SQLite

/*
** 2004 April 6
**
** The author disclaims copyright to this source code.  In place of
** a legal notice, here is a blessing:
**
**    May you do good and not evil.
**    May you find forgiveness for yourself and forgive others.
**    May you share freely, never taking more than you give.
**
*************************************************************************
** $Id: btree.c,v 1.214 2004/11/08 07:13:14 danielk1977 Exp $
**
** This file implements a external (disk-based) database using BTrees.
** For a detailed discussion of BTrees, refer to
**
**     Donald E. Knuth, THE ART OF COMPUTER PROGRAMMING, Volume 3:
**     "Sorting And Searching", pages 473-480. Addison-Wesley
**     Publishing Company, Reading, Massachusetts.

** be defined locally, but now we use the varint routines in the util.c
** file.
*/
#define getVarint    sqlite3GetVarint
#define getVarint32  sqlite3GetVarint32
#define putVarint    sqlite3PutVarint

/* The database page the PENDING_BYTE occupies. This page is never used.
** TODO: This macro is very similary to PAGER_MJ_PGNO() in pager.c. They
** should possibly be consolidated (presumably in pager.h).
*/
#define PENDING_BYTE_PAGE(pBt) ((PENDING_BYTE/(pBt)->pageSize)+1)

#ifndef SQLITE_OMIT_AUTOVACUUM
/*
** These two macros define the location of the pointer-map entry for a 
** database page. The first argument to each is the page size used 
** by the database (often 1024). The second is the page number to look
** up in the pointer map.
**
** PTRMAP_PAGENO returns the database page number of the pointer-map
** page that stores the required pointer. PTRMAP_PTROFFSET returns
** the offset of the requested map entry.
**
** If the pgno argument passed to PTRMAP_PAGENO is a pointer-map page,
** then pgno is returned. So (pgno==PTRMAP_PAGENO(pgsz, pgno)) can be
** used to test if pgno is a pointer-map page. PTRMAP_ISPAGE implements
** this test.
*/
#define PTRMAP_PAGENO(pgsz, pgno) (((pgno-2)/(pgsz/5+1))*(pgsz/5+1)+2)
#define PTRMAP_PTROFFSET(pgsz, pgno) (((pgno-2)%(pgsz/5+1)-1)*5)

#define PTRMAP_ISPAGE(pgsz, pgno) (PTRMAP_PAGENO(pgsz,pgno)==pgno)

/*
** The pointer map is a lookup table that contains an entry for each database
** page in the file except for page 1. In this context 'database page' refers
** to any page that is not part of the pointer map itself.  Each pointer map
** entry consists of a single byte 'type' and a 4 byte page number. The

*/
static int ptrmapPut(Btree *pBt, Pgno key, u8 eType, Pgno pgno){
  u8 *pPtrmap;    /* The pointer map page */
  Pgno iPtrmap;   /* The pointer map page number */
  int offset;     /* Offset in pointer map page */
  int rc;

  assert( key!=0 );
  iPtrmap = PTRMAP_PAGENO(pBt->pageSize, key);
  rc = sqlite3pager_get(pBt->pPager, iPtrmap, (void **)&pPtrmap);
  if( rc!=SQLITE_OK ){
    return rc;
  }
  offset = PTRMAP_PTROFFSET(pBt->pageSize, key);

    *nTrunc = 0;
    return SQLITE_OK;
  }

  origSize = sqlite3pager_pagecount(pPager);
  nPtrMap = (nFreeList-origSize+PTRMAP_PAGENO(pgsz, origSize)+pgsz/5)/(pgsz/5);
  finSize = origSize - nFreeList - nPtrMap;
  if( origSize>PENDING_BYTE_PAGE(pBt) && finSize<=PENDING_BYTE_PAGE(pBt) ){
    finSize--;
    if( PTRMAP_ISPAGE(pBt->pageSize, finSize) ){
      finSize--;
    }
  }
  TRACE(("AUTOVACUUM: Begin (db size %d->%d)\n", origSize, finSize));

