db/btree/bt_delete.c

   1 /*
   2  * Copyright (c) 1999 Apple Computer, Inc. All rights reserved.
   3  *
   4  * @APPLE_LICENSE_HEADER_START@
   5  *
   6  * Copyright (c) 1999-2003 Apple Computer, Inc.  All Rights Reserved.
   7  *
   8  * This file contains Original Code and/or Modifications of Original Code
   9  * as defined in and that are subject to the Apple Public Source License
  10  * Version 2.0 (the 'License'). You may not use this file except in
  11  * compliance with the License. Please obtain a copy of the License at
  12  * http://www.opensource.apple.com/apsl/ and read it before using this
  13  * file.
  14  *
  15  * The Original Code and all software distributed under the License are
  16  * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
  17  * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
  18  * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
  19  * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
  20  * Please see the License for the specific language governing rights and
  21  * limitations under the License.
  22  *
  23  * @APPLE_LICENSE_HEADER_END@
  24  */
  25 /*
  26  * Copyright (c) 1990, 1993
  27  *      The Regents of the University of California.  All rights reserved.
  28  *
  29  * This code is derived from software contributed to Berkeley by
  30  * Mike Olson.
  31  *
  32  * Redistribution and use in source and binary forms, with or without
  33  * modification, are permitted provided that the following conditions
  34  * are met:
  35  * 1. Redistributions of source code must retain the above copyright
  36  *    notice, this list of conditions and the following disclaimer.
  37  * 2. Redistributions in binary form must reproduce the above copyright
  38  *    notice, this list of conditions and the following disclaimer in the
  39  *    documentation and/or other materials provided with the distribution.
  40  * 3. All advertising materials mentioning features or use of this software
  41  *    must display the following acknowledgement:
  42  *      This product includes software developed by the University of
  43  *      California, Berkeley and its contributors.
  44  * 4. Neither the name of the University nor the names of its contributors
  45  *    may be used to endorse or promote products derived from this software
  46  *    without specific prior written permission.
  47  *
  48  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  49  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  50  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  51  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  52  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  53  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  54  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  55  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  56  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  57  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  58  * SUCH DAMAGE.
  59  */
  60
  61
  62 #include <sys/types.h>
  63
  64 #include <errno.h>
  65 #include <stdio.h>
  66 #include <string.h>
  67
  68 #include <db.h>
  69 #include "btree.h"
  70
  71 static int bt_bdelete __P((BTREE *, const DBT *));
  72
  73 /*
  74  * __BT_DELETE -- Delete the item(s) referenced by a key.
  75  *
  76  * Parameters:
  77  *      dbp:    pointer to access method
  78  *      key:    key to delete
  79  *      flags:  R_CURSOR if deleting what the cursor references
  80  *
  81  * Returns:
  82  *      RET_ERROR, RET_SUCCESS and RET_SPECIAL if the key not found.
  83  */
  84 int
  85 __bt_delete(dbp, key, flags)
  86         const DB *dbp;
  87         const DBT *key;
  88         u_int flags;
  89 {
  90         BTREE *t;
  91         int status;
  92
  93         t = dbp->internal;
  94
  95         /* Toss any page pinned across calls. */
  96         if (t->bt_pinned != NULL) {
  97                 mpool_put(t->bt_mp, t->bt_pinned, 0);
  98                 t->bt_pinned = NULL;
  99         }
 100
 101         if (ISSET(t, B_RDONLY)) {
 102                 errno = EPERM;
 103                 return (RET_ERROR);
 104         }
 105
 106         switch(flags) {
 107         case 0:
 108                 status = bt_bdelete(t, key);
 109                 break;
 110         case R_CURSOR:
 111                 /*
 112                  * If flags is R_CURSOR, delete the cursor; must already have
 113                  * started a scan and not have already deleted the record.  For
 114                  * the delete cursor bit to have been set requires that the
 115                  * scan be initialized, so no reason to check.
 116                  */
 117                 if (!ISSET(t, B_SEQINIT))
 118                         goto einval;
 119                 status = ISSET(t, B_DELCRSR) ?
 120                     RET_SPECIAL : __bt_crsrdel(t, &t->bt_bcursor);
 121                 break;
 122         default:
 123 einval:         errno = EINVAL;
 124                 return (RET_ERROR);
 125         }
 126         if (status == RET_SUCCESS)
 127                 SET(t, B_MODIFIED);
 128         return (status);
 129 }
 130
 131 /*
 132  * BT_BDELETE -- Delete all key/data pairs matching the specified key.
