Function find_common_columns

datafusion/physical-expr/src/utils/guarantee.rs:505–532

Find columns that appear in all termsets

(
    termsets: &[Vec<ColOpLitOrInList<'a>>],
)

/// Find columns that appear in all termsets
fn find_common_columns<'a>(
    termsets: &[Vec<ColOpLitOrInList<'a>>],
) -> Vec<&'a crate::expressions::Column> {
    if termsets.is_empty() {
        return Vec::new();
    }

    // Start with columns from the first termset
    let mut common_cols: HashSet<_> = termsets[0].iter().map(|term| term.col()).collect();

    // check if any common_col in one termset occur many times
    // e.g. (a = 1 AND a = 2) OR (a = 2 AND b = 3), should not infer a guarantee
    // TODO: for above case, we can infer a IN (2) AND b IN (3)
    if common_cols.len() != termsets[0].len() {
        return Vec::new();
    }

    // Intersect with columns from remaining termsets
    for termset in termsets.iter().skip(1) {
        let termset_cols: HashSet<_> = termset.iter().map(|term| term.col()).collect();
        if termset_cols.len() != termset.len() {
            return Vec::new();
        }
        common_cols = common_cols.intersection(&termset_cols).cloned().collect();
    }

    common_cols.into_iter().collect()
}

#[cfg(test)]
mod test {