diff PubchemEntry.R @ 2:20d69a062da3 draft

planemo upload for repository https://github.com/workflow4metabolomics/lcmsmatching.git commit d4048accde6bdfd5b3e14f5394902d38991854f8

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/PubchemEntry.R	Thu Mar 02 08:55:00 2017 -0500
@@ -0,0 +1,108 @@
+#####################
+# CLASS DECLARATION #
+#####################
+
+PubchemEntry <- methods::setRefClass("PubchemEntry", contains = "BiodbEntry")
+PubchemSubstance <- methods::setRefClass("PubchemSubstance", contains = "BiodbEntry")
+
+#####################
+# SUBSTANCE FACTORY #
+#####################
+
+createPubchemSubstanceFromXml <- function(contents, drop = TRUE) {
+
+	entries <- list()
+
+	# Define xpath expressions
+	xpath.expr <- character()
+	xpath.expr[[BIODB.ACCESSION]] <- "//PC-ID_id"
+	#xpath.expr[[BIODB.PUBCHEMCOMP.ID]] <- "//PC-CompoundType_id_cid" --> Apparently that can be more than one CID for a substance.
+
+	for (content in contents) {
+
+		# Create instance
+		entry <- PubchemEntry$new()
+
+		if ( ! is.null(content) && ! is.na(content)) {
+
+			# Parse XML
+			xml <-  XML::xmlInternalTreeParse(content, asText = TRUE)
+
+			# Unknown entry
+			fault <- XML::xpathSApply(xml, "/Fault", XML::xmlValue)
+			if (length(fault) == 0) {
+
+				# Test generic xpath expressions
+				for (field in names(xpath.expr)) {
+					v <- XML::xpathSApply(xml, xpath.expr[[field]], XML::xmlValue)
+					if (length(v) > 0)
+						entry$setField(field, v)
+				}
+			}
+		}
+
+		entries <- c(entries, entry)
+	}
+
+	# Replace elements with no accession id by NULL
+	entries <- lapply(entries, function(x) if (is.na(x$getField(BIODB.ACCESSION))) NULL else x)
+
+	# If the input was a single element, then output a single object
+	if (drop && length(contents) == 1)
+		entries <- entries[[1]]
+
+	return(entries)
+}
+
+####################
+# COMPOUND FACTORY #
+####################
+
+createPubchemEntryFromXml <- function(contents, drop = TRUE) {
+
+	entries <- list()
+
+	# Define xpath expressions
+	xpath.expr <- character()
+	xpath.expr[[BIODB.ACCESSION]] <- "//PC-CompoundType_id_cid"
+	xpath.expr[[BIODB.INCHI]] <- "//PC-Urn_label[text()='InChI']/../../..//PC-InfoData_value_sval"
+	xpath.expr[[BIODB.INCHIKEY]] <- "//PC-Urn_label[text()='InChIKey']/../../..//PC-InfoData_value_sval"
+	xpath.expr[[BIODB.FORMULA]] <- "//PC-Urn_label[text()='Molecular Formula']/../../..//PC-InfoData_value_sval"
+	xpath.expr[[BIODB.MASS]] <- "//PC-Urn_label[text()='Mass']/../../..//PC-InfoData_value_fval"
+	xpath.expr[[BIODB.COMP.IUPAC.NAME.SYST]] <- "//PC-Urn_label[text()='IUPAC Name']/../PC-Urn_name[text()='Systematic']/../../..//PC-InfoData_value_sval"
+
+	for (content in contents) {
+
+		# Create instance
+		entry <- PubchemEntry$new()
+
+		if ( ! is.null(content) && ! is.na(content)) {
+
+			# Parse XML
+			xml <-  XML::xmlInternalTreeParse(content, asText = TRUE)
+
+			# Unknown entry
+			fault <- XML::xpathSApply(xml, "/Fault", XML::xmlValue)
+			if (length(fault) == 0) {
+
+				# Test generic xpath expressions
+				for (field in names(xpath.expr)) {
+					v <- XML::xpathSApply(xml, xpath.expr[[field]], XML::xmlValue)
+					if (length(v) > 0)
+						entry$setField(field, v)
+				}
+			}
+		}
+
+		entries <- c(entries, entry)
+	}
+
+	# Replace elements with no accession id by NULL
+	entries <- lapply(entries, function(x) if (is.na(x$getField(BIODB.ACCESSION))) NULL else x)
+
+	# If the input was a single element, then output a single object
+	if (drop && length(contents) == 1)
+		entries <- entries[[1]]
+
+	return(entries)
+}