Closed eddelbuettel closed 3 years ago
Late seeing this, but everything LGTM.
(Quasi-related to the HTML to PDF printing: I have a longer-term goal of building a PNAS template for pagedown, but that's in the back of the queue with a bunch of other unrealised ideas!)
Go for it :)
I consider html rendering to be semi-random and html-to-pdf a fool's errand as I don't trust browsers. Make pdf files like the man himself intended. Related: https://twitter.com/keenanisalive/status/1309826322944978945
Update to pinp
worked great when the (much) smaller and otherwise nice html vignette framework turned out to be "CRAN incomatible" as Uwe uses a pandoc.exe
version from the very dark ages. Pdf has the very strong thing going for it that I can slurp it into a Sweave document and then not depend on anything that may change or break. And CI runs are quicker as there is not phantom bloat of Depends/Imports/LinkingTo/Suggests as you don't need rmarkdown
and knitr
.
Still a shame as the CSS I use for the mini html vignette was awesome for the purpose. Ah well.
Both @coatless and I already know and cherish (given that we talked about it in a different context) that pandoc lets us use multiple columns in a document using
::::::
blocks with inner:::
blocks.But only wizard @grantmcdermott knows how to do it in general or even in one document over html and pdf (giving the credit back to @yihui 's book).
As multi-column is really good for e.g. code snippets, I recently "borrowed" the css trick for a html vignette I tried (with a currently less widely-used markdown wrapper) but that hit the wall at CRAN over pandoc 2.0 lacking on Windoze so I converted it to pdf. Using this package.
Which was lacking the multi-column setup for LaTeX so I copied it into the document header, with an added
\noindent
. That is useful, so I will add this the package now and plan to let the PR sit there for a day or so.