xref: /sqlite-3.40.0/src/btreeInt.h (revision d5578433)
1 /*
2 ** 2004 April 6
3 **
4 ** The author disclaims copyright to this source code.  In place of
5 ** a legal notice, here is a blessing:
6 **
7 **    May you do good and not evil.
8 **    May you find forgiveness for yourself and forgive others.
9 **    May you share freely, never taking more than you give.
10 **
11 *************************************************************************
12 ** This file implements a external (disk-based) database using BTrees.
13 ** For a detailed discussion of BTrees, refer to
14 **
15 **     Donald E. Knuth, THE ART OF COMPUTER PROGRAMMING, Volume 3:
16 **     "Sorting And Searching", pages 473-480. Addison-Wesley
17 **     Publishing Company, Reading, Massachusetts.
18 **
19 ** The basic idea is that each page of the file contains N database
20 ** entries and N+1 pointers to subpages.
21 **
22 **   ----------------------------------------------------------------
23 **   |  Ptr(0) | Key(0) | Ptr(1) | Key(1) | ... | Key(N-1) | Ptr(N) |
24 **   ----------------------------------------------------------------
25 **
26 ** All of the keys on the page that Ptr(0) points to have values less
27 ** than Key(0).  All of the keys on page Ptr(1) and its subpages have
28 ** values greater than Key(0) and less than Key(1).  All of the keys
29 ** on Ptr(N) and its subpages have values greater than Key(N-1).  And
30 ** so forth.
31 **
32 ** Finding a particular key requires reading O(log(M)) pages from the
33 ** disk where M is the number of entries in the tree.
34 **
35 ** In this implementation, a single file can hold one or more separate
36 ** BTrees.  Each BTree is identified by the index of its root page.  The
37 ** key and data for any entry are combined to form the "payload".  A
38 ** fixed amount of payload can be carried directly on the database
39 ** page.  If the payload is larger than the preset amount then surplus
40 ** bytes are stored on overflow pages.  The payload for an entry
41 ** and the preceding pointer are combined to form a "Cell".  Each
42 ** page has a small header which contains the Ptr(N) pointer and other
43 ** information such as the size of key and data.
44 **
45 ** FORMAT DETAILS
46 **
47 ** The file is divided into pages.  The first page is called page 1,
48 ** the second is page 2, and so forth.  A page number of zero indicates
49 ** "no such page".  The page size can be any power of 2 between 512 and 65536.
50 ** Each page can be either a btree page, a freelist page, an overflow
51 ** page, or a pointer-map page.
52 **
53 ** The first page is always a btree page.  The first 100 bytes of the first
54 ** page contain a special header (the "file header") that describes the file.
55 ** The format of the file header is as follows:
56 **
57 **   OFFSET   SIZE    DESCRIPTION
58 **      0      16     Header string: "SQLite format 3\000"
59 **     16       2     Page size in bytes.
60 **     18       1     File format write version
61 **     19       1     File format read version
62 **     20       1     Bytes of unused space at the end of each page
63 **     21       1     Max embedded payload fraction
64 **     22       1     Min embedded payload fraction
65 **     23       1     Min leaf payload fraction
66 **     24       4     File change counter
67 **     28       4     Reserved for future use
68 **     32       4     First freelist page
69 **     36       4     Number of freelist pages in the file
70 **     40      60     15 4-byte meta values passed to higher layers
71 **
72 **     40       4     Schema cookie
73 **     44       4     File format of schema layer
74 **     48       4     Size of page cache
75 **     52       4     Largest root-page (auto/incr_vacuum)
76 **     56       4     1=UTF-8 2=UTF16le 3=UTF16be
77 **     60       4     User version
78 **     64       4     Incremental vacuum mode
79 **     68       4     unused
80 **     72       4     unused
81 **     76       4     unused
82 **
83 ** All of the integer values are big-endian (most significant byte first).
84 **
85 ** The file change counter is incremented when the database is changed
86 ** This counter allows other processes to know when the file has changed
87 ** and thus when they need to flush their cache.
