We can't do lazy size estimates after crossmatching since we drop metadata from one catalog. It would be very hard to estimate the size anyway due to how crossmatching works in general, but we could at least warn user with estimates of how much data will be downloaded or read from I/O from the total size of partitions of input catalogs. Blocked by #71
We can't do lazy size estimates after crossmatching since we drop metadata from one catalog. It would be very hard to estimate the size anyway due to how crossmatching works in general, but we could at least warn user with estimates of how much data will be downloaded or read from I/O from the total size of partitions of input catalogs. Blocked by #71