000001  /*
000002  ** 2003 April 6
000003  **
000004  ** The author disclaims copyright to this source code.  In place of
000005  ** a legal notice, here is a blessing:
000006  **
000007  **    May you do good and not evil.
000008  **    May you find forgiveness for yourself and forgive others.
000009  **    May you share freely, never taking more than you give.
000010  **
000011  *************************************************************************
000012  ** This file contains code used to implement the VACUUM command.
000013  **
000014  ** Most of the code in this file may be omitted by defining the
000015  ** SQLITE_OMIT_VACUUM macro.
000016  */
000017  #include "sqliteInt.h"
000018  #include "vdbeInt.h"
000019  
000020  #if !defined(SQLITE_OMIT_VACUUM) && !defined(SQLITE_OMIT_ATTACH)
000021  
000022  /*
000023  ** Execute zSql on database db.
000024  **
000025  ** If zSql returns rows, then each row will have exactly one
000026  ** column.  (This will only happen if zSql begins with "SELECT".)
000027  ** Take each row of result and call execSql() again recursively.
000028  **
000029  ** The execSqlF() routine does the same thing, except it accepts
000030  ** a format string as its third argument
000031  */
000032  static int execSql(sqlite3 *db, char **pzErrMsg, const char *zSql){
000033    sqlite3_stmt *pStmt;
000034    int rc;
000035  
000036    /* printf("SQL: [%s]\n", zSql); fflush(stdout); */
000037    rc = sqlite3_prepare_v2(db, zSql, -1, &pStmt, 0);
000038    if( rc!=SQLITE_OK ) return rc;
000039    while( SQLITE_ROW==(rc = sqlite3_step(pStmt)) ){
000040      const char *zSubSql = (const char*)sqlite3_column_text(pStmt,0);
000041      assert( sqlite3_strnicmp(zSql,"SELECT",6)==0 );
000042      /* The secondary SQL must be one of CREATE TABLE, CREATE INDEX,
000043      ** or INSERT.  Historically there have been attacks that first
000044      ** corrupt the sqlite_schema.sql field with other kinds of statements
000045      ** then run VACUUM to get those statements to execute at inappropriate
000046      ** times. */
000047      if( zSubSql
000048       && (strncmp(zSubSql,"CRE",3)==0 || strncmp(zSubSql,"INS",3)==0)
000049      ){
000050        rc = execSql(db, pzErrMsg, zSubSql);
000051        if( rc!=SQLITE_OK ) break;
000052      }
000053    }
000054    assert( rc!=SQLITE_ROW );
000055    if( rc==SQLITE_DONE ) rc = SQLITE_OK;
000056    if( rc ){
000057      sqlite3SetString(pzErrMsg, db, sqlite3_errmsg(db));
000058    }
000059    (void)sqlite3_finalize(pStmt);
000060    return rc;
000061  }
000062  static int execSqlF(sqlite3 *db, char **pzErrMsg, const char *zSql, ...){
000063    char *z;
000064    va_list ap;
000065    int rc;
000066    va_start(ap, zSql);
000067    z = sqlite3VMPrintf(db, zSql, ap);
000068    va_end(ap);
000069    if( z==0 ) return SQLITE_NOMEM;
000070    rc = execSql(db, pzErrMsg, z);
000071    sqlite3DbFree(db, z);
000072    return rc;
000073  }
000074  
000075  /*
000076  ** The VACUUM command is used to clean up the database,
000077  ** collapse free space, etc.  It is modelled after the VACUUM command
000078  ** in PostgreSQL.  The VACUUM command works as follows:
000079  **
000080  **   (1)  Create a new transient database file
000081  **   (2)  Copy all content from the database being vacuumed into
000082  **        the new transient database file
000083  **   (3)  Copy content from the transient database back into the
000084  **        original database.
000085  **
000086  ** The transient database requires temporary disk space approximately
000087  ** equal to the size of the original database.  The copy operation of
000088  ** step (3) requires additional temporary disk space approximately equal
000089  ** to the size of the original database for the rollback journal.
000090  ** Hence, temporary disk space that is approximately 2x the size of the
000091  ** original database is required.  Every page of the database is written
000092  ** approximately 3 times:  Once for step (2) and twice for step (3).
