Skip to content

feat: scaffolding work for expression #177

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

Merged
merged 2 commits into from
Oct 9, 2025
Merged

feat: scaffolding work for expression #177

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
876cd9f
feat: scaffolding work for expression
wgtmac Aug 15, 2025
363f01e
negate and/or/true/false
wgtmac Oct 9, 2025
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
3 changes: 3 additions & 0 deletions src/iceberg/CMakeLists.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,10 @@ set(ICEBERG_INCLUDES "$<BUILD_INTERFACE:${PROJECT_BINARY_DIR}/src>"
set(ICEBERG_SOURCES
catalog/in_memory_catalog.cc
expression/expression.cc
expression/expressions.cc
expression/literal.cc
expression/predicate.cc
expression/term.cc
file_reader.cc
file_writer.cc
inheritable_metadata.cc
Expand Down
120 changes: 110 additions & 10 deletions src/iceberg/expression/expression.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,6 +20,10 @@
#include "iceberg/expression/expression.h"

#include <format>
#include <utility>

#include "iceberg/util/formatter_internal.h"
#include "iceberg/util/macros.h"

namespace iceberg {

Expand All @@ -29,15 +33,15 @@ const std::shared_ptr<True>& True::Instance() {
return instance;
}

std::shared_ptr<Expression> True::Negate() const { return False::Instance(); }
Result<std::shared_ptr<Expression>> True::Negate() const { return False::Instance(); }

// False implementation
const std::shared_ptr<False>& False::Instance() {
static const std::shared_ptr<False> instance = std::shared_ptr<False>(new False());
return instance;
}

std::shared_ptr<Expression> False::Negate() const { return True::Instance(); }
Result<std::shared_ptr<Expression>> False::Negate() const { return True::Instance(); }

// And implementation
And::And(std::shared_ptr<Expression> left, std::shared_ptr<Expression> right)
Expand All @@ -47,11 +51,11 @@ std::string And::ToString() const {
return std::format("({} and {})", left_->ToString(), right_->ToString());
}

std::shared_ptr<Expression> And::Negate() const {
Result<std::shared_ptr<Expression>> And::Negate() const {
// De Morgan's law: not(A and B) = (not A) or (not B)
auto left_negated = left_->Negate();
auto right_negated = right_->Negate();
return std::make_shared<Or>(left_negated, right_negated);
ICEBERG_ASSIGN_OR_RAISE(auto left_negated, left_->Negate());
ICEBERG_ASSIGN_OR_RAISE(auto right_negated, right_->Negate());
return std::make_shared<Or>(std::move(left_negated), std::move(right_negated));
}

bool And::Equals(const Expression& expr) const {
Expand All @@ -71,11 +75,11 @@ std::string Or::ToString() const {
return std::format("({} or {})", left_->ToString(), right_->ToString());
}

std::shared_ptr<Expression> Or::Negate() const {
Result<std::shared_ptr<Expression>> Or::Negate() const {
// De Morgan's law: not(A or B) = (not A) and (not B)
auto left_negated = left_->Negate();
auto right_negated = right_->Negate();
return std::make_shared<And>(left_negated, right_negated);
ICEBERG_ASSIGN_OR_RAISE(auto left_negated, left_->Negate());
ICEBERG_ASSIGN_OR_RAISE(auto right_negated, right_->Negate());
return std::make_shared<And>(std::move(left_negated), std::move(right_negated));
}

bool Or::Equals(const Expression& expr) const {
Expand All @@ -87,4 +91,100 @@ bool Or::Equals(const Expression& expr) const {
return false;
}

std::string_view ToString(Expression::Operation op) {
switch (op) {
case Expression::Operation::kAnd:
return "AND";
case Expression::Operation::kOr:
return "OR";
case Expression::Operation::kTrue:
return "TRUE";
case Expression::Operation::kFalse:
return "FALSE";
case Expression::Operation::kIsNull:
return "IS_NULL";
case Expression::Operation::kNotNull:
return "NOT_NULL";
case Expression::Operation::kIsNan:
return "IS_NAN";
case Expression::Operation::kNotNan:
return "NOT_NAN";
case Expression::Operation::kLt:
return "LT";
case Expression::Operation::kLtEq:
return "LT_EQ";
case Expression::Operation::kGt:
return "GT";
case Expression::Operation::kGtEq:
return "GT_EQ";
case Expression::Operation::kEq:
return "EQ";
case Expression::Operation::kNotEq:
return "NOT_EQ";
case Expression::Operation::kIn:
return "IN";
case Expression::Operation::kNotIn:
return "NOT_IN";
case Expression::Operation::kStartsWith:
return "STARTS_WITH";
case Expression::Operation::kNotStartsWith:
return "NOT_STARTS_WITH";
case Expression::Operation::kCount:
return "COUNT";
case Expression::Operation::kNot:
return "NOT";
case Expression::Operation::kCountStar:
return "COUNT_STAR";
case Expression::Operation::kMax:
return "MAX";
case Expression::Operation::kMin:
return "MIN";
}
std::unreachable();
}

