Closed chad-klumb closed 4 years ago
To my surprise R CMD check
did not demand that I add el
as an argument to get.edge.attribute.active
despite it now appearing in the generic. (Apparently before vs. after the dots makes a difference in this regard.) Nonetheless, "Writing R Extensions" is pretty unequivocal,
A method must have all the arguments of the generic, including … if the generic does.
so I've added it anyway. If el
is not missing in a call to get.edge.attribute.active
, an error is now produced.
I also added a documentation entry for get.edge.attribute.active
, which apparently didn't have one before.
updated to depend on master
branch of network
@chad-klumb this is great! I'm looking over PR, reviewing docs etc this week. Do you know if reverse depends of networkDynamic have been tested with these changes?
I checked out revdeps for tsna and ndtv, seem great! I did find a broken link in the vignette (unrelated to these changes). I think we should merge this PR and then I can do a couple of tweaks like that
I have not run reverse dependencies of networkDynamic
, but let me know if you need me to.
Any objection to merging this now? We are getting various packages ready for CRAN release and it's convenient if we can test all the updates together by using master
versions of everything.
apparently package cross-reference requirements are changing with r-devel, which is what 0e7decac60ede79b16b34eba4c2a7b8a6ba7d6dd addressed