fn not_starts_with()

in crates/iceberg/src/expr/visitors/row_group_metrics_evaluator.rs [393:460]

• 48 lines of code
• 6 McCabe index (conditional complexity)

    fn not_starts_with(
        &mut self,
        reference: &BoundReference,
        datum: &Datum,
        _predicate: &BoundPredicate,
    ) -> Result<bool> {
        let field_id = reference.field().id;

        if self.may_contain_null(field_id) {
            return ROW_GROUP_MIGHT_MATCH;
        }

        // notStartsWith will match unless all values must start with the prefix.
        // This happens when the lower and upper bounds both start with the prefix.

        let PrimitiveLiteral::String(prefix) = datum.literal() else {
            return Err(Error::new(
                ErrorKind::Unexpected,
                "Cannot use StartsWith operator on non-string values",
            ));
        };

        let Some(lower_bound) = self.min_value(field_id)? else {
            return ROW_GROUP_MIGHT_MATCH;
        };

        let PrimitiveLiteral::String(lower_bound_str) = lower_bound.literal() else {
            return Err(Error::new(
                ErrorKind::Unexpected,
                "Cannot use NotStartsWith operator on non-string lower_bound value",
            ));
        };

        if lower_bound_str < prefix {
            // if lower is shorter than the prefix then lower doesn't start with the prefix
            return ROW_GROUP_MIGHT_MATCH;
        }

        let prefix_len = prefix.chars().count();

        if lower_bound_str.chars().take(prefix_len).collect::<String>() == *prefix {
            // lower bound matches the prefix

            let Some(upper_bound) = self.max_value(field_id)? else {
                return ROW_GROUP_MIGHT_MATCH;
            };

            let PrimitiveLiteral::String(upper_bound) = upper_bound.literal() else {
                return Err(Error::new(
                    ErrorKind::Unexpected,
                    "Cannot use NotStartsWith operator on non-string upper_bound value",
                ));
            };

            // if upper is shorter than the prefix then upper can't start with the prefix
            if upper_bound.chars().count() < prefix_len {
                return ROW_GROUP_MIGHT_MATCH;
            }

            if upper_bound.chars().take(prefix_len).collect::<String>() == *prefix {
                // both bounds match the prefix, so all rows must match the
                // prefix and therefore do not satisfy the predicate
                return ROW_GROUP_CANT_MATCH;
            }
        }

        ROW_GROUP_MIGHT_MATCH
    }