ADDED ext/misc/rot13.c Index: ext/misc/rot13.c ================================================================== --- ext/misc/rot13.c +++ ext/misc/rot13.c @@ -0,0 +1,114 @@ +/* +** 2013-05-15 +** +** 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. +** +****************************************************************************** +** +** This SQLite extension implements a rot13() function and a rot13 +** collating sequence. +*/ +#include "sqlite3ext.h" +SQLITE_EXTENSION_INIT1 +#include +#include + +/* +** Perform rot13 encoding on a single ASCII character. +*/ +static unsigned char rot13(unsigned char c){ + if( c>='a' && c<='z' ){ + c += 13; + if( c>'z' ) c -= 26; + }else if( c>='A' && c<='Z' ){ + c += 13; + if( c>'Z' ) c -= 26; + } + return c; +} + +/* +** Implementation of the rot13() function. +** +** Rotate ASCII alphabetic characters by 13 character positions. +** Non-ASCII characters are unchanged. rot13(rot13(X)) should always +** equal X. +*/ +static void rot13func( + sqlite3_context *context, + int argc, + sqlite3_value **argv +){ + const unsigned char *zIn; + int nIn; + unsigned char *zOut; + char *zToFree = 0; + int i; + char zTemp[100]; + assert( argc==1 ); + if( sqlite3_value_type(argv[0])==SQLITE_NULL ) return; + zIn = (const unsigned char*)sqlite3_value_text(argv[0]); + nIn = sqlite3_value_bytes(argv[0]); + if( nInpPage1 = 0; return rc; } +#ifndef NDEBUG +/* +** Return the number of cursors open on pBt. This is for use +** in assert() expressions, so it is only compiled if NDEBUG is not +** defined. +** +** Only write cursors are counted if wrOnly is true. If wrOnly is +** false then all cursors are counted. +** +** For the purposes of this routine, a cursor is any cursor that +** is capable of reading or writing to the databse. Cursors that +** have been tripped into the CURSOR_FAULT state are not counted. +*/ +static int countValidCursors(BtShared *pBt, int wrOnly){ + BtCursor *pCur; + int r = 0; + for(pCur=pBt->pCursor; pCur; pCur=pCur->pNext){ + if( (wrOnly==0 || pCur->wrFlag) && pCur->eState!=CURSOR_FAULT ) r++; + } + return r; +} +#endif + /* ** If there are no outstanding cursors and we are not in the middle ** of a transaction but there is a read lock on the database, then ** this routine unrefs the first page of the database file which ** has the effect of releasing the read lock. @@ -2525,11 +2548,11 @@ ** ** If there is a transaction in progress, this routine is a no-op. */ static void unlockBtreeIfUnused(BtShared *pBt){ assert( sqlite3_mutex_held(pBt->mutex) ); - assert( pBt->pCursor==0 || pBt->inTransaction>TRANS_NONE ); + assert( countValidCursors(pBt,0)==0 || pBt->inTransaction>TRANS_NONE ); if( pBt->inTransaction==TRANS_NONE && pBt->pPage1!=0 ){ assert( pBt->pPage1->aData ); assert( sqlite3PagerRefcount(pBt->pPager)==1 ); assert( pBt->pPage1->aData ); releasePage(pBt->pPage1); @@ -3253,11 +3276,10 @@ assert( sqlite3BtreeHoldsMutex(p) ); #ifndef SQLITE_OMIT_AUTOVACUUM pBt->bDoTruncate = 0; #endif - btreeClearHasContent(pBt); if( p->inTrans>TRANS_NONE && p->db->activeVdbeCnt>1 ){ /* If there are other active statements that belong to this database ** handle, downgrade to a read-only transaction. The other statements ** may still be reading from the database. */ downgradeAllSharedCacheTableLocks(p); @@ -3328,10 +3350,11 @@ if( rc!=SQLITE_OK && bCleanup==0 ){ sqlite3BtreeLeave(p); return rc; } pBt->inTransaction = TRANS_READ; + btreeClearHasContent(pBt); } btreeEndTransaction(p); sqlite3BtreeLeave(p); return SQLITE_OK; @@ -3349,31 +3372,10 @@ } sqlite3BtreeLeave(p); return rc; } -#ifndef NDEBUG -/* -** Return the number of write-cursors open on this handle. This is for use -** in assert() expressions, so it is only compiled if NDEBUG is not -** defined. -** -** For the purposes of this routine, a write-cursor is any cursor that -** is capable of writing to the databse. That means the cursor was -** originally opened for writing and the cursor has not be disabled -** by having its state changed to CURSOR_FAULT. -*/ -static int countWriteCursors(BtShared *pBt){ - BtCursor *pCur; - int r = 0; - for(pCur=pBt->pCursor; pCur; pCur=pCur->pNext){ - if( pCur->wrFlag && pCur->eState!