Tests for text encoding conversion functions. Also new sqlite3_bindXX APIs. (CVS...

FossilOrigin-Name: 4e602bb473e22cc45de2f5788c035d18586cb836

diff --git a/manifest b/manifest
index a54c16f03308de7db53665f0883a03cbe50732e6..48f6318da32ea1c7d04183d31e974803c21549d2 100644 (file)
--- a/manifest
+++ b/manifest
@@ -1,5 +1,5 @@
-C Add\sdefinitions\sof\sthe\sCollSeq\sand\sKeyInfo\sstructures.\s(CVS\s1399)
-D 2004-05-18T23:21:36
+C Tests\sfor\stext\sencoding\sconversion\sfunctions.\sAlso\snew\ssqlite3_bindXX\sAPIs.\s(CVS\s1400)
+D 2004-05-19T10:34:52
 F Makefile.in ab7b0d5118e2da97bac66be8684a1034e3500f5a
 F Makefile.linux-gcc b86a99c493a5bfb402d1d9178dcdc4bd4b32f906
 F README f1de682fbbd94899d50aca13d387d1b3fd3be2dd
@@ -48,7 +48,7 @@ F src/printf.c ef750e8e2398ca7e8b58be991075f08c6a7f0e53
 F src/random.c eff68e3f257e05e81eae6c4d50a51eb88beb4ff3
 F src/select.c 2981cafe3b21ca06c97ddec8c7181f209a06ee82
 F src/shell.c 0c4662e13bfbfd3d13b066c5859cc97ad2f95d21
-F src/sqlite.h.in 799c5e726296ec7bc20e6407cdf4df0e0bc00c0c
+F src/sqlite.h.in 8c62076ea226b1870df977d7438bf99383d02387
 F src/sqliteInt.h 1da4a9c26da43f235c45c1de0c49fadf6c40d0e6
 F src/table.c af14284fa36c8d41f6829e3f2819dce07d3e2de2
 F src/tclsqlite.c fbf0fac73624ae246551a6c671f1de0235b5faa1
@@ -56,14 +56,14 @@ F src/test1.c 12ef76b8aaba4408422f21f269256b630d4dd627
 F src/test2.c 6195a1ca2c8d0d2d93644e86da3289b403486872
 F src/test3.c 5e4a6d596f982f6f47a5f9f75ede9b4a3b739968
 F src/test4.c b3fab9aea7a8940a8a7386ce1c7e2157b09bd296
-F src/test5.c eb39aac8fed61bd930b92613cd705c145244074a
+F src/test5.c c92dca7028b19b9c8319d55e0a5037fc183640a6
 F src/tokenize.c e7536dd31205d5afb76c1bdc832dea009c7a3847
 F src/trigger.c 11afe9abfba13a2ba142944c797c952e162d117f
 F src/update.c 0cc7291dd0e0f82cf93085e49c973e8ef9e51fd5
-F src/utf.c fc799748d43fe1982d157b871e3e420a19c85d4f
-F src/util.c 6d4339b7f05ccdacaebcce67e7fb8c5b880620e8
+F src/utf.c 48c537bf7990ce32a36b051401874d024ec2a07b
+F src/util.c b72f775a6c3fa404d70250382f63d708e17bc332
 F src/vacuum.c c134702e023db8778e6be59ac0ea7b02315b5476
-F src/vdbe.c 37c3d0c45fd6bd7096ae50d08e864acf41924b34
+F src/vdbe.c f5451f1f71e46acebe0be797a138302b4b7ead43
 F src/vdbe.h 314e9c07db73a42a6ba91ab7539e27652fc88870
 F src/vdbeInt.h 5bac5f0f468205f6e43a4ba86a807abff4953abb
 F src/vdbeaux.c 95f5a9ff770794f5165d5827d2343eb2d83d7a6d
@@ -193,7 +193,7 @@ F www/sqlite.tcl 3c83b08cf9f18aa2d69453ff441a36c40e431604
 F www/tclsqlite.tcl b9271d44dcf147a93c98f8ecf28c927307abd6da
 F www/vdbe.tcl 9b9095d4495f37697fd1935d10e14c6015e80aa1
 F www/whentouse.tcl a8335bce47cc2fddb07f19052cb0cb4d9129a8e4
-P ae37e8a0bff39fd1568eae83f8761c34cd0184fc
-R 3c5b4370c9c3ea00ed8ad90f1394d1c1
-U drh
-Z b84fedd6f94a949476c1655f09bd2d6c
+P cd1be81569aa6d5a365eb27b7d067a153079ce45
+R 334bdded4c677d317efa922d114eed4d
+U danielk1977
+Z 68bec39604761b12cf9824e9ba6afe63

