1 /* $Id: rectext.c,v 1.19 2004-08-06 12:55:03 adam Exp $
2 Copyright (C) 1995,1996,1997,1998,1999,2000,2001,2002,2003,2004
5 This file is part of the Zebra server.
7 Zebra is free software; you can redistribute it and/or modify it under
8 the terms of the GNU General Public License as published by the Free
9 Software Foundation; either version 2, or (at your option) any later
12 Zebra is distributed in the hope that it will be useful, but WITHOUT ANY
13 WARRANTY; without even the implied warranty of MERCHANTABILITY or
14 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
17 You should have received a copy of the GNU General Public License
18 along with Zebra; see the file LICENSE.zebra. If not, write to the
19 Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA
35 static void *text_init (RecType recType)
37 struct text_info *tinfo = (struct text_info *) xmalloc(sizeof(*tinfo));
42 static void text_destroy (void *clientData)
44 struct text_info *tinfo = clientData;
50 struct recExtractCtrl *p;
56 struct buf_info *buf_open (struct recExtractCtrl *p)
58 struct buf_info *fi = (struct buf_info *) xmalloc (sizeof(*fi));
61 fi->buf = (char *) xmalloc (4096);
67 int buf_read (struct text_info *tinfo, struct buf_info *fi, char *dst)
69 if (fi->offset >= fi->max)
73 fi->max = (*fi->p->readf)(fi->p->fh, fi->buf, 4096);
78 *dst = fi->buf[(fi->offset)++];
79 if (tinfo->sep && *dst == *tinfo->sep)
81 off_t off = (*fi->p->tellf)(fi->p->fh);
82 (*fi->p->endf)(fi->p->fh, off - (fi->max - fi->offset));
88 void buf_close (struct buf_info *fi)
94 static int text_extract (void *clientData, struct recExtractCtrl *p)
96 struct text_info *tinfo = clientData;
100 struct buf_info *fi = buf_open (p);
103 yaz_log(LOG_LOG, "text_extract off=%ld",
104 (long) (*fi->p->tellf)(fi->p->fh));
109 if (!strncmp(p->subType, "sep=", 4))
110 tinfo->sep = xstrdup(p->subType+4);
112 (*p->init)(p, &recWord);
113 recWord.reg_type = 'w';
118 r = buf_read (tinfo, fi, w);
119 while (r > 0 && i < 511 && w[i] != '\n' && w[i] != '\r')
122 r = buf_read (tinfo, fi, w + i);
128 (*p->tokenAdd)(&recWord);
132 return RECCTRL_EXTRACT_OK;
135 static int text_retrieve (void *clientData, struct recRetrieveCtrl *p)
138 static char *text_buf = NULL;
139 static int text_size = 0;
142 const char *elementSetName = NULL;
145 if (p->comp && p->comp->which == Z_RecordComp_simple &&
146 p->comp->u.simple->which == Z_ElementSetNames_generic)
147 elementSetName = p->comp->u.simple->u.generic;
151 /* don't make header for the R(aw) element set name */
152 if (!strcmp(elementSetName, "R"))
157 /* only make header for the H(eader) element set name */
158 else if (!strcmp(elementSetName, "H"))
166 if (text_ptr + 4096 >= text_size)
170 text_size = 2*text_size + 8192;
171 nb = (char *) xmalloc (text_size);
174 memcpy (nb, text_buf, text_ptr);
179 if (make_header && text_ptr == 0)
183 sprintf (text_buf, "Rank: %d\n", p->score);
184 text_ptr = strlen(text_buf);
186 sprintf (text_buf + text_ptr, "Local Number: " ZINT_FORMAT "\n",
188 text_ptr = strlen(text_buf);
191 sprintf (text_buf + text_ptr, "Filename: %s\n", p->fname);
192 text_ptr = strlen(text_buf);
194 strcpy(text_buf+text_ptr++, "\n");
198 r = (*p->readf)(p->fh, text_buf + text_ptr, 4096);
203 text_buf[text_ptr] = '\0';
206 if (!strcmp (elementSetName, "B"))
208 if (!strcmp (elementSetName, "M"))
216 while (++i <= no_lines && (p = strchr (p, '\n')))
221 text_ptr = p-text_buf;
224 p->output_format = VAL_SUTRS;
225 p->rec_buf = text_buf;
226 p->rec_len = text_ptr;
230 static struct recType text_type = {
238 RecType recTypeText = &text_type;