=CURSOR_FAULT ) r++; - } - return r; -} -#endif - /* ** This routine sets the state to CURSOR_FAULT and the error ** code to errCode for every cursor on BtShared that pBtree ** references. ** @@ -3449,12 +3451,13 @@ if( nPage==0 ) sqlite3PagerPagecount(pBt->pPager, &nPage); testcase( pBt->nPage!=nPage ); pBt->nPage = nPage; releasePage(pPage1); } - assert( countWriteCursors(pBt)==0 ); + assert( countValidCursors(pBt, 1)==0 ); pBt->inTransaction = TRANS_READ; + btreeClearHasContent(pBt); } btreeEndTransaction(p); sqlite3BtreeLeave(p); return rc; Index: src/build.c ================================================================== --- src/build.c +++ src/build.c @@ -2657,14 +2657,12 @@ ** specified collation sequence names. */ for(i=0; inExpr; i++){ Expr *pExpr = pList->a[i].pExpr; if( pExpr ){ - CollSeq *pColl = sqlite3ExprCollSeq(pParse, pExpr); - if( pColl ){ - nExtra += (1 + sqlite3Strlen30(pColl->zName)); - } + assert( pExpr->op==TK_COLLATE ); + nExtra += (1 + sqlite3Strlen30(pExpr->u.zToken)); } } /* ** Allocate the index structure. @@ -2721,11 +2719,10 @@ */ for(i=0, pListItem=pList->a; inExpr; i++, pListItem++){ const char *zColName = pListItem->zName; Column *pTabCol; int requestedSortOrder; - CollSeq *pColl; /* Collating sequence */ char *zColl; /* Collation sequence name */ for(j=0, pTabCol=pTab->aCol; jnCol; j++, pTabCol++){ if( sqlite3StrICmp(zColName, pTabCol->zName)==0 ) break; } @@ -2734,26 +2731,23 @@ pTab->zName, zColName); pParse->checkSchema = 1; goto exit_create_index; } pIndex->aiColumn[i] = j; - if( pListItem->pExpr - && (pColl = sqlite3ExprCollSeq(pParse, pListItem->pExpr))!=0 - ){ + if( pListItem->pExpr ){ int nColl; - zColl = pColl->zName; + assert( pListItem->pExpr->op==TK_COLLATE ); + zColl = pListItem->pExpr->u.zToken; nColl = sqlite3Strlen30(zColl) + 1; assert( nExtra>=nColl ); memcpy(zExtra, zColl, nColl); zColl = zExtra; zExtra += nColl; nExtra -= nColl; }else{ zColl = pTab->aCol[j].zColl; - if( !zColl ){ - zColl = "BINARY"; - } + if( !zColl ) zColl = "BINARY"; } if( !db->init.busy && !sqlite3LocateCollSeq(pParse, zColl) ){ goto exit_create_index; } pIndex->azColl[i] = zColl; Index: src/expr.c ================================================================== --- src/expr.c +++ src/expr.c @@ -114,16 +114,11 @@ p = p->pLeft; continue; } assert( op!=TK_REGISTER || p->op2!=TK_COLLATE ); if( op==TK_COLLATE ){ - if( db->init.busy ){ - /* Do not report errors when parsing while the schema */ - pColl = sqlite3FindCollSeq(db, ENC(db), p->u.zToken, 0); - }else{ - pColl = sqlite3GetCollSeq(pParse, ENC(db), 0, p->u.zToken); - } + pColl = sqlite3GetCollSeq(pParse, ENC(db), 0, p->u.zToken); break; } if( p->pTab!=0 && (op==TK_AGG_COLUMN || op==TK_COLUMN || op==TK_REGISTER || op==TK_TRIGGER) Index: src/main.c ================================================================== --- src/main.c +++ src/main.c @@ -846,10 +846,16 @@ "statements or unfinished backups"); sqlite3_mutex_leave(db->mutex); return SQLITE_BUSY; } + /* If a transaction is open, roll it back. This also ensures that if + ** any database schemas have been modified by the current transaction + ** they are reset. And that the required b-tree mutex is held to make + ** the the pager rollback and schema reset an atomic operation. */ + sqlite3RollbackAll(db, SQLITE_OK); + #ifdef SQLITE_ENABLE_SQLLOG if( sqlite3GlobalConfig.xSqllog ){ /* Closing the handle. Fourth parameter is passed the value 2. */ sqlite3GlobalConfig.xSqllog(sqlite3GlobalConfig.pSqllogArg, db, 0, 2); } @@ -1000,10 +1006,11 @@ void sqlite3RollbackAll(sqlite3 *db, int tripCode){ int i; int inTrans = 0; assert( sqlite3_mutex_held(db->mutex) ); sqlite3BeginBenignMalloc(); + sqlite3BtreeEnterAll(db); for(i=0; inDb; i++){ Btree *p = db->aDb[i].pBt; if( p ){ if( sqlite3BtreeIsInTrans(p) ){ inTrans = 1; @@ -1017,10 +1024,11 @@ if( (db->flags&SQLITE_InternChanges)!