rowwithout

Rowwithout is a data-processing operation that yields a new tabular dataset by removing rows that satisfy a given condition or belong to a specified set of row identifiers. The term emphasizes exclusion: the resulting table contains all rows not marked for removal.

In typical usage, rowwithout is expressed as rowwithout(dataset, predicate) or rowwithout(dataset, indices). The predicate is a boolean expression evaluated on each row; rows for which the predicate is false are retained, while those for which it is true are dropped. When using indices, rows whose position is listed are omitted from the result. Implementations may support combining both approaches, or providing a complementary function such as rowfilter for the inclusive case.

Examples illustrate common patterns. For a table named sales, rowwithout(sales, region == 'West') returns all rows not in the West region. For a dataframe df, rowwithout(df, [3, 5, 9]) excludes the rows with indices 3, 5, and 9. In some systems, rowwithout can be chained with other operations, enabling exclusions to be applied before or after transformations like aggregation or joins.

Performance and semantics depend on the language or framework. Rowwithout can be implemented with eager filtering or lazy evaluation, and its efficiency is influenced by how predicates are evaluated and how row indices are stored. It is related to, but distinct from, similar operations such as rowfilter (inclusion) and drop or subset functions that perform equivalent exclusions in different naming conventions. See also: rowfilter, drop, subset.