Result<Expression::Operation> Negate(Expression::Operation op) {
switch (op) {
case Expression::Operation::kIsNull:
return Expression::Operation::kNotNull;
case Expression::Operation::kNotNull:
return Expression::Operation::kIsNull;
case Expression::Operation::kIsNan:
return Expression::Operation::kNotNan;
case Expression::Operation::kNotNan:
return Expression::Operation::kIsNan;
case Expression::Operation::kLt:
return Expression::Operation::kGtEq;
case Expression::Operation::kLtEq:
return Expression::Operation::kGt;
case Expression::Operation::kGt:
return Expression::Operation::kLtEq;
case Expression::Operation::kGtEq:
return Expression::Operation::kLt;
case Expression::Operation::kEq:
return Expression::Operation::kNotEq;
case Expression::Operation::kNotEq:
return Expression::Operation::kEq;
case Expression::Operation::kIn:
return Expression::Operation::kNotIn;
case Expression::Operation::kNotIn:
return Expression::Operation::kIn;
case Expression::Operation::kStartsWith:
return Expression::Operation::kNotStartsWith;
case Expression::Operation::kNotStartsWith:
return Expression::Operation::kStartsWith;
case Expression::Operation::kTrue:
case Expression::Operation::kFalse:
case Expression::Operation::kNot:
Comment on lines 176 to 184
Copy link
Member

@zeroshade zeroshade Sep 23, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

shouldn't there be negations for these?

Copy link
Member Author

@wgtmac wgtmac Sep 24, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm following the Java impl: https://github.com/apache/iceberg/blob/main/api/src/main/java/org/apache/iceberg/expressions/Expression.java#L64-L97

case Expression::Operation::kCountStar:
case Expression::Operation::kMax:
case Expression::Operation::kMin:
case Expression::Operation::kAnd:
case Expression::Operation::kOr:
Copy link
Contributor

@Fokko Fokko Oct 7, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is interesting, we can negate and and or through the Morgan's law

Copy link
Member Author

@wgtmac wgtmac Oct 9, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Done!

case Expression::Operation::kCount:
return InvalidArgument("No negation for operation: {}", op);
}
std::unreachable();
}

} // namespace iceberg
25 changes: 16 additions & 9 deletions src/iceberg/expression/expression.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,13 +25,14 @@
#include <memory>
#include <string>

#include "iceberg/exception.h"
#include "iceberg/iceberg_export.h"
#include "iceberg/result.h"
#include "iceberg/util/formattable.h"

namespace iceberg {

/// \brief Represents a boolean expression tree.
class ICEBERG_EXPORT Expression {
class ICEBERG_EXPORT Expression : public util::Formattable {
public:
/// Operation types for expressions
enum class Operation {
Expand Down Expand Up @@ -66,8 +67,8 @@ class ICEBERG_EXPORT Expression {
virtual Operation op() const = 0;

/// \brief Returns the negation of this expression, equivalent to not(this).
virtual std::shared_ptr<Expression> Negate() const {
throw IcebergError("Expression cannot be negated");
virtual Result<std::shared_ptr<Expression>> Negate() const {
return NotSupported("Expression cannot be negated");
}

/// \brief Returns whether this expression will accept the same values as another.
Expand All @@ -78,7 +79,7 @@ class ICEBERG_EXPORT Expression {
return false;
}

virtual std::string ToString() const { return "Expression"; }
std::string ToString() const override { return "Expression"; }
};

/// \brief An Expression that is always true.
Expand All @@ -93,7 +94,7 @@ class ICEBERG_EXPORT True : public Expression {

std::string ToString() const override { return "true"; }

std::shared_ptr<Expression> Negate() const override;
Result<std::shared_ptr<Expression>> Negate() const override;

bool Equals(const Expression& other) const override {
return other.op() == Operation::kTrue;
Expand All @@ -113,7 +114,7 @@ class ICEBERG_EXPORT False : public Expression {

std::string ToString() const override { return "false"; }

std::shared_ptr<Expression> Negate() const override;
Result<std::shared_ptr<Expression>> Negate() const override;

bool Equals(const Expression& other) const override {
return other.op() == Operation::kFalse;
Expand Down Expand Up @@ -149,7 +150,7 @@ class ICEBERG_EXPORT And : public Expression {

std::string ToString() const override;

std::shared_ptr<Expression> Negate() const override;
Result<std::shared_ptr<Expression>> Negate() const override;

bool Equals(const Expression& other) const override;

Expand Down Expand Up @@ -184,7 +185,7 @@ class ICEBERG_EXPORT Or : public Expression {

std::string ToString() const override;

std::shared_ptr<Expression> Negate() const override;
Result<std::shared_ptr<Expression>> Negate() const override;

bool Equals(const Expression& other) const override;

Expand All @@ -193,4 +194,10 @@ class ICEBERG_EXPORT Or : public Expression {
std::shared_ptr<Expression> right_;
};

/// \brief Returns a string representation of an expression operation.
ICEBERG_EXPORT std::string_view ToString(Expression::Operation op);

/// \brief Returns the negated operation.
ICEBERG_EXPORT Result<Expression::Operation> Negate(Expression::Operation op);

} // namespace iceberg
Loading
Loading