Background
So, one could be surprised by the fact that:
Hold @ Grid[{{1, 2}}]
(*or*)
Hold @ Graphics @ Disk[]
This issue and methods of prevention are discussed in Prevent graphics render inside held expression.
I need further explanation though.
Questions
Recall that the renderi...
@VitaliyKaurov Hi. Are you aware of that issue? I'm receiving e-mail with new posts in topics I've participated quite a long time before I can read them on community. So i can read a reply in e-mail but it isn't online when I follow the link.
If yes, was it ever an issue that it is not possible to pass empty graphs to functions that expect directed graphs?
That's because for empty graphs Mathematica does not distinguish between directed and undirected (or rather: empty graphs are both). The translation to igraph format treats empty graphs as directed.
There are certain rare cases when empty directed graphs are actually needed. For example, searching for empty induced subgraphs in a directed graph. Up to now I was thinking that it's just not worth bothering with this in practice, but now I'm not sure anymore.
I was considering introducing a symbol such as IGEmptyDirectedGraph which would translate to, well, an empty directed graph. I'm not happy about it though because the translation mechanism is already very complicated, with many different code paths for special cases, in an effort to get good performance ...
@Kuba I dunno, throwing a message does things like increment message counters, send the info to some message channels/lists, etc. so I think it might be more trouble than it's worth.
Why not just save them in downvalues/associations and use $Language to pick out the right one?
@R.M. :: is shorter, but I don't insist, I wouldn't use Message itself, only MyPackageTooltips::theOneOnTheRigh::Polish to show that string or with StringExpression/Template in case of slots.
@Kuba Ah, in that case I think it should be fine, I think. This info is stored on the symbol itself, so shouldn't affect anything else (or can't think of any)