/*
** Implementation of SQLite REGEXP operator. This scalar function takes
** two arguments. The first is a regular expression pattern to compile
** the second is a string to match against that pattern. If either
** argument is an SQL NULL, then NULL Is returned. Otherwise, the result
** is 1 if the string matches the pattern, or 0 otherwise.
**
** SQLite maps the regexp() function to the regexp() operator such
** that the following two are equivalent:
**
** zString REGEXP zPattern
** regexp(zPattern, zString)
**
** Uses the following ICU regexp APIs:
**
** uregex_open()
** uregex_matches()
** uregex_close()
*/
static void icuRegexpFunc(sqlite3_context *p, int nArg, sqlite3_value **apArg){
UErrorCode status = U_ZERO_ERROR;
URegularExpression *pExpr;
UBool res;
const UChar *zString = sqlite3_value_text16(apArg[1]);
(void)nArg; /* Unused parameter */
/* If the left hand side of the regexp operator is NULL,
** then the result is also NULL.
*/
if( !zString ){
return;
}
pExpr = sqlite3_get_auxdata(p, 0);
if( !pExpr ){
const UChar *zPattern = sqlite3_value_text16(apArg[0]);
if( !zPattern ){
return;
}
pExpr = uregex_open(zPattern, -1, 0, 0, &status);
if( U_SUCCESS(status) ){
sqlite3_set_auxdata(p, 0, pExpr, icuRegexpDelete);
}else{
assert(!pExpr);
icuFunctionError(p, "uregex_open", status);
return;
}
}
/* Configure the text that the regular expression operates on. */
uregex_setText(pExpr, zString, -1, &status);
if( !U_SUCCESS(status) ){
icuFunctionError(p, "uregex_setText", status);
return;
}
/* Attempt the match */
res = uregex_matches(pExpr, 0, &status);
if( !U_SUCCESS(status) ){
icuFunctionError(p, "uregex_matches", status);
return;
}
/* Set the text that the regular expression operates on to a NULL
** pointer. This is not really necessary, but it is tidier than
** leaving the regular expression object configured with an invalid
** pointer after this function returns.
*/
uregex_setText(pExpr, 0, 0, &status);
/* Return 1 or 0. */
sqlite3_result_int(p, res ? 1 : 0);
}
/*
** Implementation of the scalar function icu_load_collation().
**
** This scalar function is used to add ICU collation based collation
** types to an SQLite database connection. It is intended to be called
** as follows:
**
** SELECT icu_load_collation(<locale>, <collation-name>);
**
** Where <locale> is a string containing an ICU locale identifier (i.e.
** "en_AU", "tr_TR" etc.) and <collation-name> is the name of the
** collation sequence to create.
*/
static void icuLoadCollation(
sqlite3_context *p,
int nArg,
sqlite3_value **apArg
){
sqlite3 *db = (sqlite3 *)sqlite3_user_data(p);
UErrorCode status = U_ZERO_ERROR;
const char *zLocale; /* Locale identifier - (eg. "jp_JP") */
const char *zName; /* SQL Collation sequence name (eg. "japanese") */
UCollator *pUCollator; /* ICU library collation object */
int rc; /* Return code from sqlite3_create_collation_x() */
assert(nArg==2);
zLocale = (const char *)sqlite3_value_text(apArg[0]);
zName = (const char *)sqlite3_value_text(apArg[1]);
if( !zLocale || !zName ){
return;
}
pUCollator = ucol_open(zLocale, &status);
if( !U_SUCCESS(status) ){
icuFunctionError(p, "ucol_open", status);
return;
}
assert(p);
rc = sqlite3_create_collation_v2(db, zName, SQLITE_UTF16, (void *)pUCollator,
icuCollationColl, icuCollationDel
);
if( rc!=SQLITE_OK ){
ucol_close(pUCollator);
sqlite3_result_error(p, "Error registering collation function", -1);
