1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/intl/hyphenation/src/hyphen.h Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,172 @@
1.4 +/* Hyphen - hyphenation library using converted TeX hyphenation patterns
1.5 + *
1.6 + * (C) 1998 Raph Levien
1.7 + * (C) 2001 ALTLinux, Moscow
1.8 + * (C) 2006, 2007, 2008 László Németh
1.9 + *
1.10 + * This was part of libHnj library by Raph Levien.
1.11 + *
1.12 + * Peter Novodvorsky from ALTLinux cut hyphenation part from libHnj
1.13 + * to use it in OpenOffice.org.
1.14 + *
1.15 + * Non-standard and compound word hyphenation support by László Németh.
1.16 + *
1.17 + * License is the original LibHnj license:
1.18 + *
1.19 + * LibHnj is dual licensed under LGPL and MPL. Boilerplate for both
1.20 + * licenses follows.
1.21 + */
1.22 +
1.23 +/* LibHnj - a library for high quality hyphenation and justification
1.24 + * Copyright (C) 1998 Raph Levien
1.25 + *
1.26 + * This library is free software; you can redistribute it and/or
1.27 + * modify it under the terms of the GNU Library General Public
1.28 + * License as published by the Free Software Foundation; either
1.29 + * version 2 of the License, or (at your option) any later version.
1.30 + *
1.31 + * This library is distributed in the hope that it will be useful,
1.32 + * but WITHOUT ANY WARRANTY; without even the implied warranty of
1.33 + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
1.34 + * Library General Public License for more details.
1.35 + *
1.36 + * You should have received a copy of the GNU Library General Public
1.37 + * License along with this library; if not, write to the
1.38 + * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
1.39 + * Boston, MA 02111-1307 USA.
1.40 +*/
1.41 +
1.42 +/*
1.43 + * The contents of this file are subject to the Mozilla Public License
1.44 + * Version 1.0 (the "MPL"); you may not use this file except in
1.45 + * compliance with the MPL. You may obtain a copy of the MPL at
1.46 + * http://www.mozilla.org/MPL/
1.47 + *
1.48 + * Software distributed under the MPL is distributed on an "AS IS" basis,
1.49 + * WITHOUT WARRANTY OF ANY KIND, either express or implied. See the MPL
1.50 + * for the specific language governing rights and limitations under the
1.51 + * MPL.
1.52 + *
1.53 + */
1.54 +#ifndef __HYPHEN_H__
1.55 +#define __HYPHEN_H__
1.56 +
1.57 +#ifdef __cplusplus
1.58 +extern "C" {
1.59 +#endif /* __cplusplus */
1.60 +
1.61 +typedef struct _HyphenDict HyphenDict;
1.62 +typedef struct _HyphenState HyphenState;
1.63 +typedef struct _HyphenTrans HyphenTrans;
1.64 +#define MAX_CHARS 100
1.65 +#define MAX_NAME 20
1.66 +
1.67 +struct _HyphenDict {
1.68 + /* user options */
1.69 + char lhmin; /* lefthyphenmin: min. hyph. distance from the left side */
1.70 + char rhmin; /* righthyphenmin: min. hyph. distance from the right side */
1.71 + char clhmin; /* min. hyph. distance from the left compound boundary */
1.72 + char crhmin; /* min. hyph. distance from the right compound boundary */
1.73 + char * nohyphen; /* comma separated list of characters or character
1.74 + sequences with forbidden hyphenation */
1.75 + int nohyphenl; /* count of elements in nohyphen */
1.76 + /* system variables */
1.77 + int num_states;
1.78 + char cset[MAX_NAME];
1.79 + int utf8;
1.80 + HyphenState *states;
1.81 + HyphenDict *nextlevel;
1.82 +};
1.83 +
1.84 +struct _HyphenState {
1.85 + char *match;
1.86 + char *repl;
1.87 + signed char replindex;
1.88 + signed char replcut;
1.89 + int fallback_state;
1.90 + int num_trans;
1.91 + HyphenTrans *trans;
1.92 +};
1.93 +
1.94 +struct _HyphenTrans {
1.95 + char ch;
1.96 + int new_state;
1.97 +};
1.98 +
1.99 +HyphenDict *hnj_hyphen_load (const char *fn);
1.100 +void hnj_hyphen_free (HyphenDict *dict);
1.101 +
1.102 +/* obsolete, use hnj_hyphen_hyphenate2() or *hyphenate3() functions) */
1.103 +int hnj_hyphen_hyphenate (HyphenDict *dict,
1.104 + const char *word, int word_size,
1.105 + char *hyphens);
1.106 +
1.107 +/*
1.108 +
1.109 + int hnj_hyphen_hyphenate2(): non-standard hyphenation.
