christopherbelter / pubmedxml Goto Github PK

<?xml version=\"1.0\" encoding=\"UTF-8\"?>

## clean pubmed XML returned from either the reutils or rentrez packages and save the cleaned XML to a new file
clean_api_xml <- function(infile, outfile = NULL, usedisk = FALSE) {
	if (usedisk == TRUE){
		theData <- readChar(infile, file.info(infile)$size, useBytes = TRUE)
    } else {
    	theData	 <- as.character(infile)
    }
	theData <- gsub("<?xml version=\"1.0\" encoding=\"UTF-8\"?>", "", theData, fixed = TRUE) # efetch xml results sometimes have encoding in the xml declaration
	theData <- gsub("<?xml version=\"1.0\" ?>", "", theData, fixed = TRUE)
	theData <- gsub("<!DOCTYPE PubmedArticleSet PUBLIC \"-//NLM//DTD PubMedArticle, 1st January 2019//EN\" \"https://dtd.nlm.nih.gov/ncbi/pubmed/out/pubmed_190101.dtd\">", "", theData, fixed = TRUE, useBytes = TRUE)
	theData <- gsub("<PubmedArticleSet>", "", theData, fixed = TRUE)
	theData <- gsub("</PubmedArticleSet>", "", theData, fixed = TRUE)
	theData <- gsub("<U\\+\\w{4}>", "", theData) ## note: with some files this doesn't catch everything; potential issue with <OtherAbstract> tags especially
	theData <- paste("<?xml version=\"1.0\" encoding=\"UTF-8\"?>", "<!DOCTYPE PubmedArticleSet PUBLIC \"-//NLM//DTD PubMedArticle, 1st January 2019//EN\" \"https://dtd.nlm.nih.gov/ncbi/pubmed/out/pubmed_190101.dtd\">", "<PubmedArticleSet>", theData, "</PubmedArticleSet>", sep = "\n")
	#theData <- paste(theData, "</PubmedArticleSet>")
	theData <- iconv(theData, to = "UTF-8", sub = "")
	if (usedisk == TRUE){
	    writeLines(theData, outfile, sep = " ")
    }
	return(theData)
}