X-Git-Url: http://git.indexdata.com/?p=idzebra-moved-to-github.git;a=blobdiff_plain;f=include%2Fdict.h;h=73ae9ded5671f0001361fb7f5a51a5587b5b7d44;hp=4b75eead523dc08a08ee32b935b2692d27d49864;hb=896c0427df9d8eff5de6a1735dcd992e067df844;hpb=a4e93d62564084d6d1b6670d4fe75dd70a6ed139 diff --git a/include/dict.h b/include/dict.h index 4b75eea..73ae9de 100644 --- a/include/dict.h +++ b/include/dict.h @@ -1,110 +1,32 @@ -/* - * Copyright (C) 1994-1999, Index Data - * All rights reserved. - * Sebastian Hammer, Adam Dickmeiss - * - * $Log: dict.h,v $ - * Revision 1.28 1999-03-09 13:07:06 adam - * Work on dict_compact routine. - * - * Revision 1.27 1999/02/02 14:50:32 adam - * Updated WIN32 code specific sections. Changed header. - * - * Revision 1.26 1997/09/18 08:59:18 adam - * Extra generic handle for the character mapping routines. - * - * Revision 1.25 1997/09/17 12:19:09 adam - * Zebra version corresponds to YAZ version 1.4. - * Changed Zebra server so that it doesn't depend on global common_resource. - * - * Revision 1.24 1997/09/05 15:30:00 adam - * Changed prototype for chr_map_input - added const. - * Added support for C++, headers uses extern "C" for public definitions. - * - * Revision 1.23 1996/10/29 13:45:33 adam - * Changed definition of DICT_DEFAULT_PAGESIZE. - * - * Revision 1.22 1996/06/04 10:20:10 adam - * Added support for character mapping. - * - * Revision 1.21 1996/05/24 14:46:07 adam - * Added dict_grep_cmap function to define user-mapping in grep lookups. - * - * Revision 1.20 1996/03/20 09:35:23 adam - * Function dict_lookup_grep got extra parameter, init_pos, which marks - * from which position in pattern approximate pattern matching should occur. - * - * Revision 1.19 1996/02/02 13:43:54 adam - * The public functions simply use char instead of Dict_char to represent - * search strings. Dict_char is used internally only. - * - * Revision 1.18 1996/02/01 20:41:06 adam - * Bug fix: insert didn't work on 8-bit characters due to unsigned char - * compares in dict_strcmp (strcmp) and signed Dict_char. Dict_char is - * unsigned now. - * - * Revision 1.17 1995/12/07 11:47:04 adam - * Default pagesize is 4k instead of 8k. - * - * Revision 1.16 1995/12/06 14:41:13 adam - * New function: dict_delete. - * - * Revision 1.15 1995/10/27 13:59:17 adam - * Function dict_look_grep got extra parameter max_pos that upon return - * hold length of longest prefix that matches pattern. - * - * Revision 1.14 1995/10/09 16:18:35 adam - * Function dict_lookup_grep got extra client data parameter. - * - * Revision 1.13 1995/10/06 09:03:51 adam - * First version of scan. - * - * Revision 1.12 1995/09/14 11:53:02 adam - * Grep handle function parameter info is const now. - * - * Revision 1.11 1995/09/04 09:09:51 adam - * String arg in dict lookup is const. - * Minor changes. - * - * Revision 1.10 1994/10/05 12:16:58 adam - * Pagesize is a resource now. - * - * Revision 1.9 1994/10/04 12:08:19 adam - * Minor changes. - * - * Revision 1.8 1994/10/03 17:23:11 adam - * First version of dictionary lookup with regular expressions and errors. - * - * Revision 1.7 1994/09/22 10:44:47 adam - * Don't remember what changed!! - * - * Revision 1.6 1994/09/16 15:39:21 adam - * Initial code of lookup - not tested yet. - * - * Revision 1.5 1994/09/06 13:05:29 adam - * Further development of insertion. Some special cases are - * not properly handled yet! assert(0) are put here. The - * binary search in each page definitely reduce usr CPU. - * - * Revision 1.4 1994/09/01 17:44:40 adam - * Work on insertion in dictionary. Not finished yet. - * - * Revision 1.3 1994/08/18 12:41:12 adam - * Some development of dictionary. Not finished at all! - * - * Revision 1.2 1994/08/17 13:32:33 adam - * Use cache in dict - not in bfile. - * - * Revision 1.1 1994/08/16 16:26:53 adam - * Added dict. - * - */ +/* $Id: dict.h,v 1.32 2002-08-02 19:26:55 adam Exp $ + Copyright (C) 