000093  ** Two writes per page are required in step (3) because the original
000094  ** database content must be written into the rollback journal prior to
000095  ** overwriting the database with the vacuumed content.
000096  **
000097  ** Only 1x temporary space and only 1x writes would be required if
000098  ** the copy of step (3) were replaced by deleting the original database
000099  ** and renaming the transient database as the original.  But that will
000100  ** not work if other processes are attached to the original database.
000101  ** And a power loss in between deleting the original and renaming the
000102  ** transient would cause the database file to appear to be deleted
000103  ** following reboot.
000104  */
000105  void sqlite3Vacuum(Parse *pParse, Token *pNm, Expr *pInto){
000106    Vdbe *v = sqlite3GetVdbe(pParse);
000107    int iDb = 0;
000108    if( v==0 ) goto build_vacuum_end;
000109    if( pParse->nErr ) goto build_vacuum_end;
000110    if( pNm ){
000111  #ifndef SQLITE_BUG_COMPATIBLE_20160819
000112      /* Default behavior:  Report an error if the argument to VACUUM is
000113      ** not recognized */
000114      iDb = sqlite3TwoPartName(pParse, pNm, pNm, &pNm);
000115      if( iDb<0 ) goto build_vacuum_end;
000116  #else
000117      /* When SQLITE_BUG_COMPATIBLE_20160819 is defined, unrecognized arguments
000118      ** to VACUUM are silently ignored.  This is a back-out of a bug fix that
000119      ** occurred on 2016-08-19 (https://www.sqlite.org/src/info/083f9e6270).
000120      ** The buggy behavior is required for binary compatibility with some
000121      ** legacy applications. */
000122      iDb = sqlite3FindDb(pParse->db, pNm);
000123      if( iDb<0 ) iDb = 0;
000124  #endif
000125    }
000126    if( iDb!=1 ){
000127      int iIntoReg = 0;
000128      if( pInto && sqlite3ResolveSelfReference(pParse,0,0,pInto,0)==0 ){
000129        iIntoReg = ++pParse->nMem;
000130        sqlite3ExprCode(pParse, pInto, iIntoReg);
000131      }
000132      sqlite3VdbeAddOp2(v, OP_Vacuum, iDb, iIntoReg);
000133      sqlite3VdbeUsesBtree(v, iDb);
000134    }
000135  build_vacuum_end:
000136    sqlite3ExprDelete(pParse->db, pInto);
000137    return;
000138  }
000139  
000140  /*
000141  ** This routine implements the OP_Vacuum opcode of the VDBE.
000142  */
000143  SQLITE_NOINLINE int sqlite3RunVacuum(
000144    char **pzErrMsg,        /* Write error message here */
000145    sqlite3 *db,            /* Database connection */
000146    int iDb,                /* Which attached DB to vacuum */
000147    sqlite3_value *pOut     /* Write results here, if not NULL. VACUUM INTO */
000148  ){
000149    int rc = SQLITE_OK;     /* Return code from service routines */
000150    Btree *pMain;           /* The database being vacuumed */
000151    Btree *pTemp;           /* The temporary database we vacuum into */
000152    u32 saved_mDbFlags;     /* Saved value of db->mDbFlags */
000153    u64 saved_flags;        /* Saved value of db->flags */
000154    i64 saved_nChange;      /* Saved value of db->nChange */
000155    i64 saved_nTotalChange; /* Saved value of db->nTotalChange */
000156    u32 saved_openFlags;    /* Saved value of db->openFlags */
000157    u8 saved_mTrace;        /* Saved trace settings */
000158    Db *pDb = 0;            /* Database to detach at end of vacuum */
000159    int isMemDb;            /* True if vacuuming a :memory: database */
000160    int nRes;               /* Bytes of reserved space at the end of each page */
000161    int nDb;                /* Number of attached databases */
000162    const char *zDbMain;    /* Schema name of database to vacuum */
000163    const char *zOut;       /* Name of output file */
000164    u32 pgflags = PAGER_SYNCHRONOUS_OFF; /* sync flags for output db */
000165  
000166    if( !db->autoCommit ){
000167      sqlite3SetString(pzErrMsg, db, "cannot VACUUM from within a transaction");
000168      return SQLITE_ERROR; /* IMP: R-12218-18073 */
000169    }
000170    if( db->nVdbeActive>1 ){