88 **
89 ** The max embedded payload fraction is the amount of the total usable
90 ** space in a page that can be consumed by a single cell for standard
91 ** B-tree (non-LEAFDATA) tables.  A value of 255 means 100%.  The default
92 ** is to limit the maximum cell size so that at least 4 cells will fit
93 ** on one page.  Thus the default max embedded payload fraction is 64.
94 **
95 ** If the payload for a cell is larger than the max payload, then extra
96 ** payload is spilled to overflow pages.  Once an overflow page is allocated,
97 ** as many bytes as possible are moved into the overflow pages without letting
98 ** the cell size drop below the min embedded payload fraction.
99 **
100 ** The min leaf payload fraction is like the min embedded payload fraction
101 ** except that it applies to leaf nodes in a LEAFDATA tree.  The maximum
102 ** payload fraction for a LEAFDATA tree is always 100% (or 255) and it
103 ** not specified in the header.
104 **
105 ** Each btree pages is divided into three sections:  The header, the
106 ** cell pointer array, and the cell content area.  Page 1 also has a 100-byte
107 ** file header that occurs before the page header.
108 **
109 **      |----------------|
110 **      | file header    |   100 bytes.  Page 1 only.
111 **      |----------------|
112 **      | page header    |   8 bytes for leaves.  12 bytes for interior nodes
113 **      |----------------|
114 **      | cell pointer   |   |  2 bytes per cell.  Sorted order.
115 **      | array          |   |  Grows downward
116 **      |                |   v
117 **      |----------------|
118 **      | unallocated    |
119 **      | space          |
120 **      |----------------|   ^  Grows upwards
121 **      | cell content   |   |  Arbitrary order interspersed with freeblocks.
122 **      | area           |   |  and free space fragments.
123 **      |----------------|
124 **
125 ** The page headers looks like this:
126 **
127 **   OFFSET   SIZE     DESCRIPTION
128 **      0       1      Flags. 1: intkey, 2: zerodata, 4: leafdata, 8: leaf
129 **      1       2      byte offset to the first freeblock
130 **      3       2      number of cells on this page
131 **      5       2      first byte of the cell content area
132 **      7       1      number of fragmented free bytes
133 **      8       4      Right child (the Ptr(N) value).  Omitted on leaves.
134 **
135 ** The flags define the format of this btree page.  The leaf flag means that
136 ** this page has no children.  The zerodata flag means that this page carries
137 ** only keys and no data.  The intkey flag means that the key is a integer
138 ** which is stored in the key size entry of the cell header rather than in
139 ** the payload area.
140 **
141 ** The cell pointer array begins on the first byte after the page header.
142 ** The cell pointer array contains zero or more 2-byte numbers which are
143 ** offsets from the beginning of the page to the cell content in the cell
144 ** content area.  The cell pointers occur in sorted order.  The system strives
145 ** to keep free space after the last cell pointer so that new cells can
146 ** be easily added without having to defragment the page.
147 **
148 ** Cell content is stored at the very end of the page and grows toward the
149 ** beginning of the page.
150 **
151 ** Unused space within the cell content area is collected into a linked list of
152 ** freeblocks.  Each freeblock is at least 4 bytes in size.  The byte offset
153 ** to the first freeblock is given in the header.  Freeblocks occur in
154 ** increasing order.  Because a freeblock must be at least 4 bytes in size,
155 ** any group of 3 or fewer unused bytes in the cell content area cannot
156 ** exist on the freeblock chain.  A group of 3 or fewer free bytes is called
157 ** a fragment.  The total number of bytes in all fragments is recorded.
158 ** in the page header at offset 7.
159 **
160 **    SIZE    DESCRIPTION
161 **      2     Byte offset of the next freeblock
162 **      2     Bytes in this freeblock
163 **
164 ** Cells are of variable length.  Cells are stored in the cell content area at
165 ** the end of the page.  Pointers to the cells are in the cell pointer array
166 ** that immediately follows the page header.  Cells is not necessarily
167 ** contiguous or in order, but cell pointers are contiguous and in order.