diff --git a/manifest.uuid b/manifest.uuid
index 58547655f35d8d4a4dcafd3cf46b82dd73661739..f9b5f73b79fe21b159027fde08e13fa5aad6647a 100644 (file)
--- a/manifest.uuid
+++ b/manifest.uuid
@@ -1 +1 @@
-cd1be81569aa6d5a365eb27b7d067a153079ce45
\ No newline at end of file
+4e602bb473e22cc45de2f5788c035d18586cb836
\ No newline at end of file

diff --git a/src/sqlite.h.in b/src/sqlite.h.in
index 7b3f32a209bb81039ae31167ce9441726d1ecd56..9f35eb40b8411f5cd225955a04cf315c57596949 100644 (file)
--- a/src/sqlite.h.in
+++ b/src/sqlite.h.in
@@ -12,7 +12,7 @@
 ** This header file defines the interface that the SQLite library
 ** presents to client programs.
 **
-** @(#) $Id: sqlite.h.in,v 1.63 2004/05/10 16:18:48 drh Exp $
+** @(#) $Id: sqlite.h.in,v 1.64 2004/05/19 10:34:52 danielk1977 Exp $
 */
 #ifndef _SQLITE_H_
 #define _SQLITE_H_
@@ -861,6 +861,16 @@ int sqlite_encode_binary(const unsigned char *in, int n, unsigned char *out);
 */
 int sqlite_decode_binary(const unsigned char *in, unsigned char *out);
 
+typedef sqlite_vm sqlite3_stmt;
+
+int sqlite3_bind_int32(sqlite3_stmt*, int i, int iValue);
+int sqlite3_bind_int64(sqlite3_stmt*, int i, long long int iValue);
+int sqlite3_bind_double(sqlite3_stmt*, int i, double iValue);
+int sqlite3_bind_null(sqlite3_stmt*, int i);
+int sqlite3_bind_text(sqlite3_stmt*, int i, const char*, int n, int eCopy);
+int sqlite3_bind_text16(sqlite3_stmt*, int i, const void*, int, int eCopy);
+int sqlite3_bind_blob(sqlite3_stmt*, int i, const void*, int n, int eCopy);
+
 #if 0
 
 /*
@@ -887,14 +897,6 @@ int sqlite3_prepare16(sqlite3*, const void*, sqlite3_stmt**, const void**);
 int sqlite3_finalize(sqlite3_stmt*);
 int sqlite3_reset(sqlite3_stmt*);
 
-int sqlite3_bind_int32(sqlite3_stmt*, int iParm, int iValue);
-int sqlite3_bind_int64(sqlite3_stmt*, int iParm, long long int iValue);
-int sqlite3_bind_double(sqlite3_stmt*, int iParm, double iValue);
-int sqlite3_bind_null(sqlite3_stmt*, int iParm);
-int sqlite3_bind_text(sqlite3_stmt*, int i, const char*, int n, int eCopy);
-int sqlite3_bind_text16(sqlite3_stmt*, int i, const void*, int, int eCopy);
-int sqlite3_bind_blob(sqlite3_stmt*, int i, const void*, int n, int eCopy);
-
 int sqlite3_step(sqlite3_stmt*);
 