  /* Variable 'finSize' will be the size of the file in pages after
  ** the auto-vacuum has completed (the current file size minus the number
  ** of pages on the free list). Loop through the pages that lie beyond
  ** this mark, and if they are not already on the free list, move them
  ** to a free page earlier in the file (somewhere before finSize).
  */
  for( iDbPage=finSize+1; iDbPage<=origSize; iDbPage++ ){
    /* If iDbPage is a pointer map page, or the pending-byte page, skip it. */
    if( PTRMAP_ISPAGE(pgsz, iDbPage) || iDbPage==PENDING_BYTE_PAGE(pBt) ){
      continue;
    }

    rc = ptrmapGet(pBt, iDbPage, &eType, &iPtrPage);
    if( rc!=SQLITE_OK ) goto autovacuum_out;
    assert( eType!=PTRMAP_ROOTPAGE );

    /* If iDbPage is free, do not swap it.  */


    if( eType==PTRMAP_FREEPAGE ){
      continue;
    }
    rc = getPage(pBt, iDbPage, &pDbMemPage);
    if( rc!=SQLITE_OK ) goto autovacuum_out;

    /* Find the next page in the free-list that is not already at the end 
    ** of the file. A page can be pulled off the free list using the 

#ifndef SQLITE_OMIT_AUTOVACUUM
    if( pBt->autoVacuum && PTRMAP_ISPAGE(pBt->pageSize, *pPgno) ){
      /* If *pPgno refers to a pointer-map page, allocate two new pages
      ** at the end of the file instead of one. The first allocated page
      ** becomes a new pointer-map page, the second is used by the caller.
      */
      TRACE(("ALLOCATE: %d from end of file (pointer-map page)\n", *pPgno));
      assert( *pPgno!=PENDING_BYTE_PAGE(pBt) );
      (*pPgno)++;
    }
#endif

    assert( *pPgno!=PENDING_BYTE_PAGE(pBt) );
    rc = getPage(pBt, *pPgno, ppPage);
    if( rc ) return rc;
    rc = sqlite3pager_write((*ppPage)->aData);
    TRACE(("ALLOCATE: %d from end of file\n", *pPgno));
  }

  assert( *pPgno!=PENDING_BYTE_PAGE(pBt) );
  return rc;
}

/*
** Add a page of the database file to the freelist.
**
** sqlite3pager_unref() is NOT called for pPage.

    ** root page of the new table should go. meta[3] is the largest root-page
    ** created so far, so the new root-page is (meta[3]+1).
    */
    rc = sqlite3BtreeGetMeta(pBt, 4, &pgnoRoot);
    if( rc!=SQLITE_OK ) return rc;
    pgnoRoot++;

    /* The new root-page may not be allocated on a pointer-map page, or the
    ** PENDING_BYTE page.
    */
    if( pgnoRoot==PTRMAP_PAGENO(pBt->pageSize, pgnoRoot) ||
        pgnoRoot==PENDING_BYTE_PAGE(pBt) ){
      pgnoRoot++;
    }
    assert( pgnoRoot>=3 );

    /* Allocate a page. The page that currently resides at pgnoRoot will
    ** be moved to the allocated page (unless the allocated page happens
    ** to reside at pgnoRoot).

        releasePage(pMove);
        if( rc!=SQLITE_OK ){
          return rc;
        }
        *piMoved = maxRootPgno;
      }

      /* Set the new 'max-root-page' value in the database header. This
      ** is the old value less one, less one more if that happens to
      ** be a root-page number, less one again if that is the
      ** PENDING_BYTE_PAGE.
      */
      maxRootPgno--;
      if( maxRootPgno==PENDING_BYTE_PAGE(pBt) ){
        maxRootPgno--;
      }
      if( maxRootPgno==PTRMAP_PAGENO(pBt->pageSize, maxRootPgno) ){
        maxRootPgno--;
      }
      assert( maxRootPgno!=PENDING_BYTE_PAGE(pBt) );

      rc = sqlite3BtreeUpdateMeta(pBt, 4, maxRootPgno);
    }else{
      rc = freePage(pPage);
      releasePage(pPage);
    }
#endif
  }else{

  assert( idx>=0 && idx<=15 );
  rc = sqlite3pager_get(pBt->pPager, 1, (void**)&pP1);
  if( rc ) return rc;
  *pMeta = get4byte(&pP1[36 + idx*4]);
  sqlite3pager_unref(pP1);