168 **
169 ** Cell content makes use of variable length integers.  A variable
170 ** length integer is 1 to 9 bytes where the lower 7 bits of each
171 ** byte are used.  The integer consists of all bytes that have bit 8 set and
172 ** the first byte with bit 8 clear.  The most significant byte of the integer
173 ** appears first.  A variable-length integer may not be more than 9 bytes long.
174 ** As a special case, all 8 bytes of the 9th byte are used as data.  This
175 ** allows a 64-bit integer to be encoded in 9 bytes.
176 **
177 **    0x00                      becomes  0x00000000
178 **    0x7f                      becomes  0x0000007f
179 **    0x81 0x00                 becomes  0x00000080
180 **    0x82 0x00                 becomes  0x00000100
181 **    0x80 0x7f                 becomes  0x0000007f
182 **    0x8a 0x91 0xd1 0xac 0x78  becomes  0x12345678
183 **    0x81 0x81 0x81 0x81 0x01  becomes  0x10204081
184 **
185 ** Variable length integers are used for rowids and to hold the number of
186 ** bytes of key and data in a btree cell.
187 **
188 ** The content of a cell looks like this:
189 **
190 **    SIZE    DESCRIPTION
191 **      4     Page number of the left child. Omitted if leaf flag is set.
192 **     var    Number of bytes of data. Omitted if the zerodata flag is set.
193 **     var    Number of bytes of key. Or the key itself if intkey flag is set.
194 **      *     Payload
195 **      4     First page of the overflow chain.  Omitted if no overflow
196 **
197 ** Overflow pages form a linked list.  Each page except the last is completely
198 ** filled with data (pagesize - 4 bytes).  The last page can have as little
199 ** as 1 byte of data.
200 **
201 **    SIZE    DESCRIPTION
202 **      4     Page number of next overflow page
203 **      *     Data
204 **
205 ** Freelist pages come in two subtypes: trunk pages and leaf pages.  The
206 ** file header points to the first in a linked list of trunk page.  Each trunk
207 ** page points to multiple leaf pages.  The content of a leaf page is
208 ** unspecified.  A trunk page looks like this:
209 **
210 **    SIZE    DESCRIPTION
211 **      4     Page number of next trunk page
212 **      4     Number of leaf pointers on this page
213 **      *     zero or more pages numbers of leaves
214 */
215 #include "sqliteInt.h"
216 
217 
218 /* The following value is the maximum cell size assuming a maximum page
219 ** size give above.
220 */
221 #define MX_CELL_SIZE(pBt)  ((int)(pBt->pageSize-8))
222 
223 /* The maximum number of cells on a single page of the database.  This
224 ** assumes a minimum cell size of 6 bytes  (4 bytes for the cell itself
225 ** plus 2 bytes for the index to the cell in the page header).  Such
226 ** small cells will be rare, but they are possible.
227 */
228 #define MX_CELL(pBt) ((pBt->pageSize-8)/6)
229 
230 /* Forward declarations */
231 typedef struct MemPage MemPage;
232 typedef struct BtLock BtLock;
233 
234 /*
235 ** This is a magic string that appears at the beginning of every
236 ** SQLite database in order to identify the file as a real database.
237 **
238 ** You can change this value at compile-time by specifying a
239 ** -DSQLITE_FILE_HEADER="..." on the compiler command-line.  The
240 ** header must be exactly 16 bytes including the zero-terminator so
241 ** the string itself should be 15 characters long.  If you change
242 ** the header, then your custom library will not be able to read
243 ** databases generated by the standard tools and the standard tools
244 ** will not be able to read databases created by your custom library.
245 */
246 #ifndef SQLITE_FILE_HEADER /* 123456789 123456 */
247 #  define SQLITE_FILE_HEADER "SQLite format 3"
248 #endif
249 
250 /*
251 ** Page type flags.  An ORed combination of these flags appear as the
252 ** first byte of on-disk image of every BTree page.
253 */
254 #define PTF_INTKEY    0x01
255 #define PTF_ZERODATA  0x02
256 #define PTF_LEAFDATA  0x04
257 #define PTF_LEAF      0x08
258 
259 /*
260 ** As each page of the file is loaded into memory, an instance of the following
261 ** structure is appended and initialized to zero.  This structure stores
262 ** information about the page that is decoded from the raw file page.
