Home | History | Annotate | Line # | Download | only in libuniname
      1 #!/usr/local/bin/clisp -C
      2 
      3 ;;; Creation of CLISP's uni_names.h from the UnicodeData.txt table.
      4 ;;; Bruno Haible 2000-12-28
      5 
      6 (defparameter add-comments nil)
      7 
      8 (defstruct unicode-char
      9   (code nil :type integer)
     10   (name nil :type string)
     11   word-indices
     12   word-indices-index
     13 )
     14 
     15 (defstruct word-list
     16   (hashed nil :type hash-table)
     17   (sorted nil :type list)
     18   size                          ; number of characters total
     19   length                        ; number of words
     20 )
     21 
     22 (defun main (inputfile outputfile)
     23   (declare (type string inputfile outputfile))
     24   #+UNICODE (setq *default-file-encoding* charset:utf-8)
     25   (let ((all-chars '()))
     26     ;; Read all characters and names from the input file.
     27     (with-open-file (istream inputfile :direction :input)
     28       (loop
     29         (let ((line (read-line istream nil nil)))
     30           (unless line (return))
     31           (let* ((i1 (position #\; line))
     32                  (i2 (position #\; line :start (1+ i1)))
     33                  (code-string (subseq line 0 i1))
     34                  (code (parse-integer code-string :radix 16))
     35                  (name-string (subseq line (1+ i1) i2)))
     36             ; Ignore characters whose name starts with "<".
     37             (unless (eql (char name-string 0) #\<)
     38               ; Also ignore Hangul syllables; they are treated specially.
     39               (unless (<= #xAC00 code #xD7A3)
     40                 ; Also ignore CJK compatibility ideographs; they are treated
     41                 ; specially as well.
     42                 (unless (or (<= #xF900 code #xFA2D) (<= #xFA30 code #xFA6A)
     43                             (<= #xFA70 code #xFAD9) (<= #x2F800 code #x2FA1D))
     44                   ; Transform the code so that it fits in 16 bits. In
     45                   ; Unicode 3.1 the following ranges are used.
     46                   ;   0x00000..0x04DFF  >>12=  0x00..0x04  -> 0x0..0x4
     47                   ;   0x0A000..0x0A4FF  >>12=  0x0A        -> 0x5
     48                   ;   0x0F900..0x0FFFF  >>12=  0x0F        -> 0x6
     49                   ;   0x10300..0x104FF  >>12=  0x10        -> 0x7
     50                   ;   0x1D000..0x1D7DD  >>12=  0x1D        -> 0x8
     51                   ;   0x2F800..0x2FAFF  >>12=  0x2F        -> 0x9
     52                   ;   0xE0000..0xE00FF  >>12=  0xE0        -> 0xA
     53                   (flet ((transform (x)
     54                            (dpb
     55                              (case (ash x -12)
     56                                ((#x00 #x01 #x02 #x03 #x04) (ash x -12))
     57                                (#x0A 5)
     58                                (#x0F 6)
     59                                (#x10 7)
     60                                (#x1D 8)
     61                                (#x2F 9)
     62                                (#xE0 #xA)
     63                                (t (error "Update the transform function for 0x~5,'0X" x))
     64                              )
     65                              (byte 8 12)
     66                              x
     67                         )) )
     68                     (push (make-unicode-char :code (transform code)
     69                                              :name name-string)
     70                           all-chars
     71             ) ) ) ) )
     72     ) ) ) )
     73     (setq all-chars (nreverse all-chars))
     74     ;; Split into words.
     75     (let ((words-by-length (make-array 0 :adjustable t)))
     76       (dolist (name (list* "HANGUL SYLLABLE" "CJK COMPATIBILITY" (mapcar #'unicode-char-name all-chars)))
     77         (let ((i1 0))
     78           (loop
     79             (when (>= i1 (length name)) (return))
     80             (let ((i2 (or (position #\Space name :start i1) (length name))))
     81               (let* ((word (subseq name i1 i2))
     82                      (len (length word)))
     83                 (when (>= len (length words-by-length))
     84                   (adjust-array words-by-length (1+ len))
     85                 )
     86                 (unless (aref words-by-length len)
     87                   (setf (aref words-by-length len)
     88                         (make-word-list
     89                           :hashed (make-hash-table :test #'equal)
     90                           :sorted '()
     91                 ) )     )
     92                 (let ((word-list (aref words-by-length len)))
     93                   (unless (gethash word (word-list-hashed word-list))
     94                     (setf (gethash word (word-list-hashed word-list)) t)
     95                     (push word (word-list-sorted word-list))
     96                 ) )
     97               )
     98               (setq i1 (1+ i2))
     99       ) ) ) )
    100       ;; Sort the word lists.