=0 && db->init.busy==0 ){ sqlite3ExpirePreparedStatements(db); sqlite3ResetAllSchemasOfConnection(db); } + sqlite3BtreeLeaveAll(db); /* Any deferred constraint violations have now been resolved. */ db->nDeferredCons = 0; /* If one has been configured, invoke the rollback-hook callback */ Index: test/collate3.test ================================================================== --- test/collate3.test +++ test/collate3.test @@ -52,10 +52,108 @@ } {1 {no such collation sequence: garbage}} execsql { DROP TABLE collate3t1; } + +proc caseless {a b} { string compare -nocase $a $b } +do_test collate3-1.4 { + db collate caseless caseless + execsql { + CREATE TABLE t1(a COLLATE caseless); + INSERT INTO t1 VALUES('Abc2'); + INSERT INTO t1 VALUES('abc1'); + INSERT INTO t1 VALUES('aBc3'); + } + execsql { SELECT * FROM t1 ORDER BY a } +} {abc1 Abc2 aBc3} + +do_test collate3-1.5 { + db close + sqlite3 db test.db + catchsql { SELECT * FROM t1 ORDER BY a } +} {1 {no such collation sequence: caseless}} + +do_test collate3-1.6.1 { + db collate caseless caseless + execsql { CREATE INDEX i1 ON t1(a) } + execsql { SELECT * FROM t1 ORDER BY a } +} {abc1 Abc2 aBc3} + +do_test collate3-1.6.2 { + db close + sqlite3 db test.db + catchsql { SELECT * FROM t1 ORDER BY a } +} {1 {no such collation sequence: caseless}} + +do_test collate3-1.6.3 { + db close + sqlite3 db test.db + catchsql { PRAGMA integrity_check } +} {1 {no such collation sequence: caseless}} + +do_test collate3-1.6.4 { + db close + sqlite3 db test.db + catchsql { REINDEX } +} {1 {no such collation sequence: caseless}} + +do_test collate3-1.7.1 { + db collate caseless caseless + execsql { + DROP TABLE t1; + CREATE TABLE t1(a); + CREATE INDEX i1 ON t1(a COLLATE caseless); + INSERT INTO t1 VALUES('Abc2'); + INSERT INTO t1 VALUES('abc1'); + INSERT INTO t1 VALUES('aBc3'); + SELECT * FROM t1 ORDER BY a COLLATE caseless; + } +} {abc1 Abc2 aBc3} + +do_test collate3-1.7.2 { + db close + sqlite3 db test.db + catchsql { SELECT * FROM t1 ORDER BY a COLLATE caseless} +} {1 {no such collation sequence: caseless}} + +do_test collate3-1.7.4 { + db close + sqlite3 db test.db + catchsql { REINDEX } +} {1 {no such collation sequence: caseless}} + +do_test collate3-1.7.3 { + db close + sqlite3 db test.db + catchsql { PRAGMA integrity_check } +} {1 {no such collation sequence: caseless}} + +do_test collate3-1.7.4 { + db close + sqlite3 db test.db + catchsql { REINDEX } +} {1 {no such collation sequence: caseless}} + +do_test collate3-1.7.5 { + db close + sqlite3 db test.db + db collate caseless caseless + catchsql { PRAGMA integrity_check } +} {0 ok} + +proc needed {nm} { db collate caseless caseless } +do_test collate3-1.7.6 { + db close + sqlite3 db test.db + db collation_needed needed + catchsql { PRAGMA integrity_check } +} {0 ok} + +do_test collate3-1.8 { + execsql { DROP TABLE t1 } +} {} # # Create a table with a default collation sequence, then close # and re-open the database without re-registering the collation # sequence. Then make sure the library stops us from using ADDED test/sharedA.test Index: test/sharedA.test ================================================================== --- test/sharedA.test +++ test/sharedA.test @@ -0,0 +1,188 @@ +# 2013 May 14 +# +# 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. +# +#*********************************************************************** +# +# Test some specific circumstances to do with shared cache mode. +# + + +set testdir [file dirname $argv0] +source $testdir/tester.tcl +if {[run_thread_tests]==0} { finish_test ; return } +db close +set ::testprefix sharedA + +set ::enable_shared_cache [sqlite3_enable_shared_cache 1] + +#------------------------------------------------------------------------- +# +do_test 0.1 { + sqlite3 db1 test.db + sqlite3 db2 test.db + + db1 eval { + CREATE TABLE t1(x); + INSERT INTO t1 VALUES(randomblob(100)); + INSERT INTO t1 SELECT randomblob(100) FROM t1; + INSERT INTO t1 SELECT randomblob(100) FROM t1; + INSERT INTO t1 SELECT randomblob(100) FROM t1; + INSERT INTO t1 