263 **
264 ** The pParent field points back to the parent page.  This allows us to
265 ** walk up the BTree from any leaf to the root.  Care must be taken to
266 ** unref() the parent page pointer when this page is no longer referenced.
267 ** The pageDestructor() routine handles that chore.
268 **
269 ** Access to all fields of this structure is controlled by the mutex
270 ** stored in MemPage.pBt->mutex.
271 */
272 struct MemPage {
273   u8 isInit;           /* True if previously initialized. MUST BE FIRST! */
274   u8 nOverflow;        /* Number of overflow cell bodies in aCell[] */
275   u8 intKey;           /* True if intkey flag is set */
276   u8 leaf;             /* True if leaf flag is set */
277   u8 hasData;          /* True if this page stores data */
278   u8 hdrOffset;        /* 100 for page 1.  0 otherwise */
279   u8 childPtrSize;     /* 0 if leaf==1.  4 if leaf==0 */
280   u8 max1bytePayload;  /* min(maxLocal,127) */
281   u16 maxLocal;        /* Copy of BtShared.maxLocal or BtShared.maxLeaf */
282   u16 minLocal;        /* Copy of BtShared.minLocal or BtShared.minLeaf */
283   u16 cellOffset;      /* Index in aData of first cell pointer */
284   u16 nFree;           /* Number of free bytes on the page */
285   u16 nCell;           /* Number of cells on this page, local and ovfl */
286   u16 maskPage;        /* Mask for page offset */
287   u16 aiOvfl[5];       /* Insert the i-th overflow cell before the aiOvfl-th
288                        ** non-overflow cell */
289   u8 *apOvfl[5];       /* Pointers to the body of overflow cells */
290   BtShared *pBt;       /* Pointer to BtShared that this page is part of */
291   u8 *aData;           /* Pointer to disk image of the page data */
292   u8 *aDataEnd;        /* One byte past the end of usable data */
293   u8 *aCellIdx;        /* The cell index area */
294   DbPage *pDbPage;     /* Pager page handle */
295   Pgno pgno;           /* Page number for this page */
296 };
297 
298 /*
299 ** The in-memory image of a disk page has the auxiliary information appended
300 ** to the end.  EXTRA_SIZE is the number of bytes of space needed to hold
301 ** that extra information.
302 */
303 #define EXTRA_SIZE sizeof(MemPage)
304 
305 /*
306 ** A linked list of the following structures is stored at BtShared.pLock.
307 ** Locks are added (or upgraded from READ_LOCK to WRITE_LOCK) when a cursor
308 ** is opened on the table with root page BtShared.iTable. Locks are removed
309 ** from this list when a transaction is committed or rolled back, or when
310 ** a btree handle is closed.
311 */
312 struct BtLock {
313   Btree *pBtree;        /* Btree handle holding this lock */
314   Pgno iTable;          /* Root page of table */
315   u8 eLock;             /* READ_LOCK or WRITE_LOCK */
316   BtLock *pNext;        /* Next in BtShared.pLock list */
317 };
318 
319 /* Candidate values for BtLock.eLock */
320 #define READ_LOCK     1
321 #define WRITE_LOCK    2
322 
323 /* A Btree handle
324 **
325 ** A database connection contains a pointer to an instance of
326 ** this object for every database file that it has open.  This structure
327 ** is opaque to the database connection.  The database connection cannot
328 ** see the internals of this structure and only deals with pointers to
329 ** this structure.
330 **
331 ** For some database files, the same underlying database cache might be
332 ** shared between multiple connections.  In that case, each connection
333 ** has it own instance of this object.  But each instance of this object
334 ** points to the same BtShared object.  The database cache and the
335 ** schema associated with the database file are all contained within
336 ** the BtShared object.
337 **
338 ** All fields in this structure are accessed under sqlite3.mutex.