    101       (dotimes (len (length words-by-length))
    102         (unless (aref words-by-length len)
    103           (setf (aref words-by-length len)
    104                 (make-word-list
    105                   :hashed (make-hash-table :test #'equal)
    106                   :sorted '()
    107         ) )     )
    108         (let ((word-list (aref words-by-length len)))
    109           (setf (word-list-sorted word-list)
    110                 (sort (word-list-sorted word-list) #'string<)
    111           )
    112           (setf (word-list-size word-list)
    113                 (reduce #'+ (mapcar #'length (word-list-sorted word-list)))
    114           )
    115           (setf (word-list-length word-list)
    116                 (length (word-list-sorted word-list))
    117       ) ) )
    118       ;; Output the tables.
    119       (with-open-file (ostream outputfile :direction :output
    120                        #+UNICODE :external-format #+UNICODE charset:ascii)
    121         (format ostream "/*~%")
    122         (format ostream " * ~A~%" (file-namestring outputfile))
    123         (format ostream " *~%")
    124         (format ostream " * Unicode character name table.~%")
    125         (format ostream " * Generated automatically by the gen-uninames utility.~%")
    126         (format ostream " */~%")
    127         (format ostream "~%")
    128         (format ostream "static const char unicode_name_words[~D] = {~%"
    129                         (let ((sum 0))
    130                           (dotimes (len (length words-by-length))
    131                             (let ((word-list (aref words-by-length len)))
    132                               (incf sum (word-list-size word-list))
    133                           ) )
    134                           sum
    135         )               )
    136         (dotimes (len (length words-by-length))
    137           (let ((word-list (aref words-by-length len)))
    138             (dolist (word (word-list-sorted word-list))
    139               (format ostream " ~{ '~C',~}~%" (coerce word 'list))
    140         ) ) )
    141         (format ostream "};~%")
    142         (format ostream "#define UNICODE_CHARNAME_NUM_WORDS ~D~%"
    143                         (let ((sum 0))
    144                           (dotimes (len (length words-by-length))
    145                             (let ((word-list (aref words-by-length len)))
    146                               (incf sum (word-list-length word-list))
    147                           ) )
    148                           sum
    149         )               )
    150         #| ; Redundant data
    151         (format ostream "static const uint16_t unicode_name_word_offsets[~D] = {~%"
    152                         (let ((sum 0))
    153                           (dotimes (len (length words-by-length))
    154                             (let ((word-list (aref words-by-length len)))
    155                               (incf sum (word-list-length word-list))
    156                           ) )
    157                           sum
    158         )               )
    159         (dotimes (len (length words-by-length))
    160           (let ((word-list (aref words-by-length len)))
    161             (when (word-list-sorted word-list)
    162               (format ostream " ")
    163               (do ((l (word-list-sorted word-list) (cdr l))
    164                    (offset 0 (+ offset (length (car l)))))
    165                   ((endp l))
    166                 (format ostream "~<~% ~0,79:; ~D,~>" offset)
    167               )
    168               (format ostream "~%")
    169         ) ) )
    170         (format ostream "};~%")
    171         |#
    172         (format ostream "static const struct { uint16_t extra_offset; uint16_t ind_offset; } unicode_name_by_length[~D] = {~%"
    173                         (1+ (length words-by-length))
    174         )
    175         (let ((extra-offset 0)
    176               (ind-offset 0))
    177           (dotimes (len (length words-by-length))
    178             (let ((word-list (aref words-by-length len)))
    179               (format ostream "  { ~D, ~D },~%" extra-offset ind-offset)
    180               (incf extra-offset (word-list-size word-list))
    181               (incf ind-offset (word-list-length word-list))
    182           ) )
    183           (format ostream "  { ~D, ~D }~%" extra-offset ind-offset)
    184         )
    185         (format ostream "};~%")
    186         (let ((ind-offset 0))
    187           (dotimes (len (length words-by-length))
    188             (let ((word-list (aref words-by-length len)))
    189               (dolist (word (word-list-sorted word-list))
    190                 (setf (gethash word (word-list-hashed word-list)) ind-offset)
    191                 (incf ind-offset)
    192         ) ) ) )
    193         (dolist (word '("HANGUL" "SYLLABLE" "CJK" "COMPATIBILITY"))
    194           (format ostream "#define UNICODE_CHARNAME_WORD_~A ~D~%" word
    195                           (gethash word (word-list-hashed (aref words-by-length (length word))))
    196         ) )
    197         ;; Compute the word-indices for every unicode-char.
