1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
|
# Adapt this to be able to specify an XPath expression to identify a list of nodes.
setGeneric("xmlToDataFrame",
#
# Read a relatively flat, 2-level deep XML document into a data frame.
# The document is assumed to be something of the form
# <top>
# <obs>
# <var1>value</var1>
# <var2>value</var2>
# <var3>value</var3>
# </obs>
# <obs>
# <var1>value</var1>
# <var2>value</var2>
# <var3>value</var3>
# </obs>
# </top>
#
# This can handle cases where not all observations have the same
# fields.
#
# z = xmlToDataFrame("~/size.xml")
# z = xmlToDataFrame("~/size.xml", c("integer", "integer", "numeric"))
#
function(doc, colClasses = NULL, homogeneous = NA, collectNames = TRUE, nodes = list(), stringsAsFactors = FALSE)
standardGeneric("xmlToDataFrame"))
setMethod("xmlToDataFrame", "character",
# parse the XML document if it is a file name and
# not a regular XML document already.
function(doc, colClasses = NULL, homogeneous = NA, collectNames = TRUE, nodes = list(), stringsAsFactors = FALSE)
xmlToDataFrame(xmlParse(doc), colClasses, homogeneous, collectNames, stringsAsFactors = stringsAsFactors))
setMethod("xmlToDataFrame", c("XMLInternalDocument", nodes = "missing"),
function(doc, colClasses = NULL, homogeneous = NA, collectNames = TRUE, nodes = list(), stringsAsFactors = FALSE)
xmlToDataFrame(doc, colClasses, homogeneous, collectNames, nodes = xmlChildren(xmlRoot(doc)), stringsAsFactors))
tmp =
function(doc, colClasses = NULL, homogeneous = NA, collectNames = TRUE, nodes = list(), stringsAsFactors = FALSE)
{
if(length(nodes) == 0)
return(data.frame())
# Find out how many fields there.
nfields = sapply(nodes, xmlSize)
nvar = max(nfields)
if(collectNames)
varNames = unique(unlist( lapply(nodes, names) ))
else
varNames = names(nodes[[which.max(nfields)]])
if(is.na(homogeneous))
homogeneous = all(nfields == nvar) && all(sapply(nodes[-1], function(x) all(names(x) == varNames)))
if(!homogeneous)
return(fromRaggedXML2DataFrame(nodes, varNames, c(length(nfields), length(varNames)), colClasses, stringsAsFactors))
# Function to operate on each
fun = function(x) {
tmp = xmlSApply(x, xmlValue)
length(tmp) = nvar
tmp
}
# Get the individual values
vals = unlist(lapply(nodes, fun))
ans = matrix(vals, length(nfields), byrow = TRUE)
ans =
if(length(colClasses)) {
as.data.frame(lapply(seq(along = colClasses),
function(i) {
as(ans[, i], colClasses[i])
}), stringsAsFactors = stringsAsFactors)
} else
as.data.frame(ans, stringsAsFactors = stringsAsFactors)
names(ans) = varNames
ans
}
bob =
function(doc, colClasses = NULL, homogeneous = NA, collectNames = TRUE, nodes = list(), stringsAsFactors = FALSE)
xmlToDataFrame(nodes = doc, colClasses = colClasses, homogeneous = homogeneous, collectNames = collectNames, stringsAsFactors = stringsAsFactors)
setMethod("xmlToDataFrame", c(nodes = "XMLNodeSet"), tmp)
setMethod("xmlToDataFrame", c(nodes = "list"), tmp)
setOldClass("XMLInternalNodeList")
setMethod("xmlToDataFrame", c(nodes = "XMLInternalNodeList"), tmp)
setMethod("xmlToDataFrame", "XMLNodeSet", bob)
setMethod("xmlToDataFrame", "XMLInternalNodeList", bob)
setMethod("xmlToDataFrame", "list", bob)
setMethod("xmlToDataFrame", "XMLInternalElementNode",
function(doc, colClasses = NULL, homogeneous = NA, collectNames = TRUE, nodes = list(), stringsAsFactors = FALSE)
xmlToDataFrame(nodes = xmlChildren(doc), colClasses = colClasses, homogeneous = homogeneous, collectNames = collectNames, stringsAsFactors = stringsAsFactors))