339 ** The pBt pointer itself may not be changed while there exists cursors
340 ** in the referenced BtShared that point back to this Btree since those
341 ** cursors have to go through this Btree to find their BtShared and
342 ** they often do so without holding sqlite3.mutex.
343 */
344 struct Btree {
345   sqlite3 *db;       /* The database connection holding this btree */
346   BtShared *pBt;     /* Sharable content of this btree */
347   u8 inTrans;        /* TRANS_NONE, TRANS_READ or TRANS_WRITE */
348   u8 sharable;       /* True if we can share pBt with another db */
349   u8 locked;         /* True if db currently has pBt locked */
350   int wantToLock;    /* Number of nested calls to sqlite3BtreeEnter() */
351   int nBackup;       /* Number of backup operations reading this btree */
352   Btree *pNext;      /* List of other sharable Btrees from the same db */
353   Btree *pPrev;      /* Back pointer of the same list */
354 #ifndef SQLITE_OMIT_SHARED_CACHE
355   BtLock lock;       /* Object used to lock page 1 */
356 #endif
357 };
358 
359 /*
360 ** Btree.inTrans may take one of the following values.
361 **
362 ** If the shared-data extension is enabled, there may be multiple users
363 ** of the Btree structure. At most one of these may open a write transaction,
364 ** but any number may have active read transactions.
365 */
366 #define TRANS_NONE  0
367 #define TRANS_READ  1
368 #define TRANS_WRITE 2
369 
370 /*
371 ** An instance of this object represents a single database file.
372 **
373 ** A single database file can be in use at the same time by two
374 ** or more database connections.  When two or more connections are
375 ** sharing the same database file, each connection has it own
376 ** private Btree object for the file and each of those Btrees points
377 ** to this one BtShared object.  BtShared.nRef is the number of
378 ** connections currently sharing this database file.
379 **
380 ** Fields in this structure are accessed under the BtShared.mutex
381 ** mutex, except for nRef and pNext which are accessed under the
382 ** global SQLITE_MUTEX_STATIC_MASTER mutex.  The pPager field
383 ** may not be modified once it is initially set as long as nRef>0.
384 ** The pSchema field may be set once under BtShared.mutex and
385 ** thereafter is unchanged as long as nRef>0.
386 **
387 ** isPending:
388 **
389 **   If a BtShared client fails to obtain a write-lock on a database
390 **   table (because there exists one or more read-locks on the table),
391 **   the shared-cache enters 'pending-lock' state and isPending is
392 **   set to true.
393 **
394 **   The shared-cache leaves the 'pending lock' state when either of
395 **   the following occur:
396 **
397 **     1) The current writer (BtShared.pWriter) concludes its transaction, OR
398 **     2) The number of locks held by other connections drops to zero.
399 **
400 **   while in the 'pending-lock' state, no connection may start a new
401 **   transaction.
402 **
403 **   This feature is included to help prevent writer-starvation.
404 */
405 struct BtShared {
406   Pager *pPager;        /* The page cache */
407   sqlite3 *db;          /* Database connection currently using this Btree */
408   BtCursor *pCursor;    /* A list of all open cursors */
409   MemPage *pPage1;      /* First page of the database */
410   u8 openFlags;         /* Flags to sqlite3BtreeOpen() */
411 #ifndef SQLITE_OMIT_AUTOVACUUM
412   u8 autoVacuum;        /* True if auto-vacuum is enabled */
413   u8 incrVacuum;        /* True if incr-vacuum is enabled */
414 #endif
415   u8 inTransaction;     /* Transaction state */
416   u8 max1bytePayload;   /* Maximum first byte of cell for a 1-byte payload */
417   u16 btsFlags;         /* Boolean parameters.  See BTS_* macros below */