 #define SQLITE3_INTEGER  1

diff --git a/src/test5.c b/src/test5.c
index ef0e90d394a3518fd0a09a47648a5d41096e471d..816db4cc0b06c45329ebf0a8e459f3075c2748f4 100644 (file)
--- a/src/test5.c
+++ b/src/test5.c
@@ -11,9 +11,11 @@
 *************************************************************************
 ** Code for testing the utf.c module in SQLite.  This code
 ** is not included in the SQLite library.  It is used for automated
-** testing of the SQLite library.
+** testing of the SQLite library. Specifically, the code in this file
+** is used for testing the SQLite routines for converting between
+** the various supported unicode encodings.
 **
-** $Id: 
+** $Id: test5.c,v 1.4 2004/05/19 10:34:53 danielk1977 Exp $
 */
 #include "sqliteInt.h"
 #include "tcl.h"
@@ -21,8 +23,8 @@
 #include <string.h>
 
 /*
-** Return the number of bytes up to and including the first \u0000 
-** character in *pStr.
+** Return the number of bytes up to and including the first pair of
+** 0x00 bytes in *pStr.
 */
 static int utf16_length(const unsigned char *pZ){
   const unsigned char *pC1 = pZ;
@@ -50,10 +52,10 @@ static int sqlite_utf8to16le(
     return TCL_ERROR;
   }
 
-  in = Tcl_GetByteArrayFromObj(objv[1], 0);
+  in = Tcl_GetString(objv[1]);
   out = (unsigned char *)sqlite3utf8to16le(in, -1);
   res = Tcl_NewByteArrayObj(out, utf16_length(out));
-  sqliteFree(out);
+  sqliteFree(out); 
 
   Tcl_SetObjResult(interp, res);
 
@@ -77,6 +79,7 @@ static int sqlite_utf8to16be(
   }
 
   in = Tcl_GetByteArrayFromObj(objv[1], 0);
+  in = Tcl_GetString(objv[1]);
   out = (unsigned char *)sqlite3utf8to16be(in, -1);
   res = Tcl_NewByteArrayObj(out, utf16_length(out));
   sqliteFree(out);
@@ -158,13 +161,13 @@ static int sqlite_utf16to8(
 
   if( objc!=2 ){
     Tcl_AppendResult(interp, "wrong # args: should be \"",
-        Tcl_GetStringFromObj(objv[0], 0), "<utf-16 encoded-string>", 0);
+        Tcl_GetStringFromObj(objv[0], 0), " <utf-16 encoded-string>", 0);
     return TCL_ERROR;
   }
 
   in = Tcl_GetByteArrayFromObj(objv[1], 0);
   out = sqlite3utf16to8(in, -1);
-  res = Tcl_NewByteArrayObj(out, strlen(out));
+  res = Tcl_NewByteArrayObj(out, strlen(out)+1);
   sqliteFree(out);
 
   Tcl_SetObjResult(interp, res);
@@ -172,6 +175,28 @@ static int sqlite_utf16to8(
   return TCL_OK;
 }
 
+/*
+** The first argument is a TCL UTF-8 string. Return the byte array
+** object with the encoded representation of the string, including
+** the NULL terminator.
+*/
+static int binarize(
+  void * clientData,
+  Tcl_Interp *interp,
+  int objc,
+  Tcl_Obj *CONST objv[]
+){
+  int len;
+  char *bytes;
+  Tcl_Obj *pRet;
+  assert(objc==2);
+
+  bytes = Tcl_GetStringFromObj(objv[1], &len);
+  pRet = Tcl_NewByteArrayObj(bytes, len+1);
+  Tcl_SetObjResult(interp, pRet);
+  return TCL_OK;
+}
+
 