fromRaggedXML2DataFrame =
#
# This reads data from the nodes of an XML document and assumes
# that they do not all have the same number or even names of fields.
# So this does extra work to match each observation to the union of
# the field names across all nodes.
#
# o = fromRaggedXML2DataFrame("size2.xml")
# o = fromRaggedXML2DataFrame("size1.xml")
#
function(nodes, varNames = unique(unlist( lapply(nodes, names) )),
dims = c(length(nodes), length(varNames)), colClasses = NULL,
stringsAsFactors = FALSE)
{
#XXX
if(is.character(nodes))
nodes = xmlChildren(xmlRoot(xmlParse(nodes)))
# create an empty data frame with as many rows and columns as needed.
ans = as.data.frame(replicate(dims[2], rep(as.character(NA), dims[1]), simplify = FALSE), stringsAsFactors = FALSE)
names(ans) = varNames
# Fill in the rows based on the names.
for(i in seq(length = dims[1]))
ans[i, names(nodes[[i]])] = xmlSApply(nodes[[i]], xmlValue)
# Convert the columns to the specified classes if specified.
# Should drop cols with NULL. Also guess those with NA.
if(length(colClasses)) {
i = ! sapply(colClasses, is.null)
ans = ans[ i ]
varNames = varNames[i]
colClasses = colClasses[ i ]
ans = as.data.frame(lapply(seq(length = ncol(ans)),
function(i) {
as(ans[, i], colClasses[[i]])
}), stringsAsFactors = stringsAsFactors)
}
names(ans) = varNames
ans
}
setGeneric("xmlAttrsToDataFrame",
function(doc, attrs = character(), omit = character(), ...)
standardGeneric("xmlAttrsToDataFrame"))
setMethod("xmlAttrsToDataFrame", "character",
function(doc, attrs = character(), omit = character(), ...)
xmlAttrsToDataFrame(xmlParse(doc), attrs, omit, ...))
setMethod("xmlAttrsToDataFrame", "AsIs",
function(doc, attrs = character(), omit = character(), ...)
xmlAttrsToDataFrame(xmlParse(doc), attrs, omit, ...))
setMethod("xmlAttrsToDataFrame", "XMLInternalElementNode",
function(doc, attrs = character(), omit = character(), ...)
xmlAttrsToDataFrame(xmlChildren(doc), attrs, omit, ...))
setMethod("xmlAttrsToDataFrame", "XMLNodeSet",
function(doc, attrs = character(), omit = character(), ...) {
xmlAttrsToDataFrame(as(doc, 'list'), attrs, omit, ...)
})
setMethod("xmlAttrsToDataFrame", "list",
function(doc, attrs = character(), omit = character(), ...) {
# assuming these are all nodes.
combineNamedVectors(lapply(doc, xmlAttrs), attrs, omit, ...)
})
setMethod("xmlAttrsToDataFrame", "XMLInternalNodeList",
function(doc, attrs = character(), omit = character(), ...) {
# assuming these are all nodes.
combineNamedVectors(lapply(doc, xmlAttrs), attrs, omit, ...)
})
inAllRecords =
function(x)
{
tt = table(unlist(lapply(x, names)))
names(tt)[ tt == length(x)]
}
allNames =
function(x)
unique( unlist(lapply(x, names)) )
combineNamedVectors =
function(els, attrs = character(), omit = character(), ...)
{
if(is.function(attrs))
attrs = attrs(els)
if(!length(attrs)) {
attrs = allNames(els)
if(length(omit))
attrs = setdiff(attrs, omit)
}
if(length(attrs) == 0) {
warning("no elements to combine across records")
return(data.frame())
}
values = lapply(els, function(x) {
structure(x[attrs], names = attrs)
})
ans = as.data.frame(do.call(rbind, values), row.names = NULL, ...)
rownames(ans) = NULL
ans
}
|
