n/a MC/DCmcdc n/a Subroutinesub 0.0% 0 / 7 Podpod 0.0% 0 / 4 totaltotal 0.0% 0 / 84 CC 9 SCAR 26.1
CC
9
Cov
0%
makePlane1Map
20.0
unicode_convert
20.0
unicode_mathvariant
12.0
CRAP 13.6
1 # /=====================================================================\ #
2 # | LaTeXML::Util::Unicode | #
3 # | Unicode Utilities for LaTeXML | #
4 # |=====================================================================| #
5 # | Part of LaTeXML: | #
6 # | Public domain software, produced as part of work done by the | #
7 # | United States Government & not subject to copyright in the US. | #
8 # |---------------------------------------------------------------------| #
9 # | Bruce Miller <bruce.miller@nist.gov> #_# | #
10 # | http://dlmf.nist.gov/LaTeXML/ (o o) | #
11 # \=========================================================ooo==U==ooo=/ #
12 package LaTeXML::Util::Unicode;
13 use strict;
14 use warnings;
15 use base qw(Exporter);
16 our @EXPORT = qw( &UTF &unicode_mathvariant &unicode_convert);
17  
18 #======================================================================
19 # Unicode manipulation utilities useful for LaTeXML
20 # Mostly, but not exclusively, about Mathematics
21 #======================================================================
22  
23 sub UTF {
24 my ($code) = @_;
25 return pack('U', $code); }
26  
27 #======================================================================
28 # Unicode Math Codepoints
29 # The basic latin and greek alphabets, as well as numbers appear in several
30 # (virtual) blocks.
31 # normal
32 # bold, italic, bold-italic
33 # sans-serif, bold-sans-serif, sans-serif-italic
34 # monospace
35 # script, bold-script
36 # fraktur, bold-fraktur
37 # double-struck
38  
39 # Create mapping sequences of
40 # latin (uppercase & lowercase), greek (uppercase & lowercase) and digits
41 # to blocks at the given positions within
42 sub makePlane1Map {
43 my ($latin, $GREEK, $greek, $digits) = @_;
44 return (
45 (map { (UTF(ord('A') + $_) => UTF($latin + $_)) } 0 .. 25),
46 (map { (UTF(ord('a') + $_) => UTF($latin + 26 + $_)) } 0 .. 25),
47 ($GREEK ? (map { (UTF(0x0391 + $_) => UTF($GREEK + $_)) } 0 .. 24) : ()),
48 ($greek ? (map { (UTF(0x03B1 + $_) => UTF($greek + $_)) } 0 .. 24) : ()),
49 ($digits ? (map { (UTF(ord('0') + $_) => UTF($digits + $_)) } 0 .. 9) : ())); }
50  
51 my %unicode_map = ( # CONSTANT
52 'bold' => { makePlane1Map(0x1D400, 0x1D6A8, 0x1D6C2, 0x1D7CE) },
53 'italic' => { makePlane1Map(0x1D434, 0x1D6E2, 0x1D6FC, undef),
54 h => "\x{210E}" },
55 'bold-italic' => { makePlane1Map(0x1D468, 0x1D71C, 0x1D736, undef) },
56 'sans-serif' => { makePlane1Map(0x1D5A0, undef, undef, 0x1D7E2) },
57 'bold-sans-serif' => { makePlane1Map(0x1D5D4, 0x1D756, 0x1D770, 0x1D7EC) },
58 'sans-serif-italic' => { makePlane1Map(0x1D608, undef, undef, undef) },
59 'sans-serif-bold-italic' => { makePlane1Map(0x1D63C, 0x1D790, 0x1D7AA, undef) },
60 'monospace' => { makePlane1Map(0x1D670, undef, undef, 0x1D7F6) },
61 'script' => { makePlane1Map(0x1D49C, undef, undef, undef),
62 B => "\x{212C}", E => "\x{2130}", F => "\x{2131}", H => "\x{210B}", I => "\x{2110}",
63 L => "\x{2112}", M => "\x{2133}", R => "\x{211B}",
64 e => "\x{212F}", g => "\x{210A}", o => "\x{2134}" },
65 'bold-script' => { makePlane1Map(0x1D4D0, undef, undef, undef) },
66 'fraktur' => { makePlane1Map(0x1D504, undef, undef, undef),
67 C => "\x{212D}", H => "\x{210C}", I => "\x{2111}", R => "\x{211C}", Z => "\x{2128}" },
68 'bold-fraktur' => { makePlane1Map(0x1D56C, undef, undef, undef) },
69 'double-struck' => { makePlane1Map(0x1D538, undef, undef, 0x1D7D8),
70 C => "\x{2102}", H => "\x{210D}", N => "\x{2115}", P => "\x{2119}", Q => "\x{211A}",
71 R => "\x{211D}", Z => "\x{2124}" },
72 superscript => {
73 "\x{2032}" => "'", # \prime
74 0 => "\x{2070}",
75 1 => UTF(0xB9),
76 2 => UTF(0xB2),
77 3 => UTF(0xB3),
78 4 => "\x{2074}",
79 5 => "\x{2075}",
80 6 => "\x{2076}",
81 7 => "\x{2077}",
82 8 => "\x{2078}",
83 9 => "\x{2079}",
84 '+' => "\x{207A}",
85 '-' => "\x{207B}",
86 '=' => "\x{207C}",
87 '(' => "\x{207D}",
88 ')' => "\x{207E}",
89 'n' => "\x{207F}",
90 'i' => "\x{2071}",
91 'V' => "\x{2C7D}",
92 'h' => "\x{02B0}", # aspirated!?