    198         (dolist (uc all-chars)
    199           (let ((name (unicode-char-name uc))
    200                 (indices '()))
    201             (let ((i1 0))
    202               (loop
    203                 (when (>= i1 (length name)) (return))
    204                 (let ((i2 (or (position #\Space name :start i1) (length name))))
    205                   (let* ((word (subseq name i1 i2))
    206                          (len (length word)))
    207                     (push (gethash word (word-list-hashed (aref words-by-length len)))
    208                           indices
    209                     )
    210                   )
    211                   (setq i1 (1+ i2))
    212             ) ) )
    213             (setf (unicode-char-word-indices uc)
    214                   (coerce (nreverse indices) 'vector)
    215             )
    216         ) )
    217         ;; Sort the list of unicode-chars by word-indices.
    218         (setq all-chars
    219               (sort all-chars
    220                     (lambda (vec1 vec2)
    221                       (let ((len1 (length vec1))
    222                             (len2 (length vec2)))
    223                         (do ((i 0 (1+ i)))
    224                             (nil)
    225                           (if (< i len2)
    226                             (if (< i len1)
    227                               (cond ((< (aref vec1 i) (aref vec2 i)) (return t))
    228                                     ((> (aref vec1 i) (aref vec2 i)) (return nil))
    229                               )
    230                               (return t)
    231                             )
    232                             (return nil)
    233                     ) ) ) )
    234                     :key #'unicode-char-word-indices
    235         )     )
    236         ;; Output the word-indices.
    237         (format ostream "static const uint16_t unicode_names[~D] = {~%"
    238                         (reduce #'+ (mapcar (lambda (uc) (length (unicode-char-word-indices uc))) all-chars))
    239         )
    240         (let ((i 0))
    241           (dolist (uc all-chars)
    242             (format ostream " ~{ ~D,~}"
    243                             (maplist (lambda (r) (+ (* 2 (car r)) (if (cdr r) 1 0)))
    244                                      (coerce (unicode-char-word-indices uc) 'list)
    245                             )
    246             )
    247             (when add-comments
    248               (format ostream "~40T/* ~A */" (unicode-char-name uc))
    249             )
    250             (format ostream "~%")
    251             (setf (unicode-char-word-indices-index uc) i)
    252             (incf i (length (unicode-char-word-indices uc)))
    253         ) )
    254         (format ostream "};~%")
    255         (format ostream "static const struct { uint16_t code; uint16_t name; } unicode_name_to_code[~D] = {~%"
    256                         (length all-chars)
    257         )
    258         (dolist (uc all-chars)
    259           (format ostream "  { 0x~4,'0X, ~D },"
    260                           (unicode-char-code uc)
    261                           (unicode-char-word-indices-index uc)
    262           )
    263           (when add-comments
    264             (format ostream "~21T/* ~A */" (unicode-char-name uc))
    265           )
    266           (format ostream "~%")
    267         )
    268         (format ostream "};~%")
    269         (format ostream "static const struct { uint16_t code; uint16_t name; } unicode_code_to_name[~D] = {~%"
    270                         (length all-chars)
    271         )
    272         (dolist (uc (sort (copy-list all-chars) #'< :key #'unicode-char-code))
    273           (format ostream "  { 0x~4,'0X, ~D },"
    274                           (unicode-char-code uc)
    275                           (unicode-char-word-indices-index uc)
    276           )
    277           (when add-comments
    278             (format ostream "~21T/* ~A */" (unicode-char-name uc))
    279           )
    280           (format ostream "~%")
    281         )
    282         (format ostream "};~%")
    283         (format ostream "#define UNICODE_CHARNAME_MAX_LENGTH ~D~%"
    284                         (reduce #'max (mapcar (lambda (uc) (length (unicode-char-name uc))) all-chars))
    285         )
    286         (format ostream "#define UNICODE_CHARNAME_MAX_WORDS ~D~%"
    287                         (reduce #'max (mapcar (lambda (uc) (length (unicode-char-word-indices uc))) all-chars))
    288         )
    289       )
    290 ) ) )
    291 
    292 (main (first *args*) (second *args*))
    293