jnm236;8485 wrote:
What's the risk? Worst case, the column is unusable. But the column wasn't available in the first place.
The data you're asking for isn't sitting in the UI where it can be easily obtained.
It would need to be extracted and parsed at test runtime because it relies on runtime exception objects. Which means it needs to be obtained for every test exception for every user, regardless of whether they are using the column or not.
The data needs to be piped up from the runtime environment, through the grid nodes and compressed over a network, through the engine, through the IPC barrier between the engine and the UI. There would need to be handling for edge cases such as stack traces written in different languages (i.e. German etc), or people using derived types of AggregateException ... these are the issues that on first glance I already know about, there are likely to be additional unknown unknowns.
It's not a 'big deal', but we're no longer looking at simply just adding a small piece of data to a column. To justify making these changes and carrying their weight through the lifetime of the product, I'd need to be sure there is a good number of people that would make use of the feature.