SELECT randomblob(100) FROM t1; + INSERT INTO t1 SELECT randomblob(100) FROM t1; + INSERT INTO t1 SELECT randomblob(100) FROM t1; + CREATE INDEX i1 ON t1(x); + } + + db1 eval { + BEGIN; + DROP INDEX i1; + } + + db2 close + + db1 eval { + INSERT INTO t1 SELECT randomblob(100) FROM t1; + ROLLBACK; + PRAGMA integrity_check; + } +} {ok} + +db1 close +forcedelete test.db + + +#------------------------------------------------------------------------- +# +do_test 1.1 { + sqlite3 db1 test.db + sqlite3 db2 test.db + db2 eval { + CREATE TABLE t1(x); + INSERT INTO t1 VALUES(123); + } + db1 eval { + SELECT * FROM t1; + CREATE INDEX i1 ON t1(x); + } +} {123} + +do_test 1.2 { + db2 eval { SELECT * FROM t1 ORDER BY x; } + + db1 eval { + BEGIN; DROP INDEX i1; + } + db1 close + + db2 eval { SELECT * FROM t1 ORDER BY x; } +} {123} + +do_test 1.3 { + db2 close +} {} + +#------------------------------------------------------------------------- +# +# sqlite3RollbackAll() loops through all attached b-trees and rolls +# back each one separately. Then if the SQLITE_InternChanges flag is +# set, it resets the schema. Both of the above steps must be done +# while holding a mutex, otherwise another thread might slip in and +# try to use the new schema with the old data. +# +# The following sequence of tests attempt to verify that the actions +# taken by sqlite3RollbackAll() are thread-atomic (that they cannot be +# interrupted by a separate thread.) +# +# Note that a TCL interpreter can only be used within the thread in which +# it was originally created (because it uses thread-local-storage). +# The tvfs callbacks must therefore only run on the main thread. +# There is some trickery in the read_callback procedure to ensure that +# this is the case. +# +testvfs tvfs + +# Set up two databases and two database connections. +# +# db1: main(test.db), two(test2.db) +# db2: main(test.db) +# +# The cache for test.db is shared between db1 and db2. +# +do_test 2.1 { + forcedelete test.db test.db2 + sqlite3 db1 test.db -vfs tvfs + db1 eval { ATTACH 'test.db2' AS two } + + db1 eval { + CREATE TABLE t1(x); + INSERT INTO t1 VALUES(1); + INSERT INTO t1 VALUES(2); + INSERT INTO t1 VALUES(3); + CREATE TABLE two.t2(x); + INSERT INTO t2 SELECT * FROM t1; + } + + sqlite3 db2 test.db -vfs tvfs + db2 eval { SELECT * FROM t1 } +} {1 2 3} + +# Create a prepared statement on db2 that will attempt a schema change +# in test.db. Meanwhile, start a transaction on db1 that changes +# the schema of test.db and that creates a rollback journal on test2.db +# +do_test 2.2 { + set ::STMT [sqlite3_prepare db2 "CREATE INDEX i1 ON t1(x)" -1 tail] + db1 eval { + BEGIN; + CREATE INDEX i1 ON t1(x); + INSERT INTO t2 VALUES('value!'); + } +} {} + +# Set up a callback that will cause db2 to try to execute its +# schema change when db1 accesses the journal file of test2.db. +# +# This callback will be invoked after the content of test.db has +# be rolled back but before the schema has been reset. If the +# sqlite3RollbackAll() operation is not thread-atomic, then the +# db2 statement in the callback will see old content with the newer +# schema, which is wrong. +# +tvfs filter xRead +tvfs script read_callback +unset -nocomplain ::some_time_laster +unset -nocomplain ::thread_result +proc read_callback {call file args} { + if {[string match *test.db2-journal $file]} { + tvfs filter {} ;# Ensure that tvfs callbacks to do run on the + # child thread + sqlthread spawn ::thread_result [subst -nocommands { + sqlite3_step $::STMT + set rc [sqlite3_finalize $::STMT] + }] + after 1000 { set ::some_time_later 1 } + vwait ::some_time_later + } +} +do_test 2.3 { db1 eval ROLLBACK } {} + +# Verify that the db2 statement invoked by the callback detected the +# schema change. +# +if {[info exists ::thread_result]==0} { vwait ::thread_result } +do_test 2.4 { + list $::thread_result [sqlite3_errmsg db2] +} {SQLITE_SCHEMA {database schema has changed}} + +db1 close +db2 close +tvfs delete + +sqlite3_enable_shared_cache $::enable_shared_cache +finish_test