使用 R 从 csv 文件创建 xml 文件

Question

我正在尝试使用CSV文件中的XML包创建一个xml文件。我的 CSV 文件如下所示：

>head(patient)
Source         Target             weight
1 Bacteroides   Lachnospiraceae  3.80735493
2 Bacteroides        Klebsiella -1.61890983
3 Bacteroides Lachnoclostridium  3.80735493
4 Bacteroides     Streptococcus -1.77760758
5 Streptococcus   Clostridium    1.19264508
6 Streptococcus [Eubacterium]    5.58496251

我的愿望输出 xml 应如下所示：

<?xml version="1.0" encoding="iso-8859-1"?>
<gxl>
<graph id="graph id= ExtendedCallGraph edgeids=true edgemode=undirected">
<node id="1">
<attr name="Bacteroides">
</attr>
</node>
<edge from="Bacteroides" to="Lachnospiraceae" isdirected="False" id="1--2">
</edge>
<edge from="Bacteroides" to=" Klebsiella" isdirected="False" id="1--2">
</edge>
<edge from="Bacteroides" to="Lachnoclostridium" isdirected="False" id="1--3">
</edge>
<edge from="Bacteroides" to=" Streptococcus" isdirected="False" id="1--4">
</edge>
<node id="2">
<attr name="Streptococcus">
</attr>
</node>
<edge from="Streptococcus" to="Clostridium" isdirected="False" id="2--3">
</edge>
<edge from="Streptococcus" to="Eubacterium" isdirected="False" id="2--4">
</edge>
:
:
:
:
</graph>
</gxl>

我尝试了以下代码：

DD = xmlHashTree()
top1<-addNode(xmlNode("gxl"), character(), DD)
addNode(xmlNode("graph id= ExtendedCallGraph edgeids=true edgemode=directed"),top1,DD,close=FALSE)
lapply(unique(patient_1$Source),function(x){
b=addNode(xmlNode("node",attrs = c('id' = as.integer(x))),top1,DD)
c=addNode(xmlNode("attr",attrs = c('name' = as.character(x))),b,DD)})
#####I am trying to add edge node from source to Target########
apply(unique(patient_1[,1:2]),1,function(x){
e=addNode(xmlNode("edge",attrs = c("from"= as.character(patient_1$Source[1]), 
"to"=as.character(patient_1$target[1]), isdirected="false")),top1,DD)})

但我在这一点上陷入了困境。如何放置"from"属性的源列和"to"属性的目标。我如何实现这一点？ 提前致谢

Answer 1

考虑一个更简单的嵌套for循环，它沿着每个唯一的源及其子集观测值向下走动。与应用族解决方案相反，您可以保留所需@id属性的小版本号并增加 XML 树。此外，请考虑使用newXMLNode方法构建元素及其attrs参数，用于传入带有c()的命名向量的属性

# CREATE XML FILE
doc = newXMLDoc()
root = newXMLNode("gxl", doc = doc)
graph = newXMLNode("graph", parent = root,
attrs = c(id="ExtendedCallGraph", edgeids="true", edgemode="directed"))
# WRITE XML NODES AND DATA
sources <- unique(patient$Source)
for(i in seq_along(sources)){
# NODE nodes
grp_node = newXMLNode("node", parent = graph, attrs=c(name=i))
attr_node = newXMLNode("attr", sources[i], parent = grp_node)
sub_df <- subset(patient, Source == sources[i])
# EDGE nodes
for(j in 1:nrow(sub_df)){
edge_node = newXMLNode("edge", parent=graph,
attrs=c(from=sub_df$Source[j], to=sub_df$Target[j],
isdirected="False", id=paste0(i, '--', j)))
}
}    
# OUTPUT XML CONTENT TO SCREEN
print(doc)
# OUTPUT XML CONTENT TO FILE
saveXML(doc, file="Output.xml")

输出

print(doc)    
# <?xml version="1.0"?>
# <gxl>
#   <graph id="ExtendedCallGraph" edgeids="true" edgemode="directed">
#     <node name="1">
#       <attr>Bacteroides</attr>
#     </node>
#     <edge from="Bacteroides" to="Lachnospiraceae" isdirected="False" id="1--1"/>
#     <edge from="Bacteroides" to="Klebsiella" isdirected="False" id="1--2"/>
#     <edge from="Bacteroides" to="Lachnoclostridium" isdirected="False" id="1--3"/>
#     <edge from="Bacteroides" to="Streptococcus" isdirected="False" id="1--4"/>
#     <node name="2">
#       <attr>Streptococcus</attr>
#     </node>
#     <edge from="Streptococcus" to="Clostridium" isdirected="False" id="2--1"/>
#     <edge from="Streptococcus" to="[Eubacterium]" isdirected="False" id="2--2"/>
#   </graph>
# </gxl>

Answer 2

我也使用 foreach 函数实现了。但这需要相当长的时间。

Fin_Doc = newXMLDoc()
root = newXMLNode("gxl", doc = Fin_Doc)
graph = newXMLNode("graph", parent = root,
attrs = c(id="Co-occurance Network", edgeids="true", edgemode="undirected"))
##########adding the node id and attribute_name##########
foreach(w=as.vector(unique(patient_1$Otu_1)),y = as.vector(unique(patient_1$taxonomy.y)), x=as.vector(patient_1$taxonomy.x)) %do%{
(grp_node = newXMLNode("node", parent = graph, attrs=c(id= "_")))
(attr_name = newXMLNode("attr",parent = grp_node, text="", attrs=c(name="OTU")))
(otu_id=newXMLNode("int",parent = attr_name, text="",w ))
(bacteria=newXMLNode("attr",parent = grp_node, text="", attrs=c(name="Bacteria")))
(string_name=newXMLNode("string",parent = bacteria, text="",y))
}
####################edge from otuids########################## 
foreach(w=as.vector(patient_1$Otu_1),q=as.vector(patient_1$Otu_2), z=as.vector(patient_1$patient1))%do% {   (edge_node1 = newXMLNode("edge", parent=graph,text="n", attrs=c(from= w, to=q)))
(attrs_node1=newXMLNode("attr", parent=edge_node1, text=" ", attrs=c("logratio")))   
(weight_node1= newXMLNode("float", as.character(z), parent=attrs_node1, text=" ")) }