Error when diffing list containing empty list

[wch]

It's not clear to me whether diffObj is meant to be used for complex structures like nested lists, but it seems that at it at least shouldn't give an error (or that the error is more informative). Example:

> diffObj(list(list()), list(list()))
Error in trim_sub(obj.as.chr, stripped) : 
  Logic Error: trimmed string is not a substring of orginal, contact maintainer
Error in apply_trim(target, tar.capt.p, etc@trim) : 
  `*trim` method produced an error when attempting to trim ; If you did not define custom `*trim` methods contact maintainer (see `?trim`).

[brodieG]

This is definitely a bug. It is supposed to work well with nested structures, but obviously I didn't properly handle the case you bring up. I'll fix this in the next release (and hopefully get development branch fixed today). In the meantime, you can use:

diffStr(list(list()), list(list()))

or

diffObj(list(list()), list(list()), trim=FALSE)

to work around the issue. Thanks for reporting.

[brodieG]

This should now be working in install_github('brodieg/diffobj@development'). Unfortunately I seem to have introduced some minor regressions for R 3.4.x in the process, but I don't have time to look at those now. The R release tests all pass. I am hoping to get version 0.1.6 with these fixes and others out to CRAN in the next week or two.

Let me know if for whatever reason this does not work for you.

Also, feel free to add feature request issues if you have use cases for diffobj that don't appear currently covered.