swihart / repeated

Non-normal repeated measurements models
GNU General Public License v2.0
0 stars 0 forks source link

on TRAVIS-CI, "links to gnlm, growth and event" complaint #2

Open swihart opened 7 years ago

swihart commented 7 years ago

I have rmutil on CRAN. Yes!

Now I can start putting up the remaining Lindsey packages. To identify where the links are, I will uninstall (growth, gnlm, and event) from my RLIBs and re-run CMD-SHIFT-E and try to comment out or modify all the Xrefs.

swihart commented 7 years ago

In /R:

swihartbj@L01950966:~/github/repeated$ find ./R -type f -print0 | xargs -0 grep -l 'gnlm'
./R/catmiss.r
./R/chidden.r
./R/cphidden.r
./R/gar.r
./R/gausscop.r
./R/glmm.r
./R/gnlmix.r
./R/gnlmm.r
./R/gnlmm3.r
./R/hidden.r
./R/hnlmix.r
./R/kalcount.r
./R/kalseries.r
./R/logitord.r
./R/nbkal.r
swihartbj@L01950966:~/github/repeated$ find ./R -type f -print0 | xargs -0 grep -l 'growth'
./R/gar.r
./R/gausscop.r
./R/gnlmix.r
./R/hnlmix.r
./R/kalcount.r
./R/kalseries.r
swihartbj@L01950966:~/github/repeated$ find ./R -type f -print0 | xargs -0 grep -l 'event'
./R/chidden.r
./R/cmcre.r
./R/gar.r
./R/kalcount.r
./R/kalseries.r

In /man:

swihartbj@L01950966:~/github/repeated$ find ./man -type f -print0 | xargs -0 grep -l 'gnlm'
./man/catmiss.Rd
./man/chidden.Rd
./man/cphidden.Rd
./man/gar.Rd
./man/gausscop.Rd
./man/glmm.Rd
./man/gnlmix.Rd
./man/gnlmm.Rd
./man/gnlmm3.Rd
./man/hidden.Rd
./man/hnlmix.Rd
./man/kalcount.Rd
./man/kalseries.Rd
./man/logitord.Rd
./man/nbkal.Rd
swihartbj@L01950966:~/github/repeated$ find ./man -type f -print0 | xargs -0 grep -l 'growth'
./man/gar.Rd
./man/gausscop.Rd
./man/gnlmix.Rd
./man/hnlmix.Rd
./man/kalcount.Rd
./man/kalseries.Rd
swihartbj@L01950966:~/github/repeated$ find ./man -type f -print0 | xargs -0 grep -l 'event'
./man/chidden.Rd
./man/cmcre.Rd
./man/gar.Rd
./man/kalcount.Rd
./man/kalseries.Rd
swihart commented 7 years ago

Take kalseries.R and kalseries.Rd. In repeated, we are using Roxygen. Is there a way to comment out the .R so that devtools::document() just doesn't put the Xrefs into the .Rd files and then I can cycle back after all the Lindsey packages are on CRAN to uncomment? There must be some slick Github way to do this...a branch? A fork?

Contents of kalseries.R

<snip>
##' @seealso \code{\link[growth]{carma}}, \code{\link[growth]{elliptic}},
##' \code{\link[rmutil]{finterp}}, \code{\link[repeated]{gar}},
##' \code{\link[rmutil]{gettvc}}, \code{\link[repeated]{gnlmm}},
##' \code{\link[gnlm]{gnlr}}, \code{\link[rmutil]{iprofile}},
##' \code{\link[repeated]{kalcount}}, \code{\link[event]{kalsurv}},
##' \code{\link[rmutil]{mprofile}}, \code{\link[rmutil]{read.list}},
##' \code{\link[rmutil]{restovec}}, \code{\link[rmutil]{rmna}},
##' \code{\link[rmutil]{tcctomat}}, \code{\link[rmutil]{tvctomat}}.

<snip>

Change to:

<snip>
### @seealso \code{\link[growth]{carma}}, \code{\link[growth]{elliptic}},
### \code{\link[rmutil]{finterp}}, \code{\link[repeated]{gar}},
### \code{\link[rmutil]{gettvc}}, \code{\link[repeated]{gnlmm}},
### \code{\link[gnlm]{gnlr}}, \code{\link[rmutil]{iprofile}},
### \code{\link[repeated]{kalcount}}, \code{\link[event]{kalsurv}},
### \code{\link[rmutil]{mprofile}}, \code{\link[rmutil]{read.list}},
### \code{\link[rmutil]{restovec}}, \code{\link[rmutil]{rmna}},
### \code{\link[rmutil]{tcctomat}}, \code{\link[rmutil]{tvctomat}}.

<snip>

Rerun devtools::document() and see what .Rd file looks like and what ?kalseries looks like in viewer. Ok, did that. Looks feasible.

I'm going to make a branch of the current state of the package, that is before-Xrefs-fixes in hopes that I might be able to just merge those files back in when the time is appropriate.

I'll do the fixes on master and keep working on master. the before-Xrefs-fixes branch is more of a snapshot that I may want to use later, like a massive, selective "Ctrl+z".

swihart commented 7 years ago

Ok, march through the files, comment out (ctrl+f replace in selection ##' with ###, then rerun devtools::document(), then commit the change .r and .rd files to master branch). Once we do that, we re-run CMD-SHIFT-E and hopefully get rid of Xref errors.