lapply在xml对象列表上

Question

this is my first post here, so please forgive any mistakes wrt the posting guidelines

I'm trying to read in xml data from pubmed, to extract data on author affiliations

each entry contains a set of nodes like so:

<AuthorList>
          <Author>
            <LastName>Serra-Blasco</LastName>
            <ForeName>Maria</ForeName>
            <Initials>M</Initials>
            <AffiliationInfo>
              <Affiliation>Department of Psychiatry, Hospital de la Santa Creu i Sant Pau, Biomedical Research Institute Sant Pau (IIB Sant Pau), Universitat Autònoma de Barcelona (UAB), Centro de Investigación Biomédica en Red de Salud Mental (CIBERSAM), Barcelona, Catalonia, Spain.</Affiliation>
            </AffiliationInfo>
          </Author>
          ...

I would like to end up with a dataframe that contains each authors name and affiliation in a row.

I tried to do this using xpathSApply to parse nodes reading "//Author", and ended up with a list of xml nodes.

Further parsing is proving to be a problem: i've written code that works on an individual element of this list;

for eg if the list is authorlist

I can extract an appropriate array for authorlist[[1]] using this function (that uses xpathSApply within the element)

But when I try to wrap lapply around this function, it gives me an error that says that it cannot perform xpathApply on a list. The exact error call is:

Error in UseMethod("xpathApply") : no applicable method for 'xpathApply' applied to an object of class "list"

I surmise that lapply calls the list subsetting with the equivalent of [i] whereas what I need is [[i]]. Is there a way around this? Or will I have to rewrite with some other rules in mind?

I'm open to rewriting (this is just some goofing around I'm doing) but this problem seemed interesting, hope you can help!

Answer 1

I prefer using the package rvest when working with html/xml files. Based on your simple example:

library(rvest)
myxml<-read_xml("author.xml")

lastname<-xml_text(xml_nodes(myxml,"LastName"))
firstname<-xml_text(xml_nodes(myxml,"ForeName"))
affiliation<-xml_text(xml_nodes(myxml,"Affiliation"))
df<-data.frame(firstname, lastname, affiliation)

If the structure of the xml file changes, then then call to data.frame command will error and some additional work is required to properly parse the file.

Answer 2

It would help to show your code that produced the error, but you could try xmlToDataFrame

url <- "http://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=pubmed&id=23620451&rettype=XML"
doc <- xmlParse(url)

xmlToDataFrame(doc["//Author"])
           LastName ForeName Initials                   AffiliationInfo
1      Serra-Blasco    Maria        M Department of Psychiatry...Spain.
2          Portella  Maria J       MJ                              <NA>
3       Gómez-Ansón  Beatriz        B                              <NA>
...

If you get nodes that have zero or many tags, I usually create a function to set missing tags to NA and a delimiter for joining multiple tags.

authors <- getNodeSet(doc, "//Author")

xpath2 <-function(x, path){
     y <- xpathSApply(x, path, xmlValue)
     ifelse(length(y)==0, NA, 
        ifelse(length(y)>1, paste(y, collapse=", "), y))
}

last <- sapply(authors, xpath2, ".//LastName")
aff <- sapply(authors, xpath2, ".//Affiliation")
data.frame(last, aff)
               last                               aff
1      Serra-Blasco Department of Psychiatry...Spain.
2          Portella                              <NA>
3       Gómez-Ansón                              <NA>