Closed sckott closed 9 years ago
Please remember that the work-citation part looks different depending on how the article was imported. The BibTeX probably comes from Scopus, for a work imported from CrossRef it will look like this:
<work-citation>
<work-citation-type>formatted-unspecified</work-citation-type>
<citation>Fenner, M, 2008, 'Blogs, Wikis und Podcasts im Unterricht', <i>Biologie in unserer Zeit</i>, vol. 38, no. 5, pp. 284-286.</citation>
</work-citation>
You should parse the XML based on the work-citation-type.
Thanks for the note.
Hey Martin, dear all,
can I expect <work-external-identifier-id>
to contain DOI or are you considering others? What happens if none is provided?
sketch:
require(XML)
url <- c("http://pub.orcid.org/0000-0003-1419-2405/orcid-works/")
doc <- xmlTreeParse(url, useInternal=T)
doi <- xpathSApply(doc,"//r:work-external-identifier-id",
namespaces = (c(r= "http://www.orcid.org/ns/orcid")), xmlValue)
Najko, content in ORCID can come from several places. If imported from CrossRef, the work will have a DOI, but that might not always be true for other content, e.g. when imported from a file.
Thanks Martin, actually I struggle to parse <orcid-works>
nodes: within <orcid-work>
ORCID API exposes a node referencing to an external ID, a bibtex
within <citation>
or something else encapsulated within <citation>
. Thus, R functions might require if else
handling.
To illustrate, have a look at:
require(XML)
url <- c("http://pub.orcid.org/0000-0003-1419-2405/orcid-works/")
doc <- xmlTreeParse(url, useInternal=T)
xpathSApply(doc,"//r:work-citation//r:citation",
namespaces = (c(r= "http://www.orcid.org/ns/orcid")), xmlValue)
Do you know whether the ORCID API team seeks to better distinguish between linking to external sources (for fetching bibliographic metadata in a next step) and a format that describes bibliographic data as it is used within the digital library domain (such as bibo or MODS)?
Sorry, I do not know if rorcid
is the appropriate forum to address my issues. Please let me know if I have to consult ORCID support.
using json now, this seems fine now
Hey @cboettig, hoping you can help figure out how to parse these citations from Orcid? This is an example call using Peter Binfields Orcid ID. You have any functions from
knitcitations
that we can use to parse this? And probably use those bibtext like citations too?