Vcat broadcast to all columns?

[GHTaarn]

@user664303

My intention was to propose an addition to the functionality of vcat (and other concatenation functions): to broadcast scalars across columns. I get that some of you might not see a need for this in your line of work, but it is definitely useful in projective geometry applications. PaddedView is probably good enough for my purpose, though the issues around mutability and having to call collect() make it not as flexible.

To come at my suggestion from a different angle, are there any downsides to vcat broadcasting scalars?

I have a few comments about your proposal for additional functionality:

1. The “General Usage” category is normally used for here and now solutions to peoples problems, proposals for changes to Julia are better to post in the “Internals” category or in the issue tracker on Github
2. A lot of people have ideas for improving Julia and many post them, the core developers have to be very very selective just for the sake of keeping the language clean and and for the good proposals they have to select the few that are the best use of their limited resources unless the person that proposes them implements them him or herself.
3. My own opinion about your proposal is that it does not really feel right. It is mixing two functionalities into one call that should not be mixed in that way. I would prefer changing the resize function to be able to handle multi dimensional arrays, but this has already been proposed long ago and Stefan Karpinsky was not in favor of this See this thread

The choice is yours, but I would suggest looking in external packages for the functionality that you desire or use some of the other suggestions in this thread. You are not alone in wishing for more functionality than Julia has to offer.