database/sqlite
1
0
Fork 0
mirror of https://github.com/sqlite/sqlite.git synced 2025-07-30 19:03:16 +03:00
Code Activity

Proper surrogate pair decoding added to JSON functions. See the mailing list

Browse Source 
bug report and [https://bugs.python.org/issue38749].  More test cases
needed here, but it seems to work so far.

FossilOrigin-Name: 51027f08c0478f1bf9d7545d9e268c772c0a5cd5dda4b03d78f16c7d94f2f50d
This commit is contained in:
drh
2019-11-10 11:09:06 +00:00
parent 9576802dba
commit 48eb03bd0e
4 changed files with 75 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files

64
ext/misc/json1.c
View File

@ -522,6 +522,37 @@ static void jsonReturnJson(
sqlite3_result_subtype(pCtx, JSON_SUBTYPE);
}
/*
** Translate a single byte of Hex into an integer.
** This routine only works if h really is a valid hexadecimal
** character: 0..9a..fA..F
*/
static u8 jsonHexToInt(int h){
assert( (h>='0' && h<='9') || (h>='a' && h<='f') || (h>='A' && h<='F') );
#ifdef SQLITE_EBCDIC
h += 9*(1&~(h>>4));
#else
h += 9*(1&(h>>6));
#endif
return (u8)(h & 0xf);
}
/*
** Convert a 4-byte hex string into an integer
*/
static u32 jsonHexToInt4(const char *z){
u32 v;
assert( safe_isxdigit(z[0]) );
assert( safe_isxdigit(z[1]) );
assert( safe_isxdigit(z[2]) );
assert( safe_isxdigit(z[3]) );
v = (jsonHexToInt(z[0])<<12)
+ (jsonHexToInt(z[1])<<8)
+ (jsonHexToInt(z[2])<<4)
+ jsonHexToInt(z[3]);
return v;
}
/*
** Make the JsonNode the return value of the function.
*/
@ -615,15 +646,8 @@ static void jsonReturn(
}else{
c = z[++i];
if( c=='u' ){
u32 v = 0, k;
for(k=0; k<4; i++, k++){
assert( i<n-2 );
c = z[i+1];
assert( safe_isxdigit(c) );
if( c<='9' ) v = v*16 + c - '0';
else if( c<='F' ) v = v*16 + c - 'A' + 10;
else v = v*16 + c - 'a' + 10;
}
u32 v = jsonHexToInt4(z+i+1);
i += 4;
if( v==0 ) break;
if( v<=0x7f ){
zOut[j++] = (char)v;
@ -631,9 +655,25 @@ static void jsonReturn(
zOut[j++] = (char)(0xc0 | (v>>6));
zOut[j++] = 0x80 | (v&0x3f);
}else{
zOut[j++] = (char)(0xe0 | (v>>12));
zOut[j++] = 0x80 | ((v>>6)&0x3f);
zOut[j++] = 0x80 | (v&0x3f);
u32 vlo;
if( (v&0xfc00)==0xd800
&& i<n-6
&& z[i+1]=='\\'
&& z[i+2]=='u'
&& ((vlo = jsonHexToInt4(z+i+3))&0xfc00)==0xdc00
){
/* We have a surrogate pair */
v = ((v&0x3ff)<<10) + (vlo&0x3ff) + 0x10000;
i += 6;
zOut[j++] = 0xf0 | (v>>18);
zOut[j++] = 0x80 | ((v>>12)&0x3f);
zOut[j++] = 0x80 | ((v>>6)&0x3f);
zOut[j++] = 0x80 | (v&0x3f);
}else{
zOut[j++] = 0xe0 | (v>>12);
zOut[j++] = 0x80 | ((v>>6)&0x3f);
zOut[j++] = 0x80 | (v&0x3f);
}
}
}else{
if( c=='b' ){