418   u16 maxLocal;         /* Maximum local payload in non-LEAFDATA tables */
419   u16 minLocal;         /* Minimum local payload in non-LEAFDATA tables */
420   u16 maxLeaf;          /* Maximum local payload in a LEAFDATA table */
421   u16 minLeaf;          /* Minimum local payload in a LEAFDATA table */
422   u32 pageSize;         /* Total number of bytes on a page */
423   u32 usableSize;       /* Number of usable bytes on each page */
424   int nTransaction;     /* Number of open transactions (read + write) */
425   u32 nPage;            /* Number of pages in the database */
426   void *pSchema;        /* Pointer to space allocated by sqlite3BtreeSchema() */
427   void (*xFreeSchema)(void*);  /* Destructor for BtShared.pSchema */
428   sqlite3_mutex *mutex; /* Non-recursive mutex required to access this object */
429   Bitvec *pHasContent;  /* Set of pages moved to free-list this transaction */
430 #ifndef SQLITE_OMIT_SHARED_CACHE
431   int nRef;             /* Number of references to this structure */
432   BtShared *pNext;      /* Next on a list of sharable BtShared structs */
433   BtLock *pLock;        /* List of locks held on this shared-btree struct */
434   Btree *pWriter;       /* Btree with currently open write transaction */
435 #endif
436   u8 *pTmpSpace;        /* BtShared.pageSize bytes of space for tmp use */
437 };
438 
439 /*
440 ** Allowed values for BtShared.btsFlags
441 */
442 #define BTS_READ_ONLY        0x0001   /* Underlying file is readonly */
443 #define BTS_PAGESIZE_FIXED   0x0002   /* Page size can no longer be changed */
444 #define BTS_SECURE_DELETE    0x0004   /* PRAGMA secure_delete is enabled */
445 #define BTS_INITIALLY_EMPTY  0x0008   /* Database was empty at trans start */
446 #define BTS_NO_WAL           0x0010   /* Do not open write-ahead-log files */
447 #define BTS_EXCLUSIVE        0x0020   /* pWriter has an exclusive lock */
448 #define BTS_PENDING          0x0040   /* Waiting for read-locks to clear */
449 
450 /*
451 ** An instance of the following structure is used to hold information
452 ** about a cell.  The parseCellPtr() function fills in this structure
453 ** based on information extract from the raw disk page.
454 */
455 typedef struct CellInfo CellInfo;
456 struct CellInfo {
457   i64 nKey;      /* The key for INTKEY tables, or number of bytes in key */
458   u8 *pCell;     /* Pointer to the start of cell content */
459   u32 nData;     /* Number of bytes of data */
460   u32 nPayload;  /* Total amount of payload */
461   u16 nHeader;   /* Size of the cell content header in bytes */
462   u16 nLocal;    /* Amount of payload held locally */
463   u16 iOverflow; /* Offset to overflow page number.  Zero if no overflow */
464   u16 nSize;     /* Size of the cell content on the main b-tree page */
465 };
466 
467 /*
468 ** Maximum depth of an SQLite B-Tree structure. Any B-Tree deeper than
469 ** this will be declared corrupt. This value is calculated based on a
470 ** maximum database size of 2^31 pages a minimum fanout of 2 for a
471 ** root-node and 3 for all other internal nodes.
472 **
473 ** If a tree that appears to be taller than this is encountered, it is
474 ** assumed that the database is corrupt.
475 */
476 #define BTCURSOR_MAX_DEPTH 20
477 
478 /*
479 ** A cursor is a pointer to a particular entry within a particular
480 ** b-tree within a database file.
481 **
482 ** The entry is identified by its MemPage and the index in
483 ** MemPage.aCell[] of the entry.
484 **
485 ** A single database file can be shared by two more database connections,
486 ** but cursors cannot be shared.  Each cursor is associated with a
487 ** particular database connection identified BtCursor.pBtree.db.
488 **
489 ** Fields in this structure are accessed under the BtShared.mutex
490 ** found at self->pBt->mutex.
