Parse single entry from string

[dmi3kno]

I am fetching citations from Zotero with BetterBibtex-interfacing {rbbt} by @paleolimbot. I would love to be able to edit bibtex entry before it gets written to ".bib" file, i.e. while it is still in string format.

Ideally I would do that with {RefManageR}, but currently {RefManageR} relies on {bibtex} to parse .bib. If {bibtex} had an exposed function for reading literal bib string, then I could use {RefManageR} to edit it. Would you consider adding a parser for literal bibtex?

my_ref <- " @book{McElreath_2020, edition={2}, 
   title={Statistical Rethinking: A Bayesian Course with Examples in R and Stan}, ISBN={978-0-429-02960-8},
   url={https://www.taylorfrancis.com/books/9780429642319}, DOI={10.1201/9780429029608}, 
   publisher={Chapman and Hall/CRC}, author={McElreath, Richard}, year={2020}, month={Mar} }"

my_bibentry <- read.bib(text=my_ref)

It needs to be vectorized, of course, i.e. accepting character vectors of length()>1.

Thank you for providing such an important infrastructure package for bibliography management infrastructure in R.

[dmi3kno]

Related to #1

[trashbirdecology]

@dmi3kno, I too came here to try to find guidance on reading in .bib files, but it looks like read.bib is designed to only read .bib from inside packages, and not from .bib files directly.

See package 'bibliometrix' for reading in bib from specific sources, but I am still trying to find someone who wrote code to read in basic bib.
https://warin.ca/shiny/bibliometrix/#section-import-convert

[dmi3kno]

There are some names and addresses in the vignettes. I still believe it is possible to wrap the underlying C library in such a way that would make it possible to read a character vector with bib content (from file from disk). At least generalizing the connection (to enable textConnection, as Romain implies in #1).

[dieghernan]

Hi:

You can write the string to a temporal bib file and read it with read.bib(). This is already implemented on a test:

bibtex/tests/testthat/test-bibtex.R

Lines 14 to 20 in 1c0e12d

	test_that("read.bib can use ? in key (#9)", {
	tmp <- tempfile(fileext = ".bib")
	entry <- "@Misc{key?,\n author = \"Smith, Bob\",\n title = \"The Title\",\n year = 2012, \n}"
	writeLines(entry, tmp)
	out <- read.bib(tmp)
	expect_equal( names(out), "key?" )
	})

And a full reprex with your string:

my_ref <- " @book{McElreath_2020, edition={2},
   title={Statistical Rethinking: A Bayesian Course with Examples in R and Stan}, ISBN={978-0-429-02960-8},
   url={https://www.taylorfrancis.com/books/9780429642319}, DOI={10.1201/9780429029608},
   publisher={Chapman and Hall/CRC}, author={McElreath, Richard}, year={2020}, month={Mar} }"

tmp1 <- tempfile(fileext = ".bib")

writeLines(my_ref, tmp1)

library(bibtex)

bib <- read.bib(tmp1)

bib
#> McElreath R (2020). _Statistical Rethinking: A Bayesian Course with
#> Examples in R and Stan_, 2 edition. Chapman and Hall/CRC. ISBN
#> 978-0-429-02960-8, doi: 10.1201/9780429029608 (URL:
#> https://doi.org/10.1201/9780429029608), <URL:
#> https://www.taylorfrancis.com/books/9780429642319>.

toBibtex(bib)
#> @Book{McElreath_2020,
#>   edition = {2},
#>   title = {Statistical Rethinking: A Bayesian Course with Examples in R and Stan},
#>   isbn = {978-0-429-02960-8},
#>   url = {https://www.taylorfrancis.com/books/9780429642319},
#>   doi = {10.1201/9780429029608},
#>   publisher = {Chapman and Hall/CRC},
#>   author = {Richard McElreath},
#>   year = {2020},
#>   month = {Mar},
#> }

^{Created on 2021-12-09 by the reprex package (v2.0.1)}