  /* If autovacuumed is disabled in this build but we are trying to 
  ** access an autovacuumed database, then make the database readonly. 
  */
#ifdef SQLITE_OMIT_AUTOVACUUM
  if( idx==4 && *pMeta>0 ) pBt->readOnly = 1;
#endif

  return SQLITE_OK;
}

** file format.  Depending on how it is changed, you might not notice
** the incompatibility right away, even running a full regression test.
** The default location of PENDING_BYTE is the first byte past the
** 1GB boundary.
**
*/
#define PENDING_BYTE      0x40000000  /* First byte past the 1GB boundary */
/* #define PENDING_BYTE     0x5400   // Page 22 - for testing */
#define RESERVED_BYTE     (PENDING_BYTE+1)
#define SHARED_FIRST      (PENDING_BYTE+2)
#define SHARED_SIZE       510


int sqlite3OsDelete(const char*);
int sqlite3OsFileExists(const char*);

** The pager is used to access a database disk file.  It implements
** atomic commit and rollback through the use of a journal file that
** is separate from the database file.  The pager also implements file
** locking to prevent two processes from writing the same database
** file simultaneously, or one process from reading the database while
** another is writing.
**
** @(#) $Id: pager.c,v 1.175 2004/11/08 07:13:14 danielk1977 Exp $
*/
#include "sqliteInt.h"
#include "os.h"
#include "pager.h"
#include <assert.h>
#include <string.h>

/*
** Macros for troubleshooting.  Normally turned off
*/
#if 0
#define TRACE1(X)       sqlite3DebugPrintf(X)
#define TRACE2(X,Y)     sqlite3DebugPrintf(X,Y)
#define TRACE3(X,Y,Z)   sqlite3DebugPrintf(X,Y,Z)
#define TRACE4(X,Y,Z,W) sqlite3DebugPrintf(X,Y,Z,W)
#define TRACE5(X,Y,Z,W,V) sqlite3DebugPrintf(X,Y,Z,W, V)
#else
#define TRACE1(X)
#define TRACE2(X,Y)
#define TRACE3(X,Y,Z)
#define TRACE4(X,Y,Z,W)
#define TRACE5(X,Y,Z,W,V)
#endif

/*
** The following two macros are used within the TRACEX() macros above
** to print out file-descriptors. They are required so that tracing
** can be turned on when using both the regular os_unix.c and os_test.c
** backends.
**
** PAGERID() takes a pointer to a Pager struct as it's argument. The
** associated file-descriptor is returned. FILEHANDLEID() takes an OsFile
** struct as it's argument.
*/
#ifdef OS_TEST
#define PAGERID(p) (p->fd->fd.h)
#define FILEHANDLEID(fd) (fd->fd.h)
#else
#define PAGERID(p) (p->fd.h)
#define FILEHANDLEID(fd) (fd.h)
#endif

/*
** Page number PAGER_MJ_PGNO is never used in an SQLite database (it is
** reserved for working around a windows/posix incompatibility). It is
** used in the journal to signify that the remainder of the journal file 
** is devoted to storing a master journal name - there are no more pages to
** roll back. See comments for function writeMasterJournal() for details.
*/
/* #define PAGER_MJ_PGNO(x) (PENDING_BYTE/((x)->pageSize)) */
#define PAGER_MJ_PGNO(x) ((PENDING_BYTE/((x)->pageSize))+1)

/*
** Enable reference count tracking (for debugging) here:
*/
#ifdef SQLITE_TEST
  int pager3_refinfo_enable = 0;
  static void pager_refinfo(PgHdr *p){

  }

sync_exit:
  return rc;
}

#ifndef SQLITE_OMIT_AUTOVACUUM

/*
** Move the page identified by pData to location pgno in the file. 
**
** There must be no references to the current page pgno. If current page
** pgno is not already in the rollback journal, it is not written there by
** by this routine. The same applies to the page pData refers to on entry to
** this routine.

  PgHdr *pPg = DATA_TO_PGHDR(pData);
  PgHdr *pPgOld; 
  int h;

  assert( !pPager->stmtInUse );
  assert( pPg->nRef>0 );

  TRACE5("MOVE %d page %d (needSync=%d) moves to %d\n", 
      PAGERID(pPager), pPg->pgno, pPg->needSync, pgno);

  /* Unlink pPg from it's hash-chain */
  unlinkHashChain(pPager, pPg);

  /* If the cache contains a page with page-number pgno, remove it
  ** from it's hash chain. Also, if the PgHdr.needSync was set for 
  ** page pgno before the 'move' operation, it needs to be retained 
  ** for the page moved there.
  */
  pPgOld = pager_lookup(pPager, pgno);
  if( pPgOld ){
    assert( pPgOld->nRef==0 );
    unlinkHashChain(pPager, pPgOld);
    pPgOld->dirty = 0;
    if( pPgOld->needSync ){