r.fold generates profiling data whose size is proportional to the length of the sequence folded

For example, the query r.range(100).fold(0, function (acc, item) { return acc.add(item); }) generates a massive chunk of profiling data.

Profiling data size should never be proportional to the data size.

Looks like it slows down a trivial fold (like the one above) by about 20x.

We've had similar issues with other commands before (forEach I think among others). Indeed something we need to fix.

Navigation Menu