Open jb388 opened 2 years ago
The thing that has been preventing this has (in my mind, but I am absolutely no expert!) is that the pages are in different branches of the repository. If there's a way to put the files into two different branches during the build, then this should be an easy fix!
In terms of the characters and regex, I'm sure it isn't/can't be our top priority (maybe we can put it on the plate of the next-phase ISRaD manager).
Right, makes sense. It's pretty easy to pull graphics/etc. from other branches, but I'm not sure about whole .md files. Can you insert an .md file into an .md file?!?
Realistically I think the best bet would be to create a separate Rmd file that can be knit to a github-friendly .md file that contains our needed elements (headers, etc.). As a test, I pasted in the code for printing the ISRaD map from the gh-pages branch into Credits.Rmd, and the output credits.md file generates the image when you look at it on github. So anyway, I think it is doable, but it will require a bit of work.
I'll leave this issue up for now as a reminder of something to address, but in the meanwhile lets just copy and paste the updated file manually.
Thanks for updating this manually for now. And agreed with Shane. Not worried about the character formatting - it would be great to have it stay up to date automatically if that were the only tradeoff.
Also, something else we should update for the credits is mentioning the other compilations we've now included, which are synthesis papers by Claire Treat (https://agupubs.onlinelibrary.wiley.com/doi/10.1002/2015JG003061) and Cristian Estop-Aragones (https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2020GB006672)
@aahoyt Yes, thanks for the reminder about the Treat and Estop-Aragones syntheses. They are credited on the webpage, it was just the internal document that didn't include them (based on an old template). When I updated the webpage document I only did the individual entries, so all four syntheses are still credited.
I looked into it a little more and I think it should be easy enough to make a new github flavor markdown document with the R code on the gh-pages branch that can run the code and look nice. I will tackle it later in the week.
@ShaneStoner @aahoyt
Question about how the credits page of the website is generated. I created a script that generates nicely formatted .md and .html files of the ISRaD credits, but not sure of the best way to make sure those updated files make it to the webpage. Obviously the simplest way is cut & paste, but if we could have the webpage simply point to the relevant file (credits.md) on the master branch that would be both more robust and easier.
The output is not perfect, but it's good. I wrote some simple regex scripts to fix the "updelta" (upgreek characters not rendering) issue we had been having for a while now. But for some reason citations from the journal Biogeosciences still come out with weird characters: specifically, sub/superscripts generate a character printed as "mathsemicolon". Since it's a PITA to use regex to find and fix those (I'm terrible with regex), I extend the invitation to fix it to whomever is willing. Otherwise, I'm fine with it the way it is.