1995,1996,1997,1998,1999,2000,2001,2002 + Index Data Aps + +This file is part of the Zebra server. + +Zebra is free software; you can redistribute it and/or modify it under +the terms of the GNU General Public License as published by the Free +Software Foundation; either version 2, or (at your option) any later +version. + +Zebra is distributed in the hope that it will be useful, but WITHOUT ANY +WARRANTY; without even the implied warranty of MERCHANTABILITY or +FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License +for more details. + +You should have received a copy of the GNU General Public License +along with Zebra; see the file LICENSE.zebra. If not, write to the +Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA +02111-1307, USA. +*/ + + #ifndef DICT_H #define DICT_H #include -#include +#include #ifdef __cplusplus extern "C" { @@ -116,7 +38,8 @@ typedef unsigned char Dict_char; struct Dict_head { char magic_str[8]; int page_size; - Dict_ptr free_list, last; + int compact_flag; + Dict_ptr root, last, freelist; }; struct Dict_file_block @@ -126,6 +49,7 @@ struct Dict_file_block void *data; int dirty; int no; + int nbytes; }; typedef struct Dict_file_struct @@ -144,6 +68,7 @@ typedef struct Dict_file_struct int block_size; int hits; int misses; + int compact_flag; } *Dict_BFile; typedef struct Dict_struct { @@ -154,22 +79,26 @@ typedef struct Dict_struct { struct Dict_head head; } *Dict; -#define DICT_MAGIC "dict00" +#define DICT_MAGIC "dict01" #define DICT_DEFAULT_PAGESIZE 4096 int dict_bf_readp (Dict_BFile bf, int no, void **bufp); -int dict_bf_newp (Dict_BFile bf, int no, void **bufp); +int dict_bf_newp (Dict_BFile bf, int no, void **bufp, int nbytes); int dict_bf_touch (Dict_BFile bf, int no); void dict_bf_flush_blocks (Dict_BFile bf, int no_to_flush); Dict_BFile dict_bf_open (BFiles bfs, const char *name, int block_size, int cache, int rw); int dict_bf_close (Dict_BFile dbf); +void dict_bf_compact (Dict_BFile dbf); -Dict dict_open (BFiles bfs, const char *name, int cache, int rw); +Dict dict_open (BFiles bfs, const char *name, int cache, int rw, + int compact_flag); int dict_close (Dict dict); int dict_insert (Dict dict, const char *p, int userlen, void *userinfo); int dict_delete (Dict dict, const char *p); +int dict_delete_subtree (Dict dict, const char *p, void *client, + int (*f)(const char *info, void *client)); char *dict_lookup (Dict dict, const char *p); int dict_lookup_ec (Dict dict, char *p, int range, int (*f)(char *name)); @@ -178,6 +107,7 @@ int dict_lookup_grep (Dict dict, const char *p, int range, void *client, int (*f)(char *name, const char *info, void *client)); int dict_strcmp (const Dict_char *s1, const Dict_char *s2); +int dict_strncmp (const Dict_char *s1, const Dict_char *s2, size_t n); int dict_strlen (const Dict_char *s); int dict_scan (Dict dict, char *str, int *before, int *after, void *client, @@ -187,17 +117,17 @@ int dict_scan (Dict dict, char *str, void dict_grep_cmap (Dict dict, void *vp, const char **(*cmap)(void *vp, const char **from, int len)); -int dict_compact (BFiles bfs, const char *from, const char *to); +int dict_copy_compact (BFiles bfs, const char *from, const char *to); #define DICT_EOS 0 #define DICT_type(x) 0[(Dict_ptr*) x] #define DICT_backptr(x) 1[(Dict_ptr*) x] -#define DICT_nextptr(x) 2[(Dict_ptr*) x] -#define DICT_nodir(x) 0[(short*)((char*)(x)+3*sizeof(Dict_ptr))] -#define DICT_size(x) 1[(short*)((char*)(x)+3*sizeof(Dict_ptr))] -#define DICT_infoffset (3*sizeof(Dict_ptr)+2*sizeof(short)) -#define DICT_pagesize(x) ((x)->head.page_size) +#define DICT_bsize(x) 2[(short*)((char*)(x)+2*sizeof(Dict_ptr))] +#define DICT_nodir(x) 0[(short*)((char*)(x)+2*sizeof(Dict_ptr))] +#define DICT_size(x) 1[(short*)((char*)(x)+2*sizeof(Dict_ptr))] +#define DICT_infoffset (2*sizeof(Dict_ptr)+3*sizeof(short)) +#define DICT_xxxxpagesize(x) ((x)->head.page_size) #define DICT_to_str(x) sizeof(Dict_info)+sizeof(Dict_ptr)