Skip to content

Add MySQL REGEXP_LIKE(), REGEXP_REPLACE(), REGEXP_SUBSTR(), REGEXP_INSTR() UDFs #368

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
JanJakes wants to merge 7 commits into
base: trunk
Choose a base branch
from
Draft

Add MySQL REGEXP_LIKE(), REGEXP_REPLACE(), REGEXP_SUBSTR(), REGEXP_INSTR() UDFs #368

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
bca12b6
Add REGEXP_LIKE() UDF
JanJakes Apr 21, 2026
f6e3e88
Add REGEXP_REPLACE() UDF
JanJakes Apr 21, 2026
cdd5133
Add REGEXP_SUBSTR() UDF
JanJakes Apr 21, 2026
963e1ae
Add REGEXP_INSTR() UDF
JanJakes Apr 21, 2026
36d6b56
Pin down REGEXP_* edge cases shared across functions
JanJakes Apr 21, 2026
7a7f948
Align REGEXP operators with REGEXP functions
JanJakes Jul 28, 2026
b35d758
Enforce existing UDF argument counts
JanJakes Jul 28, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
120 changes: 108 additions & 12 deletions packages/mysql-on-sqlite/src/sqlite/class-wp-pdo-mysql-on-sqlite.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3837,12 +3837,17 @@ private function translate( $node ): ?string {
return $this->translate_sequence( $node->get_children() );
case 'simpleExprBody':
return $this->translate_simple_expr_body( $node );
case 'predicate':
$operations = $node->get_first_child_node( 'predicateOperations' );
$token = null === $operations ? null : $operations->get_first_child_token();
if ( null !== $token && WP_MySQL_Lexer::REGEXP_SYMBOL === $token->id ) {
return $this->translate_regexp_predicate( $node, $operations );
}
return $this->translate_sequence( $node->get_children() );
case 'predicateOperations':
$token = $node->get_first_child_token();
if ( WP_MySQL_Lexer::LIKE_SYMBOL === $token->id ) {
return $this->translate_like( $node );
} elseif ( WP_MySQL_Lexer::REGEXP_SYMBOL === $token->id ) {
return $this->translate_regexp_functions( $node );
}
return $this->translate_sequence( $node->get_children() );
case 'runtimeFunctionCall':
Expand Down Expand Up @@ -4478,13 +4483,16 @@ private function translate_like( WP_Parser_Node $node ): string {
/**
* Translate MySQL REGEXP expression to SQLite.
*
* @param WP_Parser_Node $node The "predicateOperations" AST node.
* @param WP_Parser_Node $node The "predicate" AST node.
* @param WP_Parser_Node $operations The "predicateOperations" AST node.
* @return string The translated value.
* @throws WP_SQLite_Driver_Exception When the translation fails.
*/
private function translate_regexp_functions( WP_Parser_Node $node ): string {
$tokens = $node->get_descendant_tokens();
private function translate_regexp_predicate( WP_Parser_Node $node, WP_Parser_Node $operations ): string {
$tokens = $operations->get_descendant_tokens();
$is_binary = isset( $tokens[1] ) && WP_MySQL_Lexer::BINARY_SYMBOL === $tokens[1]->id;
$subject = $this->translate_regexp_string_arg( $node->get_first_child_node() );
$pattern = $this->translate_regexp_string_arg( $operations->get_first_child_node() );

/*
* If the query says REGEXP BINARY, the comparison is byte-by-byte
Expand All @@ -4500,9 +4508,11 @@ private function translate_regexp_functions( WP_Parser_Node $node ): string {
* regular expressions anyway.
*/
if ( true === $is_binary ) {
return 'REGEXP CHAR(0) || ' . $this->translate( $node->get_first_child_node() );
$pattern = 'CHAR(0) || ' . $pattern;
}
return 'REGEXP ' . $this->translate( $node->get_first_child_node() );