000171      sqlite3SetString(pzErrMsg, db,"cannot VACUUM - SQL statements in progress");
000172      return SQLITE_ERROR; /* IMP: R-15610-35227 */
000173    }
000174    saved_openFlags = db->openFlags;
000175    if( pOut ){
000176      if( sqlite3_value_type(pOut)!=SQLITE_TEXT ){
000177        sqlite3SetString(pzErrMsg, db, "non-text filename");
000178        return SQLITE_ERROR;
000179      }
000180      zOut = (const char*)sqlite3_value_text(pOut);
000181      db->openFlags &= ~SQLITE_OPEN_READONLY;
000182      db->openFlags |= SQLITE_OPEN_CREATE|SQLITE_OPEN_READWRITE;
000183    }else{
000184      zOut = "";
000185    }
000186  
000187    /* Save the current value of the database flags so that it can be 
000188    ** restored before returning. Then set the writable-schema flag, and
000189    ** disable CHECK and foreign key constraints.  */
000190    saved_flags = db->flags;
000191    saved_mDbFlags = db->mDbFlags;
000192    saved_nChange = db->nChange;
000193    saved_nTotalChange = db->nTotalChange;
000194    saved_mTrace = db->mTrace;
000195    db->flags |= SQLITE_WriteSchema | SQLITE_IgnoreChecks;
000196    db->mDbFlags |= DBFLAG_PreferBuiltin | DBFLAG_Vacuum;
000197    db->flags &= ~(u64)(SQLITE_ForeignKeys | SQLITE_ReverseOrder
000198                     | SQLITE_Defensive | SQLITE_CountRows);
000199    db->mTrace = 0;
000200  
000201    zDbMain = db->aDb[iDb].zDbSName;
000202    pMain = db->aDb[iDb].pBt;
000203    isMemDb = sqlite3PagerIsMemdb(sqlite3BtreePager(pMain));
000204  
000205    /* Attach the temporary database as 'vacuum_db'. The synchronous pragma
000206    ** can be set to 'off' for this file, as it is not recovered if a crash
000207    ** occurs anyway. The integrity of the database is maintained by a
000208    ** (possibly synchronous) transaction opened on the main database before
000209    ** sqlite3BtreeCopyFile() is called.
000210    **
000211    ** An optimization would be to use a non-journaled pager.
000212    ** (Later:) I tried setting "PRAGMA vacuum_db.journal_mode=OFF" but
000213    ** that actually made the VACUUM run slower.  Very little journalling
000214    ** actually occurs when doing a vacuum since the vacuum_db is initially
000215    ** empty.  Only the journal header is written.  Apparently it takes more
000216    ** time to parse and run the PRAGMA to turn journalling off than it does
000217    ** to write the journal header file.
000218    */
000219    nDb = db->nDb;
000220    rc = execSqlF(db, pzErrMsg, "ATTACH %Q AS vacuum_db", zOut);
000221    db->openFlags = saved_openFlags;
000222    if( rc!=SQLITE_OK ) goto end_of_vacuum;
000223    assert( (db->nDb-1)==nDb );
000224    pDb = &db->aDb[nDb];
000225    assert( strcmp(pDb->zDbSName,"vacuum_db")==0 );
000226    pTemp = pDb->pBt;
000227    if( pOut ){
000228      sqlite3_file *id = sqlite3PagerFile(sqlite3BtreePager(pTemp));
000229      i64 sz = 0;
000230      if( id->pMethods!=0 && (sqlite3OsFileSize(id, &sz)!=SQLITE_OK || sz>0) ){
000231        rc = SQLITE_ERROR;
000232        sqlite3SetString(pzErrMsg, db, "output file already exists");
000233        goto end_of_vacuum;
000234      }
000235      db->mDbFlags |= DBFLAG_VacuumInto;
000236  
000237      /* For a VACUUM INTO, the pager-flags are set to the same values as
000238      ** they are for the database being vacuumed, except that PAGER_CACHESPILL
000239      ** is always set. */
000240      pgflags = db->aDb[iDb].safety_level | (db->flags & PAGER_FLAGS_MASK);
000241    }
000242    nRes = sqlite3BtreeGetRequestedReserve(pMain);
000243  
000244    sqlite3BtreeSetCacheSize(pTemp, db->aDb[iDb].pSchema->cache_size);
000245    sqlite3BtreeSetSpillSize(pTemp, sqlite3BtreeSetSpillSize(pMain,0));
000246    sqlite3BtreeSetPagerFlags(pTemp, pgflags|PAGER_CACHESPILL);
000247  
000248    /* Begin a transaction and take an exclusive lock on the main database
000249    ** file. This is done before the sqlite3BtreeGetPageSize(pMain) call below,
000250    ** to ensure that we do not try to change the page-size on a WAL database.
