1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
|
#!/usr/local/bin/clisp -C
;;; Creation of CLISP's uni_names.h from the UnicodeData.txt table.
;;; Bruno Haible 2000-12-28
(defparameter add-comments nil)
(defstruct unicode-char
(code nil :type integer)
(name nil :type string)
word-indices
word-indices-index
)
(defstruct word-list
(hashed nil :type hash-table)
(sorted nil :type list)
size ; number of characters total
length ; number of words
)
(defun main (inputfile outputfile)
(declare (type string inputfile outputfile))
#+UNICODE (setq *default-file-encoding* charset:utf-8)
(let ((all-chars '()))
;; Read all characters and names from the input file.
(with-open-file (istream inputfile :direction :input)
(loop
(let ((line (read-line istream nil nil)))
(unless line (return))
(let* ((i1 (position #\; line))
(i2 (position #\; line :start (1+ i1)))
(code-string (subseq line 0 i1))
(code (parse-integer code-string :radix 16))
(name-string (subseq line (1+ i1) i2)))
; Ignore characters whose name starts with "<".
(unless (eql (char name-string 0) #\<)
; Also ignore Hangul syllables; they are treated specially.
(unless (<= #xAC00 code #xD7A3)
(push (make-unicode-char :code code :name name-string)
all-chars
) ) )
) ) ) )
(setq all-chars (nreverse all-chars))
;; Split into words.
(let ((words-by-length (make-array 0 :adjustable t)))
(dolist (name (cons "HANGUL SYLLABLE" (mapcar #'unicode-char-name all-chars)))
(let ((i1 0))
(loop
(when (>= i1 (length name)) (return))
(let ((i2 (or (position #\Space name :start i1) (length name))))
(let* ((word (subseq name i1 i2))
(len (length word)))
(when (>= len (length words-by-length))
(adjust-array words-by-length (1+ len))
)
(unless (aref words-by-length len)
(setf (aref words-by-length len)
(make-word-list
:hashed (make-hash-table :test #'equal)
:sorted '()
) ) )
(let ((word-list (aref words-by-length len)))
(unless (gethash word (word-list-hashed word-list))
(setf (gethash word (word-list-hashed word-list)) t)
(push word (word-list-sorted word-list))
) )
)
(setq i1 (1+ i2))
) ) ) )
;; Sort the word lists.
(dotimes (len (length words-by-length))
(unless (aref words-by-length len)
(setf (aref words-by-length len)
(make-word-list
:hashed (make-hash-table :test #'equal)
:sorted '()
) ) )
(let ((word-list (aref words-by-length len)))
(setf (word-list-sorted word-list)
(sort (word-list-sorted word-list) #'string<)
)
(setf (word-list-size word-list)
(reduce #'+ (mapcar #'length (word-list-sorted word-list)))
)
(setf (word-list-length word-list)
(length (word-list-sorted word-list))
) ) )
;; Output the tables.
(with-open-file (ostream outputfile :direction :output
#+UNICODE :external-format #+UNICODE charset:ascii)
(format ostream "/*~%")
(format ostream " * ~A~%" (file-namestring outputfile))
(format ostream " *~%")
(format ostream " * Unicode character name table.~%")
(format ostream " * Generated automatically by the gen-uninames utility.~%")
(format ostream " */~%")
(format ostream "~%")
(format ostream "static const char unicode_name_words[~D] = {~%"
(let ((sum 0))
(dotimes (len (length words-by-length))
(let ((word-list (aref words-by-length len)))
(incf sum (word-list-size word-list))
) )
sum
) )
(dotimes (len (length words-by-length))
(let ((word-list (aref words-by-length len)))
(dolist (word (word-list-sorted word-list))
(format ostream " ~{ '~C',~}~%" (coerce word 'list))
) ) )
(format ostream "};~%")
(format ostream "#define UNICODE_CHARNAME_NUM_WORDS ~D~%"
(let ((sum 0))
(dotimes (len (length words-by-length))
(let ((word-list (aref words-by-length len)))
(incf sum (word-list-length word-list))
) )
sum
) )
#| ; Redundant data
(format ostream "static const uint16 unicode_name_word_offsets[~D] = {~%"
(let ((sum 0))
(dotimes (len (length words-by-length))
(let ((word-list (aref words-by-length len)))
(incf sum (word-list-length word-list))
) )
sum
) )
(dotimes (len (length words-by-length))
(let ((word-list (aref words-by-length len)))
(when (word-list-sorted word-list)
(format ostream " ")
(do ((l (word-list-sorted word-list) (cdr l))
(offset 0 (+ offset (length (car l)))))
((endp l))
(format ostream "~<~% ~0,79:; ~D,~>" offset)
)
(format ostream "~%")
) ) )
(format ostream "};~%")
|#
(format ostream "static const struct { uintL extra_offset; uint16 ind_offset; } unicode_name_by_length[~D] = {~%"
(1+ (length words-by-length))
)
(let ((extra-offset 0)
(ind-offset 0))
(dotimes (len (length words-by-length))
(let ((word-list (aref words-by-length len)))
(format ostream " { ~D, ~D },~%" extra-offset ind-offset)
(incf extra-offset (word-list-size word-list))
(incf ind-offset (word-list-length word-list))
) )
(format ostream " { ~D, ~D }~%" extra-offset ind-offset)
)
(format ostream "};~%")
(let ((ind-offset 0))
(dotimes (len (length words-by-length))
(let ((word-list (aref words-by-length len)))
(dolist (word (word-list-sorted word-list))
(setf (gethash word (word-list-hashed word-list)) ind-offset)
(incf ind-offset)
) ) ) )
(dolist (word '("HANGUL" "SYLLABLE"))
(format ostream "#define UNICODE_CHARNAME_WORD_~A ~D~%" word
(gethash word (word-list-hashed (aref words-by-length (length word))))
) )
;; Compute the word-indices for every unicode-char.