491 */
492 struct BtCursor {
493   Btree *pBtree;            /* The Btree to which this cursor belongs */
494   BtShared *pBt;            /* The BtShared this cursor points to */
495   BtCursor *pNext, *pPrev;  /* Forms a linked list of all cursors */
496   struct KeyInfo *pKeyInfo; /* Argument passed to comparison function */
497 #ifndef SQLITE_OMIT_INCRBLOB
498   Pgno *aOverflow;          /* Cache of overflow page locations */
499 #endif
500   Pgno pgnoRoot;            /* The root page of this tree */
501   sqlite3_int64 cachedRowid; /* Next rowid cache.  0 means not valid */
502   CellInfo info;            /* A parse of the cell we are pointing at */
503   i64 nKey;        /* Size of pKey, or last integer key */
504   void *pKey;      /* Saved key that was cursor's last known position */
505   int skipNext;    /* Prev() is noop if negative. Next() is noop if positive */
506   u8 wrFlag;                /* True if writable */
507   u8 atLast;                /* Cursor pointing to the last entry */
508   u8 validNKey;             /* True if info.nKey is valid */
509   u8 eState;                /* One of the CURSOR_XXX constants (see below) */
510 #ifndef SQLITE_OMIT_INCRBLOB
511   u8 isIncrblobHandle;      /* True if this cursor is an incr. io handle */
512 #endif
513   u8 hints;                             /* As configured by CursorSetHints() */
514   i16 iPage;                            /* Index of current page in apPage */
515   u16 aiIdx[BTCURSOR_MAX_DEPTH];        /* Current index in apPage[i] */
516   MemPage *apPage[BTCURSOR_MAX_DEPTH];  /* Pages from root to current page */
517 };
518 
519 /*
520 ** Potential values for BtCursor.eState.
521 **
522 ** CURSOR_VALID:
523 **   Cursor points to a valid entry. getPayload() etc. may be called.
524 **
525 ** CURSOR_INVALID:
526 **   Cursor does not point to a valid entry. This can happen (for example)
527 **   because the table is empty or because BtreeCursorFirst() has not been
528 **   called.
529 **
530 ** CURSOR_REQUIRESEEK:
531 **   The table that this cursor was opened on still exists, but has been
532 **   modified since the cursor was last used. The cursor position is saved
533 **   in variables BtCursor.pKey and BtCursor.nKey. When a cursor is in
534 **   this state, restoreCursorPosition() can be called to attempt to
535 **   seek the cursor to the saved position.
536 **
537 ** CURSOR_FAULT:
538 **   A unrecoverable error (an I/O error or a malloc failure) has occurred
539 **   on a different connection that shares the BtShared cache with this
540 **   cursor.  The error has left the cache in an inconsistent state.
541 **   Do nothing else with this cursor.  Any attempt to use the cursor
542 **   should return the error code stored in BtCursor.skip
543 */
544 #define CURSOR_INVALID           0
545 #define CURSOR_VALID             1
546 #define CURSOR_REQUIRESEEK       2
547 #define CURSOR_FAULT             3
548 
549 /*
550 ** The database page the PENDING_BYTE occupies. This page is never used.
551 */
552 # define PENDING_BYTE_PAGE(pBt) PAGER_MJ_PGNO(pBt)
553 
554 /*
555 ** These macros define the location of the pointer-map entry for a
556 ** database page. The first argument to each is the number of usable
557 ** bytes on each page of the database (often 1024). The second is the
558 ** page number to look up in the pointer map.
559 **
560 ** PTRMAP_PAGENO returns the database page number of the pointer-map
561 ** page that stores the required pointer. PTRMAP_PTROFFSET returns
562 ** the offset of the requested map entry.
563 **
564 ** If the pgno argument passed to PTRMAP_PAGENO is a pointer-map page,
565 ** then pgno is returned. So (pgno==PTRMAP_PAGENO(pgsz, pgno)) can be
566 ** used to test if pgno is a pointer-map page. PTRMAP_ISPAGE implements
567 ** this test.
568 */
569 #define PTRMAP_PAGENO(pBt, pgno) ptrmapPageno(pBt, pgno)
570 #define PTRMAP_PTROFFSET(pgptrmap, pgno) (5*(pgno-pgptrmap-1))
571 #define PTRMAP_ISPAGE(pBt, pgno) (PTRMAP_PAGENO((pBt),(pgno))==(pgno))
572 
573 /*
574 ** The pointer map is a lookup table that identifies the parent page for
575 ** each child page in the database file.  The parent page is the page that
576 ** contains a pointer to the child.  Every page in the database contains
577 ** 0 or 1 parent pages.  (In this context 'database page' refers
578 ** to any page that is not part of the pointer map itself.)  Each pointer map
579 ** entry consists of a single byte 'type' and a 4 byte parent page number.