1.110 +
1.111 + (It supports Catalan, Dutch, German, Hungarian, Norwegian, Swedish
1.112 + etc. orthography, see documentation.)
1.113 +
1.114 + input data:
1.115 + word: input word
1.116 + word_size: byte length of the input word
1.117 +
1.118 + hyphens: allocated character buffer (size = word_size + 5)
1.119 + hyphenated_word: allocated character buffer (size ~ word_size * 2) or NULL
1.120 + rep, pos, cut: pointers (point to the allocated and _zeroed_ buffers
1.121 + (size=word_size) or with NULL value) or NULL
1.122 +
1.123 + output data:
1.124 + hyphens: hyphenation vector (hyphenation points signed with odd numbers)
1.125 + hyphenated_word: hyphenated input word (hyphens signed with `='),
1.126 + optional (NULL input)
1.127 + rep: NULL (only standard hyph.), or replacements (hyphenation points
1.128 + signed with `=' in replacements);
1.129 + pos: NULL, or difference of the actual position and the beginning
1.130 + positions of the change in input words;
1.131 + cut: NULL, or counts of the removed characters of the original words
1.132 + at hyphenation,
1.133 +
1.134 + Note: rep, pos, cut are complementary arrays to the hyphens, indexed with the
1.135 + character positions of the input word.
1.136 +
1.137 + For example:
1.138 + Schiffahrt -> Schiff=fahrt,
1.139 + pattern: f1f/ff=f,1,2
1.140 + output: rep[5]="ff=f", pos[5] = 1, cut[5] = 2
1.141 +
1.142 + Note: hnj_hyphen_hyphenate2() can allocate rep, pos, cut (word_size
1.143 + length arrays):
1.144 +
1.145 + char ** rep = NULL;
1.146 + int * pos = NULL;
1.147 + int * cut = NULL;
1.148 + char hyphens[MAXWORDLEN];
1.149 + hnj_hyphen_hyphenate2(dict, "example", 7, hyphens, NULL, &rep, &pos, &cut);
1.150 +
1.151 + See example in the source distribution.
1.152 +
1.153 +*/
1.154 +
1.155 +int hnj_hyphen_hyphenate2 (HyphenDict *dict,
1.156 + const char *word, int word_size, char * hyphens,
1.157 + char *hyphenated_word, char *** rep, int ** pos, int ** cut);
1.158 +
1.159 +/* like hnj_hyphen_hyphenate2, but with hyphenmin parameters */
1.160 +/* lhmin: lefthyphenmin
1.161 + * rhmin: righthyphenmin
1.162 + * clhmin: compoundlefthyphemin
1.163 + * crhmin: compoundrighthyphenmin
1.164 + * (see documentation) */
1.165 +
1.166 +int hnj_hyphen_hyphenate3 (HyphenDict *dict,
1.167 + const char *word, int word_size, char * hyphens,
1.168 + char *hyphword, char *** rep, int ** pos, int ** cut,
1.169 + int lhmin, int rhmin, int clhmin, int crhmin);
1.170 +
1.171 +#ifdef __cplusplus
1.172 +}
1.173 +#endif /* __cplusplus */
1.174 +
1.175 +#endif /* __HYPHEN_H__ */