000251    */
000252    rc = execSql(db, pzErrMsg, "BEGIN");
000253    if( rc!=SQLITE_OK ) goto end_of_vacuum;
000254    rc = sqlite3BtreeBeginTrans(pMain, pOut==0 ? 2 : 0, 0);
000255    if( rc!=SQLITE_OK ) goto end_of_vacuum;
000256  
000257    /* Do not attempt to change the page size for a WAL database */
000258    if( sqlite3PagerGetJournalMode(sqlite3BtreePager(pMain))
000259                                                 ==PAGER_JOURNALMODE_WAL
000260     && pOut==0
000261    ){
000262      db->nextPagesize = 0;
000263    }
000264  
000265    if( sqlite3BtreeSetPageSize(pTemp, sqlite3BtreeGetPageSize(pMain), nRes, 0)
000266     || (!isMemDb && sqlite3BtreeSetPageSize(pTemp, db->nextPagesize, nRes, 0))
000267     || NEVER(db->mallocFailed)
000268    ){
000269      rc = SQLITE_NOMEM_BKPT;
000270      goto end_of_vacuum;
000271    }
000272  
000273  #ifndef SQLITE_OMIT_AUTOVACUUM
000274    sqlite3BtreeSetAutoVacuum(pTemp, db->nextAutovac>=0 ? db->nextAutovac :
000275                                             sqlite3BtreeGetAutoVacuum(pMain));
000276  #endif
000277  
000278    /* Query the schema of the main database. Create a mirror schema
000279    ** in the temporary database.
000280    */
000281    db->init.iDb = nDb; /* force new CREATE statements into vacuum_db */
000282    rc = execSqlF(db, pzErrMsg,
000283        "SELECT sql FROM \"%w\".sqlite_schema"
000284        " WHERE type='table'AND name<>'sqlite_sequence'"
000285        " AND coalesce(rootpage,1)>0",
000286        zDbMain
000287    );
000288    if( rc!=SQLITE_OK ) goto end_of_vacuum;
000289    rc = execSqlF(db, pzErrMsg,
000290        "SELECT sql FROM \"%w\".sqlite_schema"
000291        " WHERE type='index'",
000292        zDbMain
000293    );
000294    if( rc!=SQLITE_OK ) goto end_of_vacuum;
000295    db->init.iDb = 0;
000296  
000297    /* Loop through the tables in the main database. For each, do
000298    ** an "INSERT INTO vacuum_db.xxx SELECT * FROM main.xxx;" to copy
000299    ** the contents to the temporary database.
000300    */
000301    rc = execSqlF(db, pzErrMsg,
000302        "SELECT'INSERT INTO vacuum_db.'||quote(name)"
000303        "||' SELECT*FROM\"%w\".'||quote(name)"
000304        "FROM vacuum_db.sqlite_schema "
000305        "WHERE type='table'AND coalesce(rootpage,1)>0",
000306        zDbMain
000307    );
000308    assert( (db->mDbFlags & DBFLAG_Vacuum)!=0 );
000309    db->mDbFlags &= ~DBFLAG_Vacuum;
000310    if( rc!=SQLITE_OK ) goto end_of_vacuum;
000311  
000312    /* Copy the triggers, views, and virtual tables from the main database
000313    ** over to the temporary database.  None of these objects has any
000314    ** associated storage, so all we have to do is copy their entries
000315    ** from the schema table.
000316    */
000317    rc = execSqlF(db, pzErrMsg,
000318        "INSERT INTO vacuum_db.sqlite_schema"
000319        " SELECT*FROM \"%w\".sqlite_schema"
000320        " WHERE type IN('view','trigger')"
000321        " OR(type='table'AND rootpage=0)",
000322        zDbMain
000323    );
000324    if( rc ) goto end_of_vacuum;
000325  
000326    /* At this point, there is a write transaction open on both the 
000327    ** vacuum database and the main database. Assuming no error occurs,
000328    ** both transactions are closed by this block - the main database
000329    ** transaction by sqlite3BtreeCopyFile() and the other by an explicit
000330    ** call to sqlite3BtreeCommit().