580 ** The PTRMAP_XXX identifiers below are the valid types.
581 **
582 ** The purpose of the pointer map is to facility moving pages from one
583 ** position in the file to another as part of autovacuum.  When a page
584 ** is moved, the pointer in its parent must be updated to point to the
585 ** new location.  The pointer map is used to locate the parent page quickly.
586 **
587 ** PTRMAP_ROOTPAGE: The database page is a root-page. The page-number is not
588 **                  used in this case.
589 **
590 ** PTRMAP_FREEPAGE: The database page is an unused (free) page. The page-number
591 **                  is not used in this case.
592 **
593 ** PTRMAP_OVERFLOW1: The database page is the first page in a list of
594 **                   overflow pages. The page number identifies the page that
595 **                   contains the cell with a pointer to this overflow page.
596 **
597 ** PTRMAP_OVERFLOW2: The database page is the second or later page in a list of
598 **                   overflow pages. The page-number identifies the previous
599 **                   page in the overflow page list.
600 **
601 ** PTRMAP_BTREE: The database page is a non-root btree page. The page number
602 **               identifies the parent page in the btree.
603 */
604 #define PTRMAP_ROOTPAGE 1
605 #define PTRMAP_FREEPAGE 2
606 #define PTRMAP_OVERFLOW1 3
607 #define PTRMAP_OVERFLOW2 4
608 #define PTRMAP_BTREE 5
609 
610 /* A bunch of assert() statements to check the transaction state variables
611 ** of handle p (type Btree*) are internally consistent.
612 */
613 #define btreeIntegrity(p) \
614   assert( p->pBt->inTransaction!=TRANS_NONE || p->pBt->nTransaction==0 ); \
615   assert( p->pBt->inTransaction>=p->inTrans );
616 
617 
618 /*
619 ** The ISAUTOVACUUM macro is used within balance_nonroot() to determine
620 ** if the database supports auto-vacuum or not. Because it is used
621 ** within an expression that is an argument to another macro
622 ** (sqliteMallocRaw), it is not possible to use conditional compilation.
623 ** So, this macro is defined instead.
624 */
625 #ifndef SQLITE_OMIT_AUTOVACUUM
626 #define ISAUTOVACUUM (pBt->autoVacuum)
627 #else
628 #define ISAUTOVACUUM 0
629 #endif
630 
631 
632 /*
633 ** This structure is passed around through all the sanity checking routines
634 ** in order to keep track of some global state information.
635 **
636 ** The aRef[] array is allocated so that there is 1 bit for each page in
637 ** the database. As the integrity-check proceeds, for each page used in
638 ** the database the corresponding bit is set. This allows integrity-check to
639 ** detect pages that are used twice and orphaned pages (both of which
640 ** indicate corruption).
641 */
642 typedef struct IntegrityCk IntegrityCk;
643 struct IntegrityCk {
644   BtShared *pBt;    /* The tree being checked out */
645   Pager *pPager;    /* The associated pager.  Also accessible by pBt->pPager */
646   u8 *aPgRef;       /* 1 bit per page in the db (see above) */
647   Pgno nPage;       /* Number of pages in the database */
648   int mxErr;        /* Stop accumulating errors when this reaches zero */
649   int nErr;         /* Number of messages written to zErrMsg so far */
650   int mallocFailed; /* A memory allocation error has occurred */
651   StrAccum errMsg;  /* Accumulate the error message text here */
652 };
653 
654 /*
655 ** Routines to read or write a two- and four-byte big-endian integer values.
656 */
657 #define get2byte(x)   ((x)[0]<<8 | (x)[1])
658 #define put2byte(p,v) ((p)[0] = (u8)((v)>>8), (p)[1] = (u8)(v))
659 #define get4byte sqlite3Get4byte
660 #define put4byte sqlite3Put4byte
661