diff ChemspiderCompound.R @ 0:3afe41d3e9e7 draft

planemo upload for repository https://github.com/workflow4metabolomics/lcmsmatching.git commit bb4d3e23d99828bfee16d31d794c49a17313ec2f

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/ChemspiderCompound.R	Mon Jul 11 09:12:03 2016 -0400
@@ -0,0 +1,59 @@
+if ( ! exists('ChemspiderCompound')) { # Do not load again if already loaded
+
+	source('BiodbEntry.R')
+	
+	#####################
+	# CLASS DECLARATION #
+	#####################
+	
+	ChemspiderCompound <- setRefClass("ChemspiderCompound", contains = "BiodbEntry")
+	
+	###########
+	# FACTORY #
+	###########
+	
+	createChemspiderCompoundFromHtml <- function(contents, drop = TRUE) {
+
+		library(XML)
+
+		compounds <- list()
+
+		# Define xpath expressions
+		xpath.expr <- character()
+
+		for (html in contents) {
+
+			# Create instance
+			compound <- ChemspiderCompound$new()
+		
+			# Parse HTML
+			xml <-  htmlTreeParse(html, asText = TRUE, useInternalNodes = TRUE)
+
+			# Test generic xpath expressions
+			for (field in names(xpath.expr)) {
+				v <- xpathSApply(xml, xpath.expr[[field]], xmlValue)
+				if (length(v) > 0)
+					compound$setField(field, v)
+			}
+		
+			# Get accession
+			accession <- xpathSApply(xml, "//li[starts-with(., 'ChemSpider ID')]", xmlValue)
+			if (length(accession) > 0) {
+				accession <- sub('^ChemSpider ID([0-9]+)$', '\\1', accession, perl = TRUE)
+				compound$setField(RBIODB.ACCESSION, accession)
+			}
+
+			compounds <- c(compounds, compound)
+		}
+
+		# Replace elements with no accession id by NULL
+		compounds <- lapply(compounds, function(x) if (is.na(x$getField(RBIODB.ACCESSION))) NULL else x)
+
+		# If the input was a single element, then output a single object
+		if (drop && length(contents) == 1)
+			compounds <- compounds[[1]]
+	
+		return(compounds)
+	}
+}
+