[merged] fixing build warnings/errors (missing documentation)

[mschilli87]

@retaj: Once you got the build to work for this PR (by pushing fixes to the docufix branch), please don't merge it here. Instead, I suggest to close this PR and merge docufix directly into dev2 (as it's independent of the Travis-CI stuff I'm working on in this branch). If you prefer to that that through another PR or simply by a local git checkout dev2 && git merge docufix && git branch -d docufix && git push origin dev2 doesn't matter.

---

Once this is done, I can rebase the travis branch on dev2 (incl. your docufix changes) and keep working on that before we finally merge my changes back to dev2 when I'm done testing.

[mschilli87]

There warnings still seem to be there....

@retaj: Does R CMD check work for you locally now?

[retaj]

Looks like it crashed due to some externality (a bioconductor package didn't download), I just reran the Travis build...will not merge to travis branch, just want to see it passing :)

[mschilli87]

Let's hope this is temporal and a single successful run caches all packages for the future then.

[retaj]

ran devtools::check(), it says: R CMD check results 0 errors | 0 warnings | 3 notes

notes are mostly bs...complaining about .travis.yaml file :) and variables/functions with no visible global definition (fundamentalist way to say "you guys are using data.table and ggplot2") 🥇

let's see what Travis has to say :)

[mschilli87]

still 'Undocumented code objects'...

[retaj]

No idea why R:oldrel is still crashing, it should be running 3.2.5 according to Travis doc. SummarizedExperiment depends on R >= 3.2. Whatevs, I restarted the R:oldrel build just in case it was some trivial issue - if it crashes again, feel free to drop the oldrel from .travis.yaml and set R >= 3.3 as a dependency for ciRcus :machiavelli:

[mschilli87]

Whatever the reason, SummarizedExperiment seems to by unavailable in R 3.2.5.

---

Can you adjust the dependency R >= 3.3 in the docufix branch & merge it into dev2?

I'll keep care of dropping oldrel from Travis in the travis branch.

---

I'll close this now as all the information is here.

[mschilli87]

@retaj: Regarding the notes complaining about .travis.yml:

If I understood correctly you can make R ignore it by adding the following to .Rbuildignore:

## Ignore travis config file
^\.travis\.yml$