pub fn materialize_all_views(
lakehouse: Arc<LakehouseContext>,
views: Arc<Vec<Arc<dyn View>>>,
insert_range: TimeRange,
partition_time_delta: TimeDelta,
) -> impl Future<Output = Result<()>>Expand description
Materializes all views within a given time range.
This function iterates through the provided views, materializing partitions
for each view within the specified insert_range and partition_time_delta.
Per-view failures are isolated: a view whose materialize_partition_range
call errors is logged, counted (materialize_view_failure), and skipped —
every other view, in the same update group or a later one, still gets its
own materialization attempt this pass. This is a deliberate cross-group
policy (“continue anyway”): materialization is idempotent and re-attempted
every tick, so letting later groups proceed even after an earlier-group
failure can’t corrupt anything, and it avoids reintroducing the
group-granularity starvation this isolation is meant to fix. If any view
failed, the pass still returns a single aggregated Err listing every
failed view, so CronTask/log_task_result continues to record the pass
as failed.