93 'j' => "\x{02B2}",
94 'r' => "\x{02B3}",
95 'W' => "\x{02B7}",
96 'y' => "\x{02B8}",
97 's' => "\x{02E2}",
98 'x' => "\x{02E3}",
99 'A' => "\x{1D2C}",
100 UTF(0xC6) => "\x{1D2D}",
101 'B' => "\x{1D2E}",
102 'D' => "\x{1D30}",
103 'E' => "\x{1D31}",
104 'G' => "\x{1D33}",
105 'H' => "\x{1D34}",
106 'I' => "\x{1D35}",
107 'J' => "\x{1D36}",
108 'K' => "\x{1D37}",
109 'L' => "\x{1D38}",
110 'M' => "\x{1D39}",
111 'N' => "\x{1D3A}",
112 'O' => "\x{1D3C}",
113 'P' => "\x{1D3E}",
114 'R' => "\x{1D3F}",
115 'T' => "\x{1D40}",
116 'U' => "\x{1D41}",
117 'W' => "\x{1D42}",
118 'a' => "\x{1D43}",
119 "\x{03B1}" => "\x{1D45}", # \alpha
120 UTF(0xE6) => "\x{1D46}",
121 'b' => "\x{1D47}",
122 'd' => "\x{1D48}",
123 'e' => "\x{1D49}",
124 "\x{03B5}" => "\x{1D4B}", # \varepsilon
125 "\x{03F5}" => "\x{1D4B}", # \epsilon ??? Close enough?
126 'g' => "\x{1D4D}",
127 '!' => "\x{1D4E}",
128 'k' => "\x{1D4F}",
129 'm' => "\x{1D50}",
130 'o' => "\x{1D52}",
131 'p' => "\x{1D56}",
132 't' => "\x{1D57}",
133 'u' => "\x{1D58}",
134 'v' => "\x{1D5B}",
135 "\x{03B2}" => "\x{1D5D}", # \beta
136 "\x{03B3}" => "\x{1D5E}", # \gamma
137 "\x{03B4}" => "\x{1D5F}", # \delta
138 "\x{03C6}" => "\x{1D60}", # \varphi
139 "\x{03D5}" => "\x{1D60}", # \phi; close enough?
140 "\x{03BE}" => "\x{1D61}", # \xi
141 'H' => "\x{1D78}",
142 'c' => "\x{1D9C}",
143 'f' => "\x{1DA0}",
144 'g' => "\x{1DA2}",
145 "\x{03A6}" => "\x{1DB2}", # \Phi?
146 "\x{03C5}" => "\x{1DB7}", # \upsilon
147 'z' => "\x{1DBB}",
148 "\x{03B8}" => "\x{1DBF}", # \theta
149 },
150 subscript => {
151 0 => "\x{2080}",
152 1 => "\x{2081}",
153 2 => "\x{2082}",
154 3 => "\x{2083}",
155 4 => "\x{2084}",
156 5 => "\x{2085}",
157 6 => "\x{2086}",
158 7 => "\x{2087}",
159 8 => "\x{2088}",
160 9 => "\x{2089}",
161 '+' => "\x{208A}",
162 '-' => "\x{208B}",
163 '=' => "\x{208C}",
164 '(' => "\x{208D}",
165 ')' => "\x{208E}",
166 'a' => "\x{2090}",
167 'e' => "\x{2091}",
168 'o' => "\x{2092}",
169 'x' => "\x{2093}",
170 'upsidedowne' => "\x{2094}",
171 'h' => "\x{2095}",
172 'k' => "\x{2096}",
173 'l' => "\x{2097}",
174 'm' => "\x{2098}",
175 'n' => "\x{2099}",
176 'p' => "\x{209A}",
177 's' => "\x{209B}",
178 't' => "\x{209C}",
179 'j' => "\x{2C7C}",
180 'i' => "\x{1D62}",
181 'r' => "\x{1D63}",
182 'u' => "\x{1D64}",
183 'v' => "\x{1D65}",
184 "\x{03B2}" => "\x{1D66}", # \beta
185 "\x{03B3}" => "\x{1D67}", # \gamma
186 "\x{03C1}" => "\x{1D68}", # \rho
187 "\x{03C6}" => "\x{1D69}", # \varphi
188 "\x{03D5}" => "\x{1D69}", # \phi; close enough?