(dolist (uc all-chars)
(let ((name (unicode-char-name uc))
(indices '()))
(let ((i1 0))
(loop
(when (>= i1 (length name)) (return))
(let ((i2 (or (position #\Space name :start i1) (length name))))
(let* ((word (subseq name i1 i2))
(len (length word)))
(push (gethash word (word-list-hashed (aref words-by-length len)))
indices
)
)
(setq i1 (1+ i2))
) ) )
(setf (unicode-char-word-indices uc)
(coerce (nreverse indices) 'vector)
)
) )
;; Sort the list of unicode-chars by word-indices.
(setq all-chars
(sort all-chars
(lambda (vec1 vec2)
(let ((len1 (length vec1))
(len2 (length vec2)))
(do ((i 0 (1+ i)))
(nil)
(if (< i len2)
(if (< i len1)
(cond ((< (aref vec1 i) (aref vec2 i)) (return t))
((> (aref vec1 i) (aref vec2 i)) (return nil))
)
(return t)
)
(return nil)
) ) ) )
:key #'unicode-char-word-indices
) )
;; Output the word-indices.
(format ostream "static const uint16 unicode_names[~D] = {~%"
(reduce #'+ (mapcar (lambda (uc) (length (unicode-char-word-indices uc))) all-chars))
)
(let ((i 0))
(dolist (uc all-chars)
(format ostream " ~{ ~D,~}"
(maplist (lambda (r) (+ (* 2 (car r)) (if (cdr r) 1 0)))
(coerce (unicode-char-word-indices uc) 'list)
)
)
(when add-comments
(format ostream "~40T/* ~A */" (unicode-char-name uc))
)
(format ostream "~%")
(setf (unicode-char-word-indices-index uc) i)
(incf i (length (unicode-char-word-indices uc)))
) )
(format ostream "};~%")
(format ostream "static const struct { uint16 code; uint16 name; } unicode_name_to_code[~D] = {~%"
(length all-chars)
)
(dolist (uc all-chars)
(format ostream " { 0x~4,'0X, ~D },"
(unicode-char-code uc)
(unicode-char-word-indices-index uc)
)
(when add-comments
(format ostream "~21T/* ~A */" (unicode-char-name uc))
)
(format ostream "~%")
)
(format ostream "};~%")
(format ostream "static const struct { uint16 code; uint16 name; } unicode_code_to_name[~D] = {~%"
(length all-chars)
)
(dolist (uc (sort (copy-list all-chars) #'< :key #'unicode-char-code))
(format ostream " { 0x~4,'0X, ~D },"
(unicode-char-code uc)
(unicode-char-word-indices-index uc)
)
(when add-comments
(format ostream "~21T/* ~A */" (unicode-char-name uc))
)
(format ostream "~%")
)
(format ostream "};~%")
(format ostream "#define UNICODE_CHARNAME_MAX_LENGTH ~D~%"
(reduce #'max (mapcar (lambda (uc) (length (unicode-char-name uc))) all-chars))
)
(format ostream "#define UNICODE_CHARNAME_MAX_WORDS ~D~%"
(reduce #'max (mapcar (lambda (uc) (length (unicode-char-word-indices uc))) all-chars))
)
)
) ) )
(main (first *args*) (second *args*))
|