 /*
 ** Register commands with the TCL interpreter.
@@ -179,17 +204,18 @@ static int sqlite_utf16to8(
 int Sqlitetest5_Init(Tcl_Interp *interp){
   static struct {
     char *zName;
-    Tcl_CmdProc *xProc;
+    Tcl_ObjCmdProc *xProc;
   } aCmd[] = {
-    { "sqlite_utf16to8",         (Tcl_CmdProc*)sqlite_utf16to8    },
-    { "sqlite_utf8to16le",       (Tcl_CmdProc*)sqlite_utf8to16le  },
-    { "sqlite_utf8to16be",       (Tcl_CmdProc*)sqlite_utf8to16be  },
-    { "sqlite_utf16to16le",      (Tcl_CmdProc*)sqlite_utf16to16le },
-    { "sqlite_utf16to16be",      (Tcl_CmdProc*)sqlite_utf16to16be }
+    { "sqlite_utf16to8",         (Tcl_ObjCmdProc*)sqlite_utf16to8    },
+    { "sqlite_utf8to16le",       (Tcl_ObjCmdProc*)sqlite_utf8to16le  },
+    { "sqlite_utf8to16be",       (Tcl_ObjCmdProc*)sqlite_utf8to16be  },
+    { "sqlite_utf16to16le",      (Tcl_ObjCmdProc*)sqlite_utf16to16le },
+    { "sqlite_utf16to16be",      (Tcl_ObjCmdProc*)sqlite_utf16to16be },
+    { "binarize",                (Tcl_ObjCmdProc*)binarize }
   };
   int i;
   for(i=0; i<sizeof(aCmd)/sizeof(aCmd[0]); i++){
-    Tcl_CreateCommand(interp, aCmd[i].zName, aCmd[i].xProc, 0, 0);
+    Tcl_CreateObjCommand(interp, aCmd[i].zName, aCmd[i].xProc, 0, 0);
   }
 
   return TCL_OK;

diff --git a/src/utf.c b/src/utf.c
index 75a49877917b905ced46933ea7a6eb875fc32813..74c560986036842fc5ad0404e2612e0ee785dc70 100644 (file)
--- a/src/utf.c
+++ b/src/utf.c
@@ -12,7 +12,7 @@
 ** This file contains routines used to translate between UTF-8, 
 ** UTF-16, UTF-16BE, and UTF-16LE.
 **
-** $Id: utf.c,v 1.3 2004/05/08 08:23:40 danielk1977 Exp $
+** $Id: utf.c,v 1.4 2004/05/19 10:34:57 danielk1977 Exp $
 **
 ** Notes on UTF-8:
 **
@@ -192,9 +192,9 @@ static int writeUtf8(UtfString *pStr, u32 code){
     {0x0010FFFF, 3, 0xF7, 0xF0},
     {0x00000000, 0, 0x00, 0x00}
   };
-  static const struct Utf8WriteTblRow *pRow = &utf8tbl[0];
+  struct Utf8WriteTblRow *pRow = &utf8tbl[0];
 
-  while( code<=pRow->max_code ){
+  while( code>pRow->max_code ){
     assert( pRow->max_code );
     pRow++;
   }
@@ -367,7 +367,7 @@ unsigned char *sqlite3utf16to8(const void *pData, int N){
   ** this now.
   */
   out.n = (in.n*1.5) + 1;
-  out.pZ = sqliteMalloc(in.n);
+  out.pZ = sqliteMalloc(out.n);
   if( !out.pZ ){
     return 0;
   }
@@ -402,7 +402,7 @@ static void *utf8toUtf16(const unsigned char *pIn, int N, int big_endian){
   ** this now.
   */
   out.n = (in.n*2) + 2;
-  out.pZ = sqliteMalloc(in.n);
+  out.pZ = sqliteMalloc(out.n);
   if( !out.pZ ){
     return 0;
   }
@@ -451,10 +451,20 @@ static void utf16to16(void *pData, int N, int big_endian){
   }
 
   if( readUtf16Bom(&inout)!=big_endian ){
-    swab(&inout.pZ[inout.c], inout.pZ, inout.n-inout.c);
+    /* swab(&inout.pZ[inout.c], inout.pZ, inout.n-inout.c); */
+    int i;
+    for(i=0; i<(inout.n-inout.c); i += 2){
+      char c1 = inout.pZ[i+inout.c];
+      char c2 = inout.pZ[i+inout.c+1];
+      inout.pZ[i] = c2;
+      inout.pZ[i+1] = c1;
+    }
   }else if( inout.c ){
     memmove(inout.pZ, &inout.pZ[inout.c], inout.n-inout.c);
   }
+
+  inout.pZ[inout.n-inout.c] = 0x00;
+  inout.pZ[inout.n-inout.c+1] = 0x00;
 }
 