}
}
/*
** Implementations of scalar functions for case mapping - upper() and
** lower(). Function upper() converts its input to upper-case (ABC).
** Function lower() converts to lower-case (abc).
**
** ICU provides two types of case mapping, "general" case mapping and
** "language specific". Refer to ICU documentation for the differences
** between the two.
**
** To utilise "general" case mapping, the upper() or lower() scalar
** functions are invoked with one argument:
**
** upper('ABC') -> 'abc'
** lower('abc') -> 'ABC'
**
** To access ICU "language specific" case mapping, upper() or lower()
** should be invoked with two arguments. The second argument is the name
** of the locale to use. Passing an empty string ("") or SQL NULL value
** as the second argument is the same as invoking the 1 argument version
** of upper() or lower().
**
** lower('I', 'en_us') -> 'i'
** lower('I', 'tr_tr') -> '\u131' (small dotless i)
**
** http://www.icu-project.org/userguide/posix.html#case_mappings
*/
static void icuCaseFunc16(sqlite3_context *p, int nArg, sqlite3_value **apArg){
const UChar *zInput; /* Pointer to input string */
UChar *zOutput = 0; /* Pointer to output buffer */
int nInput; /* Size of utf-16 input string in bytes */
int nOut; /* Size of output buffer in bytes */
int cnt;
int bToUpper; /* True for toupper(), false for tolower() */
UErrorCode status;
const char *zLocale = 0;
assert(nArg==1 || nArg==2);
bToUpper = (sqlite3_user_data(p)!=0);
if( nArg==2 ){
zLocale = (const char *)sqlite3_value_text(apArg[1]);
}
zInput = sqlite3_value_text16(apArg[0]);
if( !zInput ){
return;
}
nOut = nInput = sqlite3_value_bytes16(apArg[0]);
if( nOut==0 ){
sqlite3_result_text16(p, "", 0, SQLITE_STATIC);
return;
}
for(cnt=0; cnt<2; cnt++){
UChar *zNew = sqlite3_realloc(zOutput, nOut);
if( zNew==0 ){
sqlite3_free(zOutput);
sqlite3_result_error_nomem(p);
return;
}
zOutput = zNew;
status = U_ZERO_ERROR;
if( bToUpper ){
nOut = 2*u_strToUpper(zOutput,nOut/2,zInput,nInput/2,zLocale,&status);
}else{
nOut = 2*u_strToLower(zOutput,nOut/2,zInput,nInput/2,zLocale,&status);
}
if( U_SUCCESS(status) ){
sqlite3_result_text16(p, zOutput, nOut, xFree);
}else if( status==U_BUFFER_OVERFLOW_ERROR ){
assert( cnt==0 );
continue;
}else{
icuFunctionError(p, bToUpper ? "u_strToUpper" : "u_strToLower", status);
}
return;
}
assert( 0 ); /* Unreachable */
}
/*
** Implementations of scalar functions for case mapping - upper() and
** lower(). Function upper() converts its input to upper-case (ABC).
** Function lower() converts to lower-case (abc).
**
** ICU provides two types of case mapping, "general" case mapping and
** "language specific". Refer to ICU documentation for the differences
** between the two.
**
** To utilise "general" case mapping, the upper() or lower() scalar
** functions are invoked with one argument:
**
** upper('ABC') -> 'abc'
** lower('abc') -> 'ABC'
**
** To access ICU "language specific" case mapping, upper() or lower()
** should be invoked with two arguments. The second argument is the name
** of the locale to use. Passing an empty string ("") or SQL NULL value
** as the second argument is the same as invoking the 1 argument version
** of upper() or lower().
**
** lower('I', 'en_us') -> 'i'
** lower('I', 'tr_tr') -> 'ı' (small dotless i)
**
** http://www.icu-project.org/userguide/posix.html#case_mappings
*/
static void icuCaseFunc16(sqlite3_context *p, int nArg, sqlite3_value **apArg){
const UChar *zInput;
UChar *zOutput;
int nInput;
int nOutput;
UErrorCode status = U_ZERO_ERROR;
const char *zLocale = 0;
assert(nArg==1 || nArg==2);
if( nArg==2 ){
zLocale = (const char *)sqlite3_value_text(apArg[1]);
}
zInput = sqlite3_value_text16(apArg[0]);
if( !zInput ){
return;
}
nInput = sqlite3_value_bytes16(apArg[0]);
nOutput = nInput * 2 + 2;
zOutput = sqlite3_malloc(nOutput);
if( !zOutput ){
return;
}
if( sqlite3_user_data(p) ){
u_strToUpper(zOutput, nOutput/2, zInput, nInput/2, zLocale, &status);
}else{
u_strToLower(zOutput, nOutput/2, zInput, nInput/2, zLocale, &status);
}
if( !U_SUCCESS(status) ){
icuFunctionError(p, "u_strToLower()/u_strToUpper", status);
return;
}
sqlite3_result_text16(p, zOutput, -1, xFree);
}
