Gáspár Nagy
Gáspár Nagy
@ajeckmans thx! I think the general direction is good in my opinion. Also it is clear how this helps for the dynamic / context specific conversion cases. To understand the...
@ajeckmans Makes sense. About `[StepArgumentTransformation]`, my question was more about the feasibility and your feelings and not necessarily whether this is going to be included in this PR & change...
> It is feasible I think to resolve the StepArgumentTransformation and then use those to create for instance a set from a table. But it would be inconsistent with the...
@ajeckmans you are right. The only reason why I would like to discuss this now (here or in the discussion), because the rollout strategy of the new feature depends on...
@ajeckmans OK, I get your point and have been thinking on it a lot now. In this case, however, my suggestion would be to delay marking the existing API with...
@ajeckmans Absolutely. In the docs we should push this!
@ajeckmans Do you plan to work on this PR this week? We could include it to v2 in that case.
> In any case, taking the latest discussion into account I think this change won't be a breaking one anymore and as such should be able to release with any...
> Due to how they're loaded into Roslyn, we ideally want a single package that contains all the analysers and code-generators with as much self-contained dependencies as possible. @Code-Grump Isn't...