 /*

diff --git a/src/util.c b/src/util.c
index 50c639a100e8699c7a1b1f4e188bad80fb45662b..d37a56eb20a3c14a471a2d4615d6964396e360b8 100644 (file)
--- a/src/util.c
+++ b/src/util.c
@@ -14,7 +14,7 @@
 ** This file contains functions for allocating memory, comparing
 ** strings, and stuff like that.
 **
-** $Id: util.c,v 1.84 2004/05/18 22:03:43 drh Exp $
+** $Id: util.c,v 1.85 2004/05/19 10:34:57 danielk1977 Exp $
 */
 #include "sqliteInt.h"
 #include <stdarg.h>
@@ -1180,3 +1180,4 @@ int sqlite3VarintLen(u64 v){
   }while( v!=0 && i<9 );
   return i;
 }
+

diff --git a/src/vdbe.c b/src/vdbe.c
index 72052f4274d476de6ab4e087aa1ab1d3beb031c9..a16fbb59534045f2c4049895b5971821c282b573 100644 (file)
--- a/src/vdbe.c
+++ b/src/vdbe.c
@@ -43,7 +43,7 @@
 ** in this file for details.  If in doubt, do not deviate from existing
 ** commenting and indentation practices when changing or adding code.
 **
-** $Id: vdbe.c,v 1.299 2004/05/18 22:17:46 drh Exp $
+** $Id: vdbe.c,v 1.300 2004/05/19 10:35:01 danielk1977 Exp $
 */
 #include "sqliteInt.h"
 #include "os.h"
@@ -852,14 +852,41 @@ case OP_Real: {
 */
 case OP_Variable: {
   int j = pOp->p1 - 1;
+  Mem *pVar;
+  assert( j>=0 && j<p->nVar );
+
+  /* If we need to translate between text encodings, do it now. If this is
+  ** required, then put the new string in p->azVar. This way, if the
+  ** variable is used again, even after the virtual machine is reset, the
+  ** conversion won't have to be done again.
+  **
+  ** TODO: This is where we need to support databases that use other than
+  ** UTF-8 on disk.
+  */
+  pVar = &p->azVar[j];
+  if( pVar->flags&MEM_Str && !(pVar->flags&MEM_Utf8) ){
+    char *zUtf8;
+    assert( pVar->flags&(MEM_Utf16le|MEM_Utf16be) );
+    zUtf8 = sqlite3utf16to8(pVar->z, pVar->n);
+    if( !zUtf8 ){
+      goto no_mem;
+    }
+    Release(pVar);
+    pVar->z = zUtf8;
+    pVar->n = strlen(zUtf8)+1;
+    pVar->flags = MEM_Str|MEM_Dyn;
+  }
+
+  /* Copy the value in pVar to the top of the stack. If pVar is a string or
+  ** a blob just store a pointer to the same memory, do not make a copy.
+  */
   pTos++;
-  if( j>=0 && j<p->nVar && p->azVar[j]!=0 ){
-    pTos->z = p->azVar[j];
-    pTos->n = p->anVar[j];
-    pTos->flags = MEM_Str | MEM_Static;
-  }else{
-    pTos->flags = MEM_Null;
+  memcpy(pTos, pVar, sizeof(*pVar)-NBFS);
+  if( pTos->flags&(MEM_Str|MEM_Blob) ){
+    pTos->flags &= ~(MEM_Dyn|MEM_Ephem|MEM_Short);
+    pTos->flags |= MEM_Static;
   }
+
   break;
 }
 
@@ -5166,7 +5193,7 @@ default: {
     /* Sanity checking on the top element of the stack */
     if( pTos>=p->aStack ){
       assert( pTos->flags!=0 );  /* Must define some type */
-      if( pTos->flags & MEM_Str ){
+      if( pTos->flags & (MEM_Str|MEM_Blob) ){
         int x = pTos->flags & (MEM_Static|MEM_Dyn|MEM_Ephem|MEM_Short);
         assert( x!=0 );            /* Strings must define a string subtype */
         assert( (x & (x-1))==0 );  /* Only one string subtype can be defined */