$operator = null === $node->get_first_child_node( 'notRule' ) ? 'REGEXP' : 'NOT REGEXP';
return "$subject $operator $pattern";
}

/**
Expand Down Expand Up @@ -4570,11 +4580,10 @@ private function translate_function_call( WP_Parser_Node $node ): string {
$this->unquote_sqlite_identifier( $this->translate( $nodes[0] ) )
);

$args = array();
if ( isset( $nodes[1] ) ) {
foreach ( $nodes[1]->get_child_nodes() as $child ) {
$args[] = $this->translate( $child );
}
$arg_nodes = isset( $nodes[1] ) ? $nodes[1]->get_child_nodes() : array();
$args = array();
foreach ( $arg_nodes as $child ) {
$args[] = $this->translate( $child );
}

switch ( $name ) {
Expand Down Expand Up @@ -4668,11 +4677,98 @@ private function translate_function_call( WP_Parser_Node $node ): string {
substr( $version, 3, 2 )
);
return $this->quote_sqlite_value( $value );
case 'REGEXP_LIKE':
case 'REGEXP_REPLACE':
case 'REGEXP_SUBSTR':
case 'REGEXP_INSTR':
return sprintf(
'%s(%s)',
$name,
implode( ', ', $this->translate_regexp_function_args( $name, $arg_nodes ) )
);
default:
return $this->translate_sequence( $node->get_children() );
}
}

/**
* Translate REGEXP function arguments while preserving numeric literal text.
*
* @param string $name Function name.
* @param array $arg_nodes Function argument nodes.
*
* @return array Translated arguments.
*/
private function translate_regexp_function_args( string $name, array $arg_nodes ): array {
$string_arg_indexes = array(
'REGEXP_LIKE' => array( 0, 1, 2 ),
'REGEXP_REPLACE' => array( 0, 1, 2, 5 ),
'REGEXP_SUBSTR' => array( 0, 1, 4 ),
'REGEXP_INSTR' => array( 0, 1, 5 ),
);

$args = array();
foreach ( $arg_nodes as $index => $arg_node ) {
if ( in_array( $index, $string_arg_indexes[ $name ], true ) ) {
$args[] = $this->translate_regexp_string_arg( $arg_node );
} else {
$args[] = $this->translate( $arg_node );
}
}
return $args;
}

/**
* Translate a REGEXP string-domain argument while preserving numeric literal text.
*
* @param WP_Parser_Node $node Function argument or operator operand.
*
* @return string Translated argument.
*/
private function translate_regexp_string_arg( WP_Parser_Node $node ): string {
$numeric_literal = $this->get_regexp_numeric_literal_string( $node );
return null === $numeric_literal
? $this->translate( $node )
: $this->quote_sqlite_value( $numeric_literal );
}

/**
* Read an exact decimal or floating-point literal from an expression node.
*
* @param WP_Parser_Node $node Expression node.
*
* @return string|null Literal text, or NULL for a non-literal expression.
*/
private function get_regexp_numeric_literal_string( WP_Parser_Node $node ): ?string {
$number = null;
$sign = '';
foreach ( $node->get_descendant_tokens() as $token ) {
$value = $token->get_value();
if ( '(' === $value || ')' === $value ) {
continue;
}
if ( WP_MySQL_Lexer::BINARY_SYMBOL === $token->id ) {
continue;
}
if ( ( '+' === $value || '-' === $value ) && null === $number && '' === $sign ) {
$sign = '-' === $value ? '-' : '';
continue;
}
if (
null === $number
&& (
WP_MySQL_Lexer::DECIMAL_NUMBER === $token->id
|| WP_MySQL_Lexer::FLOAT_NUMBER === $token->id
)
) {
$number = $value;
continue;
}
return null;
}
return null === $number ? null : $sign . $number;
}

/**
* Translate a MySQL datetime literal to SQLite.
*
Expand Down
Loading
Loading