1. It would be really great to be able to capture input/output parameters for some of the calls; I’d like to be able to specify what calls to track and then would get pairs of input/output parameters to be shown under the line (similar to the “# spent” line you have). Ideally, I’d be allowed to set up a callback to do normalization of input parameters to handle references, similar to what Memoize does. I understand that all that processing will impact execution time, but that’s okay. This is probably similar to what DashProfiler does and I’d like to see (potentially) better integration between the two. I’d prefer to see the results presented on one report, rather than capturing input/output myself and then trying to merge the two report together.

2. It would be also great to have a tree-view of the calls. I’d be able to set up a threshold and would see all (sub-)calls that (combined) took more time than threshold. I’d get a tree like:

Module A 5s
Module B 2s
Module C 1s
Module D 0.8s

and so on. All this in a collapsible/expandable tree ;) with links to appropriate places (like you already have in your table-based view). If this is not doable any time soon, it would at least be useful to highlight those calls that are above a threshold in the regular view (to make them stand out).

Thanks much for the great tool.

Paul.