intl/hyphenation/hyphen.h

   1 /* Hyphen - hyphenation library using converted TeX hyphenation patterns
   2  *
   3  * (C) 1998 Raph Levien
   4  * (C) 2001 ALTLinux, Moscow
   5  * (C) 2006, 2007, 2008 László Németh
   6  *
   7  * This was part of libHnj library by Raph Levien.
   8  *
   9  * Peter Novodvorsky from ALTLinux cut hyphenation part from libHnj
  10  * to use it in OpenOffice.org.
  11  *
  12  * Non-standard and compound word hyphenation support by László Németh.
  13  *
  14  * License is the original LibHnj license:
  15  *
  16  * LibHnj is dual licensed under LGPL and MPL. Boilerplate for both
  17  * licenses follows.
  18  */
  19
  20 /* LibHnj - a library for high quality hyphenation and justification
  21  * Copyright (C) 1998 Raph Levien
  22  *
  23  * This library is free software; you can redistribute it and/or
  24  * modify it under the terms of the GNU Library General Public
  25  * License as published by the Free Software Foundation; either
  26  * version 2 of the License, or (at your option) any later version.
  27  *
  28  * This library is distributed in the hope that it will be useful,
  29  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  30  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  31  * Library General Public License for more details.
  32  *
  33  * You should have received a copy of the GNU Library General Public
  34  * License along with this library; if not, write to the
  35  * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
  36  * Boston, MA  02111-1307  USA.
  37 */
  38
  39 /*
  40  * The contents of this file are subject to the Mozilla Public License
  41  * Version 1.0 (the "MPL"); you may not use this file except in
  42  * compliance with the MPL.  You may obtain a copy of the MPL at
  43  * http://www.mozilla.org/MPL/
  44  *
  45  * Software distributed under the MPL is distributed on an "AS IS" basis,
  46  * WITHOUT WARRANTY OF ANY KIND, either express or implied. See the MPL
  47  * for the specific language governing rights and limitations under the
  48  * MPL.
  49  *
  50  */
  51 #ifndef __HYPHEN_H__
  52 #define __HYPHEN_H__
  53
  54 #ifdef __cplusplus
  55 extern "C" {
  56 #endif /* __cplusplus */
  57
  58 typedef struct _HyphenDict HyphenDict;
  59 typedef struct _HyphenState HyphenState;
  60 typedef struct _HyphenTrans HyphenTrans;
  61 #define MAX_CHARS 100
  62 #define MAX_NAME 20
  63
  64 struct _HyphenDict {
  65   /* user options */
  66   char lhmin;    /* lefthyphenmin: min. hyph. distance from the left side */
  67   char rhmin;    /* righthyphenmin: min. hyph. distance from the right side */
  68   char clhmin;   /* min. hyph. distance from the left compound boundary */
  69   char crhmin;   /* min. hyph. distance from the right compound boundary */
  70   char * nohyphen; /* comma separated list of characters or character
  71                     sequences with forbidden hyphenation */
  72   int nohyphenl; /* count of elements in nohyphen */
  73   /* system variables */
  74   int num_states;
  75   char cset[MAX_NAME];
  76   int utf8;
  77   HyphenState *states;
  78   HyphenDict *nextlevel;
  79 };
  80
  81 struct _HyphenState {
  82   char *match;
  83   char *repl;
  84   signed char replindex;
  85   signed char replcut;
  86   int fallback_state;
  87   int num_trans;
  88   HyphenTrans *trans;
  89 };
  90
  91 struct _HyphenTrans {
  92   char ch;
  93   int new_state;
  94 };
  95
  96 HyphenDict *hnj_hyphen_load (const char *fn);
  97 void hnj_hyphen_free (HyphenDict *dict);
  98
  99 /* obsolete, use hnj_hyphen_hyphenate2() or *hyphenate3() functions) */
 100 int hnj_hyphen_hyphenate (HyphenDict *dict,
 101                            const char *word, int word_size,
 102                            char *hyphens);
 103
 104 /*
 105
 106  int hnj_hyphen_hyphenate2(): non-standard hyphenation.
 107
 108  (It supports Catalan, Dutch, German, Hungarian, Norwegian, Swedish
 109   etc. orthography, see documentation.)
 110
 111  input data:
 112  word:      input word
 113  word_size: byte length of the input word
 114
 115  hyphens:   allocated character buffer (size = word_size + 5)
 116  hyphenated_word: allocated character buffer (size ~ word_size * 2) or NULL
 117  rep, pos, cut: pointers (point to the allocated and _zeroed_ buffers
 118                 (size=word_size) or with NULL value) or NULL
 119
 120  output data:
 121  hyphens:   hyphenation vector (hyphenation points signed with odd numbers)
 122  hyphenated_word: hyphenated input word (hyphens signed with `='),
 123                   optional (NULL input)
 124  rep:       NULL (only standard hyph.), or replacements (hyphenation points
 125             signed with `=' in replacements);
 126  pos:       NULL, or difference of the actual position and the beginning
 127             positions of the change in input words;
 128  cut:       NULL, or counts of the removed characters of the original words
 129             at hyphenation,
 130
 131  Note: rep, pos, cut are complementary arrays to the hyphens, indexed with the
 132        character positions of the input word.
 133
 134  For example:
 135  Schiffahrt -> Schiff=fahrt,
 136  pattern: f1f/ff=f,1,2
 137  output: rep[5]="ff=f", pos[5] = 1, cut[5] = 2
 138
 139  Note: hnj_hyphen_hyphenate2() can allocate rep, pos, cut (word_size
 140        length arrays):
 141
 142  char ** rep = NULL;
 143  int * pos = NULL;
 144  int * cut = NULL;
 145  char hyphens[MAXWORDLEN];
 146  hnj_hyphen_hyphenate2(dict, "example", 7, hyphens, NULL, &rep, &pos, &cut);
 147
 148  See example in the source distribution.
 149
 150 */
 151
 152 int hnj_hyphen_hyphenate2 (HyphenDict *dict,
 153         const char *word, int word_size, char * hyphens,
 154         char *hyphenated_word, char *** rep, int ** pos, int ** cut);
 155
 156 /* like hnj_hyphen_hyphenate2, but with hyphenmin parameters */
 157 /* lhmin: lefthyphenmin
 158  * rhmin: righthyphenmin
 159  * clhmin: compoundlefthyphemin
 160  * crhmin: compoundrighthyphenmin
 161  * (see documentation) */
 162
 163 int hnj_hyphen_hyphenate3 (HyphenDict *dict,
 164         const char *word, int word_size, char * hyphens,
 165         char *hyphword, char *** rep, int ** pos, int ** cut,
 166         int lhmin, int rhmin, int clhmin, int crhmin);
 167
 168 #ifdef __cplusplus
 169 }
 170 #endif /* __cplusplus */
 171
 172 #endif /* __HYPHEN_H__ */