189 "\x{03BE}" => "\x{1D6A}", # \xi
190 },
191 );
192  
193 # Return the re-mapping of the codepoints in $string to the codepoints corresponding
194 # to the given $style.
195 # Returns undef if not all chars have a known mapping.
196 sub unicode_convert {
197 my ($string, $style) = @_;
198 if (my $mapping = $unicode_map{$style}) {
199 my @c = map { $$mapping{$_} } split(//, (defined $string ? $string : ''));
200 if (!grep { !defined $_ } @c) { # Only if ALL chars in the token could be mapped... ?????
201 return join('', @c); } }
202 return; }
203  
204 #======================================================================
205 # Normalizing a Math Font for lookup in Unicode.
206 # Basically, this results in MathML's mathvariant
207 # which corresponds to the major blocks of "styled" characters in Unicode:
208 # Mappings between (normalized) internal fonts & sizes.
209 # Default math font is roman|medium|upright.
210 my %mathvariants = ( # CONSTANT
211 'upright' => 'normal',
212 'serif' => 'normal',
213 'medium' => 'normal',
214 'bold' => 'bold',
215 'italic' => 'italic',
216 'medium italic' => 'italic',
217 'bold italic' => 'bold-italic',
218 'doublestruck' => 'double-struck',
219 'blackboard' => 'double-struck',
220 'blackboard bold' => 'double-struck', # all collapse
221 'blackboard upright' => 'double-struck', # all collapse
222 'blackboard bold upright' => 'double-struck', # all collapse
223 'fraktur' => 'fraktur',
224 'fraktur italic' => 'fraktur', # all collapse
225 'fraktur bold' => 'bold-fraktur',
226 'script' => 'script',
227 'script italic' => 'script', # all collapse
228 'script bold' => 'bold-script',
229 'caligraphic' => 'script', # NOTE: TeX caligraphic is NOT script!
230 'caligraphic bold' => 'bold-script', # collapse
231 'sansserif' => 'sans-serif',
232 'sansserif bold' => 'bold-sans-serif',
233 'sansserif italic' => 'sans-serif-italic',
234 'sansserif bold italic' => 'sans-serif-bold-italic',
235 'typewriter' => 'monospace',
236 'typewriter bold' => 'monospace',
237 'typewriter italic' => 'monospace',
238 'typewriter bold italic' => 'monospace',
239 );
240  
241 # The font differences (from the containing context) have been deciphered
242 # into font, size and color attributes. The font should match
243 # one of the above... (?)
244  
245 # Given a font string (joining the components)
246 # reduce it to a "sane" font. Note that MathML uses a single mathvariant
247 # to name the font, and doesn't inherit font components like italic or bold.
248 # Thus the font should be "complete", but also we can ignore components with
249 # default values like medium or upright (unless that is the only component).
250 sub unicode_mathvariant {
251 my ($font) = @_;
252 $font =~ s/slanted/italic/; # equivalent in math
253 $font =~ s/(?<!\w)serif// unless $font eq 'serif'; # Not needed (unless alone)
254 $font =~ s/(?<!^)upright//; # Not needed (unless 1st element)
255 $font =~ s/(?<!^)medium//; # Not needed (unless 1st element)
256 $font =~ s/^\s+//; $font =~ s/\s+$//;
257 my $variant;
258 return $variant if $variant = $mathvariants{$font};
259 # $font =~ s/\sitalic//; # try w/o italic ?
260 # return $variant if $variant = $mathvariants{$font};
261 # $font =~ s/\sbold//; # try w/o bold ?
262 # return $variant if $variant = $mathvariants{$font};
263 return 'normal'; }
264  
265 #======================================================================
266 1;
267  
268 __END__