ext/hunspell/hunspell.hxx

   1 #include "hunvisapi.h"
   2
   3 #include "hashmgr.hxx"
   4 #include "affixmgr.hxx"
   5 #include "suggestmgr.hxx"
   6 #include "langnum.hxx"
   7
   8 #define  SPELL_XML "<?xml?>"
   9
  10 #define MAXDIC 20
  11 #define MAXSUGGESTION 15
  12 #define MAXSHARPS 5
  13
  14 #define HUNSPELL_OK       (1 << 0)
  15 #define HUNSPELL_OK_WARN  (1 << 1)
  16
  17 #ifndef _MYSPELLMGR_HXX_
  18 #define _MYSPELLMGR_HXX_
  19
  20 class LIBHUNSPELL_DLL_EXPORTED Hunspell
  21 {
  22   AffixMgr*       pAMgr;
  23   HashMgr*        pHMgr[MAXDIC];
  24   int             maxdic;
  25   SuggestMgr*     pSMgr;
  26   char *          affixpath;
  27   char *          encoding;
  28   struct cs_info * csconv;
  29   int             langnum;
  30   int             utf8;
  31   int             complexprefixes;
  32   char**          wordbreak;
  33
  34 public:
  35
  36   /* Hunspell(aff, dic) - constructor of Hunspell class
  37    * input: path of affix file and dictionary file
  38    */
  39
  40   Hunspell(const char * affpath, const char * dpath, const char * key = NULL);
  41   ~Hunspell();
  42
  43   /* load extra dictionaries (only dic files) */
  44   int add_dic(const char * dpath, const char * key = NULL);
  45
  46   /* spell(word) - spellcheck word
  47    * output: 0 = bad word, not 0 = good word
  48    *
  49    * plus output:
  50    *   info: information bit array, fields:
  51    *     SPELL_COMPOUND  = a compound word
  52    *     SPELL_FORBIDDEN = an explicit forbidden word
  53    *   root: root (stem), when input is a word with affix(es)
  54    */
  55
  56   int spell(const char * word, int * info = NULL, char ** root = NULL);
  57
  58   /* suggest(suggestions, word) - search suggestions
  59    * input: pointer to an array of strings pointer and the (bad) word
  60    *   array of strings pointer (here *slst) may not be initialized
  61    * output: number of suggestions in string array, and suggestions in
  62    *   a newly allocated array of strings (*slts will be NULL when number
  63    *   of suggestion equals 0.)
  64    */
  65
  66   int suggest(char*** slst, const char * word);
  67
  68   /* deallocate suggestion lists */
  69
  70   void free_list(char *** slst, int n);
  71
  72   char * get_dic_encoding();
  73
  74  /* morphological functions */
  75
  76  /* analyze(result, word) - morphological analysis of the word */
  77
  78   int analyze(char*** slst, const char * word);
  79
  80  /* stem(result, word) - stemmer function */
  81
  82   int stem(char*** slst, const char * word);
  83
  84  /* stem(result, analysis, n) - get stems from a morph. analysis
  85   * example:
  86   * char ** result, result2;
  87   * int n1 = analyze(&result, "words");
  88   * int n2 = stem(&result2, result, n1);
  89   */
  90
  91   int stem(char*** slst, char ** morph, int n);
  92
  93  /* generate(result, word, word2) - morphological generation by example(s) */
  94
  95   int generate(char*** slst, const char * word, const char * word2);
  96
  97  /* generate(result, word, desc, n) - generation by morph. description(s)
  98   * example:
  99   * char ** result;
 100   * char * affix = "is:plural"; // description depends from dictionaries, too
 101   * int n = generate(&result, "word", &affix, 1);
 102   * for (int i = 0; i < n; i++) printf("%s\n", result[i]);
 103   */
 104
 105   int generate(char*** slst, const char * word, char ** desc, int n);
 106
 107   /* functions for run-time modification of the dictionary */
 108
 109   /* add word to the run-time dictionary */
 110
 111   int add(const char * word);
 112
 113   /* add word to the run-time dictionary with affix flags of
 114    * the example (a dictionary word): Hunspell will recognize
 115    * affixed forms of the new word, too.
 116    */
 117
 118   int add_with_affix(const char * word, const char * example);
 119
 120   /* remove word from the run-time dictionary */
 121
 122   int remove(const char * word);
 123
 124   /* other */
 125
 126   /* get extra word characters definied in affix file for tokenization */
 127   const char * get_wordchars();
 128   unsigned short * get_wordchars_utf16(int * len);
 129
 130   struct cs_info * get_csconv();
 131   const char * get_version();
 132
 133   int get_langnum() const;
 134
 135   /* experimental and deprecated functions */
 136
 137 #ifdef HUNSPELL_EXPERIMENTAL
 138   /* suffix is an affix flag string, similarly in dictionary files */
 139   int put_word_suffix(const char * word, const char * suffix);
 140   char * morph_with_correction(const char * word);
 141
 142   /* spec. suggestions */
 143   int suggest_auto(char*** slst, const char * word);
 144   int suggest_pos_stems(char*** slst, const char * word);
 145 #endif
 146
 147 private:
 148    int    cleanword(char *, const char *, int * pcaptype, int * pabbrev);
 149    int    cleanword2(char *, const char *, w_char *, int * w_len, int * pcaptype, int * pabbrev);
 150    void   mkinitcap(char *);
 151    int    mkinitcap2(char * p, w_char * u, int nc);
 152    int    mkinitsmall2(char * p, w_char * u, int nc);
 153    void   mkallcap(char *);
 154    int    mkallcap2(char * p, w_char * u, int nc);
 155    void   mkallsmall(char *);
 156    int    mkallsmall2(char * p, w_char * u, int nc);
 157    struct hentry * checkword(const char *, int * info, char **root);
 158    char * sharps_u8_l1(char * dest, char * source);
 159    hentry * spellsharps(char * base, char *, int, int, char * tmp, int * info, char **root);
 160    int    is_keepcase(const hentry * rv);
 161    int    insert_sug(char ***slst, char * word, int ns);
 162    void   cat_result(char * result, char * st);
 163    char * stem_description(const char * desc);
 164    int    spellml(char*** slst, const char * word);
 165    int    get_xml_par(char * dest, const char * par, int maxl);
 166    const char * get_xml_pos(const char * s, const char * attr);
 167    int    get_xml_list(char ***slst, char * list, const char * tag);
 168    int    check_xml_par(const char * q, const char * attr, const char * value);
 169
 170 };
 171
 172 #endif