000331    */
000332    {
000333      u32 meta;
000334      int i;
000335  
000336      /* This array determines which meta meta values are preserved in the
000337      ** vacuum.  Even entries are the meta value number and odd entries
000338      ** are an increment to apply to the meta value after the vacuum.
000339      ** The increment is used to increase the schema cookie so that other
000340      ** connections to the same database will know to reread the schema.
000341      */
000342      static const unsigned char aCopy[] = {
000343         BTREE_SCHEMA_VERSION,     1,  /* Add one to the old schema cookie */
000344         BTREE_DEFAULT_CACHE_SIZE, 0,  /* Preserve the default page cache size */
000345         BTREE_TEXT_ENCODING,      0,  /* Preserve the text encoding */
000346         BTREE_USER_VERSION,       0,  /* Preserve the user version */
000347         BTREE_APPLICATION_ID,     0,  /* Preserve the application id */
000348      };
000349  
000350      assert( SQLITE_TXN_WRITE==sqlite3BtreeTxnState(pTemp) );
000351      assert( pOut!=0 || SQLITE_TXN_WRITE==sqlite3BtreeTxnState(pMain) );
000352  
000353      /* Copy Btree meta values */
000354      for(i=0; i<ArraySize(aCopy); i+=2){
000355        /* GetMeta() and UpdateMeta() cannot fail in this context because
000356        ** we already have page 1 loaded into cache and marked dirty. */
000357        sqlite3BtreeGetMeta(pMain, aCopy[i], &meta);
000358        rc = sqlite3BtreeUpdateMeta(pTemp, aCopy[i], meta+aCopy[i+1]);
000359        if( NEVER(rc!=SQLITE_OK) ) goto end_of_vacuum;
000360      }
000361  
000362      if( pOut==0 ){
000363        rc = sqlite3BtreeCopyFile(pMain, pTemp);
000364      }
000365      if( rc!=SQLITE_OK ) goto end_of_vacuum;
000366      rc = sqlite3BtreeCommit(pTemp);
000367      if( rc!=SQLITE_OK ) goto end_of_vacuum;
000368  #ifndef SQLITE_OMIT_AUTOVACUUM
000369      if( pOut==0 ){
000370        sqlite3BtreeSetAutoVacuum(pMain, sqlite3BtreeGetAutoVacuum(pTemp));
000371      }
000372  #endif
000373    }
000374  
000375    assert( rc==SQLITE_OK );
000376    if( pOut==0 ){
000377      nRes = sqlite3BtreeGetRequestedReserve(pTemp);
000378      rc = sqlite3BtreeSetPageSize(pMain, sqlite3BtreeGetPageSize(pTemp), nRes,1);
000379    }
000380  
000381  end_of_vacuum:
000382    /* Restore the original value of db->flags */
000383    db->init.iDb = 0;
000384    db->mDbFlags = saved_mDbFlags;
000385    db->flags = saved_flags;
000386    db->nChange = saved_nChange;
000387    db->nTotalChange = saved_nTotalChange;
000388    db->mTrace = saved_mTrace;
000389    sqlite3BtreeSetPageSize(pMain, -1, 0, 1);
000390  
000391    /* Currently there is an SQL level transaction open on the vacuum
000392    ** database. No locks are held on any other files (since the main file
000393    ** was committed at the btree level). So it safe to end the transaction
000394    ** by manually setting the autoCommit flag to true and detaching the
000395    ** vacuum database. The vacuum_db journal file is deleted when the pager
000396    ** is closed by the DETACH.
000397    */
000398    db->autoCommit = 1;
000399  
000400    if( pDb ){
000401      sqlite3BtreeClose(pDb->pBt);
000402      pDb->pBt = 0;
000403      pDb->pSchema = 0;
000404    }
000405  
000406    /* This both clears the schemas and reduces the size of the db->aDb[]
000407    ** array. */ 
000408    sqlite3ResetAllSchemasOfConnection(db);
000409  
000410    return rc;
000411  }
000412  
000413  #endif  /* SQLITE_OMIT_VACUUM && SQLITE_OMIT_ATTACH */