 133  *
 134  * Parameters:
 135  *      tree:   tree
 136  *      key:    key to delete
 137  *
 138  * Returns:
 139  *      RET_ERROR, RET_SUCCESS and RET_SPECIAL if the key not found.
 140  */
 141 static int
 142 bt_bdelete(t, key)
 143         BTREE *t;
 144         const DBT *key;
 145 {
 146         EPG *e, save;
 147         PAGE *h;
 148         pgno_t cpgno, pg;
 149         indx_t cindex;
 150         int deleted, dirty1, dirty2, exact;
 151
 152         /* Find any matching record; __bt_search pins the page. */
 153         if ((e = __bt_search(t, key, &exact)) == NULL)
 154                 return (RET_ERROR);
 155         if (!exact) {
 156                 mpool_put(t->bt_mp, e->page, 0);
 157                 return (RET_SPECIAL);
 158         }
 159
 160         /*
 161          * Delete forward, then delete backward, from the found key.  The
 162          * ordering is so that the deletions don't mess up the page refs.
 163          * The first loop deletes the key from the original page, the second
 164          * unpins the original page.  In the first loop, dirty1 is set if
 165          * the original page is modified, and dirty2 is set if any subsequent
 166          * pages are modified.  In the second loop, dirty1 starts off set if
 167          * the original page has been modified, and is set if any subsequent
 168          * pages are modified.
 169          *
 170          * If find the key referenced by the cursor, don't delete it, just
 171          * flag it for future deletion.  The cursor page number is P_INVALID
 172          * unless the sequential scan is initialized, so no reason to check.
 173          * A special case is when the already deleted cursor record was the
 174          * only record found.  If so, then the delete opertion fails as no
 175          * records were deleted.
 176          *
 177          * Cycle in place in the current page until the current record doesn't
 178          * match the key or the page is empty.  If the latter, walk forward,
 179          * skipping empty pages and repeating until a record doesn't match
 180          * the key or the end of the tree is reached.
 181          */
 182         cpgno = t->bt_bcursor.pgno;
 183         cindex = t->bt_bcursor.index;
 184         save = *e;
 185         dirty1 = 0;
 186         for (h = e->page, deleted = 0;;) {
 187                 dirty2 = 0;
 188                 do {
 189                         if (h->pgno == cpgno && e->index == cindex) {
 190                                 if (!ISSET(t, B_DELCRSR)) {
 191                                         SET(t, B_DELCRSR);
 192                                         deleted = 1;
 193                                 }
 194                                 ++e->index;
 195                         } else {
 196                                 if (__bt_dleaf(t, h, e->index)) {
 197                                         if (h->pgno != save.page->pgno)
 198                                                 mpool_put(t->bt_mp, h, dirty2);
 199                                         mpool_put(t->bt_mp, save.page, dirty1);
 200                                         return (RET_ERROR);
 201                                 }
 202                                 if (h->pgno == save.page->pgno)
 203                                         dirty1 = MPOOL_DIRTY;
 204                                 else
 205                                         dirty2 = MPOOL_DIRTY;
 206                                 deleted = 1;
 207                         }
 208                 } while (e->index < NEXTINDEX(h) && __bt_cmp(t, key, e) == 0);
 209
 210                 /*
 211                  * Quit if didn't find a match, no next page, or first key on
 212                  * the next page doesn't match.  Don't unpin the original page
 213                  * unless an error occurs.
 214                  */
 215                 if (e->index < NEXTINDEX(h))
 216                         break;
 217                 for (;;) {
 218                         if ((pg = h->nextpg) == P_INVALID)
 219                                 goto done1;
 220                         if (h->pgno != save.page->pgno)
 221                                 mpool_put(t->bt_mp, h, dirty2);
 222                         if ((h = mpool_get(t->bt_mp, pg, 0)) == NULL) {
 223                                 mpool_put(t->bt_mp, save.page, dirty1);
 224                                 return (RET_ERROR);
 225                         }
 226                         if (NEXTINDEX(h) != 0) {
 227                                 e->page = h;
 228                                 e->index = 0;
 229                                 break;
 230                         }
 231                 }
 232
 233                 if (__bt_cmp(t, key, e) != 0)
 234                         break;
 235         }
 236
 237         /*
 238          * Reach here with the original page and the last page referenced
 239          * pinned (they may be the same).  Release it if not the original.
