documents fail on bookdown 0.23

[bob-carpenter]

Summary:

After upgrading the R package bookdown to the 0.23 (the latest), our documents fail to build. I know it builds in 0.18 as that's the previous version I had. See below for the gory details.

This might be an issue with bookdown, but I don't see that anywhere.

Description:

~/github/stan-dev/docs (feature/issue-375-complex-numbers)$ python build.py 2 28 reference-manual html

processing file: _main.Rmd
  |......................................................................| 100%
   inline R code fragments


output file: _main.knit.md

/usr/local/bin/pandoc +RTS -K512m -RTS _main.utf8.md --to html4 --from markdown+autolink_bare_uris+tex_math_single_backslash+smart --output _main.html --email-obfuscation none --wrap preserve --standalone --section-divs --table-of-contents --toc-depth 3 --template /Library/Frameworks/R.framework/Versions/3.6/Resources/library/bookdown/templates/gitbook.html --highlight-style pygments --number-sections --css ../stan-manual.css --include-in-header /var/folders/b3/6h4t41094vz281j7nn48ws900001jv/T//RtmpRHfYjh/rmarkdown-strd100676522e8.html --mathjax --lua-filter /Library/Frameworks/R.framework/Versions/3.6/Resources/library/rmarkdown/rmd/lua/pagebreak.lua --lua-filter /Library/Frameworks/R.framework/Versions/3.6/Resources/library/rmarkdown/rmd/lua/latex-div.lua --filter /usr/local/bin/pandoc-citeproc 

Output created: _book/index.html
[1] "/Users/bcarpenter/github/stan-dev/docs/src/reference-manual/_book/index.html"

and then

> sessionInfo()
R version 3.6.2 (2019-12-12)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS  10.16

Matrix products: default
BLAS:   /Library/Frameworks/R.framework/Versions/3.6/Resources/lib/libRblas.0.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/3.6/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] bookdown_0.18

loaded via a namespace (and not attached):
[1] compiler_3.6.2 tools_3.6.2    knitr_1.28     xfun_0.12 

but then after updating:

> install.packages('bookdown')
--- Please select a CRAN mirror for use in this session ---
...

I get this

> sessionInfo()
R version 3.6.2 (2019-12-12)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS  10.16

Matrix products: default
BLAS:   /Library/Frameworks/R.framework/Versions/3.6/Resources/lib/libRblas.0.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/3.6/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] bookdown_0.23

loaded via a namespace (and not attached):
[1] compiler_3.6.2 tools_3.6.2    knitr_1.33     xfun_0.25    

but the docs fail to build with this error:

~/github/stan-dev/docs (feature/issue-375-complex-numbers)$ python build.py 2 28 reference-manual html
Error: $ operator is invalid for atomic vectors
Execution halted

Command Rscript -e "bookdown::render_book('index.Rmd', output_format='bookdown::gitbook')" failed

Traceback (most recent call last):
  File "build.py", line 113, in <module>
    main()
  File "build.py", line 109, in main
    make_docs(docspath, stan_version, doc, formats)
  File "build.py", line 54, in make_docs
    shexec(command)
  File "build.py", line 35, in shexec
    raise Exception(returncode)
Exception: 1

Nothing informative in the error other than the line I spaced out. Is there a way to figure out what's going wrong here?

Additional Information:

Provide any additional information here.

Current Version:

v2.18.0

[mitzimorris]

https://stackoverflow.com/questions/68806645/bookdown-0-22-0-23-error-operator-is-invalid-for-atomic-vectors

problem is a bookdown problem. we could pin the version to 0.22. or follow other suggestions in the SO discussion.

[bob-carpenter]

@mitzimorris: How do we pin a version of an R package? If the last suggestion in the StackOverflow discussion works for the latest bookdown, I think we should go with that. My reasoning is that it's easier to keep an R package upgraded than downgraded.

[mitzimorris]

global namespace of R will make pinning version difficult.
the problem is that the devs at RStudio introduced a breaking change to the yaml config ("_output.yml").
we need to change the header from:

search: TRUE

to

search:
    engine: lunr

(or search engine fuse - they're saying go with fuse).

[WardBrian]

Closed by #390