comparison PubchemCompound.R @ 1:253d531a0193 draft

planemo upload for repository https://github.com/workflow4metabolomics/lcmsmatching.git commit 36c9d8099c20a1ae848f1337c16564335dd8fb2b

0:e66bb061af06

		# Set XML namespace
		ns <- c(pubchem = "http://pubchem.ncbi.nlm.nih.gov/pug_view")

		# Define xpath expressions
		xpath.expr <- character()
		xpath.expr[[RBIODB.ACCESSION]] <- "//pubchem:RecordType[text()='CID']/../pubchem:RecordNumber"
		xpath.expr[[RBIODB.INCHI]] <- "//pubchem:Name[text()='InChI']/../pubchem:StringValue"
		xpath.expr[[RBIODB.INCHIKEY]] <- "//pubchem:Name[text()='InChI Key']/../pubchem:StringValue"

		for (content in contents) {

			# Create instance
			compound <- PubchemCompound$new()

				name <- NA_character_
				tryCatch( { name <- xpathSApply(xml, "//pubchem:Name[text()='IUPAC Name']/../pubchem:StringValue", xmlValue, namespaces = ns) }, warning = function(w) {})
				if (is.na(name))
					tryCatch( { name <- xpathSApply(xml, "//pubchem:Name[text()='Record Title']/../pubchem:StringValue", xmlValue, namespaces = ns) }, warning = function(w) {})
				if ( ! is.na(name))
					compound$setField(RBIODB.NAME, name)

			}

			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]]
	

1:253d531a0193

		# Set XML namespace
		ns <- c(pubchem = "http://pubchem.ncbi.nlm.nih.gov/pug_view")

		# Define xpath expressions
		xpath.expr <- character()
		xpath.expr[[BIODB.ACCESSION]] <- "//pubchem:RecordType[text()='CID']/../pubchem:RecordNumber"
		xpath.expr[[BIODB.INCHI]] <- "//pubchem:Name[text()='InChI']/../pubchem:StringValue"
		xpath.expr[[BIODB.INCHIKEY]] <- "//pubchem:Name[text()='InChI Key']/../pubchem:StringValue"

		for (content in contents) {

			# Create instance
			compound <- PubchemCompound$new()

				name <- NA_character_
				tryCatch( { name <- xpathSApply(xml, "//pubchem:Name[text()='IUPAC Name']/../pubchem:StringValue", xmlValue, namespaces = ns) }, warning = function(w) {})
				if (is.na(name))
					tryCatch( { name <- xpathSApply(xml, "//pubchem:Name[text()='Record Title']/../pubchem:StringValue", xmlValue, namespaces = ns) }, warning = function(w) {})
				if ( ! is.na(name))
					compound$setField(BIODB.NAME, name)

			}

			compounds <- c(compounds, compound)
		}

		# Replace elements with no accession id by NULL
		compounds <- lapply(compounds, function(x) if (is.na(x$getField(BIODB.ACCESSION))) NULL else x)

		# If the input was a single element, then output a single object
		if (drop && length(contents) == 1)
			compounds <- compounds[[1]]