 240          */
 241 done1:  if (h->pgno != save.page->pgno)
 242                 mpool_put(t->bt_mp, h, dirty2);
 243
 244         /*
 245          * Walk backwards from the record previous to the record returned by
 246          * __bt_search, skipping empty pages, until a record doesn't match
 247          * the key or reach the beginning of the tree.
 248          */
 249         *e = save;
 250         for (;;) {
 251                 if (e->index)
 252                         --e->index;
 253                 for (h = e->page; e->index; --e->index) {
 254                         if (__bt_cmp(t, key, e) != 0)
 255                                 goto done2;
 256                         if (h->pgno == cpgno && e->index == cindex) {
 257                                 if (!ISSET(t, B_DELCRSR)) {
 258                                         SET(t, B_DELCRSR);
 259                                         deleted = 1;
 260                                 }
 261                         } else {
 262                                 if (__bt_dleaf(t, h, e->index) == RET_ERROR) {
 263                                         mpool_put(t->bt_mp, h, dirty1);
 264                                         return (RET_ERROR);
 265                                 }
 266                                 if (h->pgno == save.page->pgno)
 267                                         dirty1 = MPOOL_DIRTY;
 268                                 deleted = 1;
 269                         }
 270                 }
 271
 272                 if ((pg = h->prevpg) == P_INVALID)
 273                         goto done2;
 274                 mpool_put(t->bt_mp, h, dirty1);
 275                 dirty1 = 0;
 276                 if ((e->page = mpool_get(t->bt_mp, pg, 0)) == NULL)
 277                         return (RET_ERROR);
 278                 e->index = NEXTINDEX(e->page);
 279         }
 280
 281         /*
 282          * Reach here with the last page that was looked at pinned.  Release
 283          * it.
 284          */
 285 done2:  mpool_put(t->bt_mp, h, dirty1);
 286         return (deleted ? RET_SUCCESS : RET_SPECIAL);
 287 }
 288
 289 /*
 290  * __BT_DLEAF -- Delete a single record from a leaf page.
 291  *
 292  * Parameters:
 293  *      t:      tree
 294  *      index:  index on current page to delete
 295  *
 296  * Returns:
 297  *      RET_SUCCESS, RET_ERROR.
 298  */
 299 int
 300 __bt_dleaf(t, h, index)
 301         BTREE *t;
 302         PAGE *h;
 303         indx_t index;
 304 {
 305         register BLEAF *bl;
 306         register indx_t cnt, *ip, offset;
 307         register size_t nbytes;
 308         char *from;
 309         void *to;
 310
 311         /*
 312          * Delete a record from a btree leaf page.  Internal records are never
 313          * deleted from internal pages, regardless of the records that caused
 314          * them to be added being deleted.  Pages made empty by deletion are
 315          * not reclaimed.  They are, however, made available for reuse.
 316          *
 317          * Pack the remaining entries at the end of the page, shift the indices
 318          * down, overwriting the deleted record and its index.  If the record
 319          * uses overflow pages, make them available for reuse.
 320          */
 321         to = bl = GETBLEAF(h, index);
 322         if (bl->flags & P_BIGKEY && __ovfl_delete(t, bl->bytes) == RET_ERROR)
 323                 return (RET_ERROR);
 324         if (bl->flags & P_BIGDATA &&
 325             __ovfl_delete(t, bl->bytes + bl->ksize) == RET_ERROR)
 326                 return (RET_ERROR);
 327         nbytes = NBLEAF(bl);
 328
 329         /*
 330          * Compress the key/data pairs.  Compress and adjust the [BR]LEAF
 331          * offsets.  Reset the headers.
 332          */
 333         from = (char *)h + h->upper;
 334         memmove(from + nbytes, from, (char *)to - from);
 335         h->upper += nbytes;
 336
 337         offset = h->linp[index];
 338         for (cnt = index, ip = &h->linp[0]; cnt--; ++ip)
 339                 if (ip[0] < offset)
 340                         ip[0] += nbytes;
 341         for (cnt = NEXTINDEX(h) - index; --cnt; ++ip)
 342                 ip[0] = ip[1] < offset ? ip[1] + nbytes : ip[1];
 343         h->lower -= sizeof(indx_t);
 344         return (RET_SUCCESS);
 345 }