File: spss.R

package info (click to toggle)
foreign 0.8.40-1
  • links: PTS
  • area: main
  • in suites: squeeze
  • size: 1,576 kB
  • ctags: 657
  • sloc: ansic: 6,997; asm: 4; sh: 2; makefile: 1
file content (179 lines) | stat: -rw-r--r-- 7,046 bytes parent folder | download
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
### This file is part of the 'foreign' package for R.

###
###		Read SPSS system data files
###
### Copyright 2000-2002 Saikat DebRoy <saikat$stat.wisc.edu>
###			Douglas M. Bates <bates$stat.wisc.edu>,
###			Thomas Lumley
### Copyright 2007-9 R Core Development Team
### This file is part of the `foreign' package for R and related languages.
### It is made available under the terms of the GNU General Public
### License, version 2, or at your option, any later version,
### incorporated herein by reference.
###
### This program is distributed in the hope that it will be
### useful, but WITHOUT ANY WARRANTY; without even the implied
### warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR
### PURPOSE.  See the GNU General Public License for more
### details.
###
### You should have received a copy of the GNU General Public
### License along with this program; if not, a copy is available at
### http://www.r-project.org/Licenses/

read.spss <- function(file, use.value.labels = TRUE, to.data.frame = FALSE,
		      max.value.labels = Inf, trim.factor.names = FALSE,
                      trim_values = TRUE, reencode = NA,
                      use.missings = to.data.frame)
{

    trim <- function(strings, trim=TRUE)
	if (trim) sub(" +$","",strings) else strings

    ## mappings taken from win-iconv
    knownCP <- c("UCS-2LE" = 1200, "UCS-2BE" = 1201,
                 "macroman" = 10000, " UCS-4LE" = 12000, "UCS-4BE" = 12001,
                 "koi8-r" = 20866, "koi8-u" = 21866,
                 "latin1" = 28591, "latin2" = 28592, "latin3" = 28593,
                 ## latin-9 seems most portable, but only on Windows
                 ## R >= 2.10.0.  libiconv doesn't know latin9.
                 "latin4" = 28594, "latin-9" = 28605,
                 "ISO-2022-JP" = 50221, "euc-jp" = 51932,
                 "UTF-8" = 65001,
                 "ASCII" = 20127,
                 ## pages known to glibc and libiconv
                 "CP1250" = 1250,
                 "CP1251" = 1251,
                 "CP1252" = 1252,
                 "CP1253" = 1253,
                 "CP1254" = 1254,
                 "CP1255" = 1255,
                 "CP1256" = 1256,
                 "CP1257" = 1257,
                 "CP1258" = 1258,
                 "CP874" = 874,
                 "CP936" = 936)

    if(length(grep("^(http|ftp|https)://", file))) {
        tmp <- tempfile()
        download.file(file, tmp, quiet = TRUE, mode = "wb")
        file <- tmp
        on.exit(unlink(file))
    }
    rval <- .Call(do_read_SPSS, file)
    codepage <- attr(rval, "codepage")
    if(is.null(codepage)) codepage <- 2 # .por files
    if(!capabilities("iconv")) reencode <- FALSE
    if(!identical(reencode, FALSE)) {
        cp <- "unknown"
        if(is.character(reencode)) {
            cp <- reencode
            reencode <- TRUE
        } else if(codepage == 20127) {
            reencode <- FALSE # ASCII
        } else if(m <- match(codepage, knownCP, 0L)) {
            cp <-names(knownCP)[m]
        } else if (codepage < 200) {
            ## small numbers are not codepages, and real codepages are large
            attr(rval, "codepage") <- NULL
            reencode <- FALSE
        } else cp <- paste("CP", codepage, sep="")
        if(is.na(reencode)) reencode <- l10n_info()[["UTF-8"]]

        if(reencode) {
            message("re-encoding from ", cp)
            names(rval) <- iconv(names(rval), cp, "")
            vl <- attr(rval, "variable.labels")
            nm <- names(vl)
            vl <- iconv(vl, cp, "")
            names(vl) <- iconv(nm, cp, "")
            attr(rval, "variable.labels") <- vl
            for(i in seq_along(rval)) {
                xi <- rval[[i]]
                if(is.character(xi)) rval[[i]] <- iconv(xi, cp, "")
            }
        }
    }

    miss <- attr(rval, "missings")
    if(!is.null(miss)) {
        if(reencode) {
            nm <- names(miss)
            nm <- iconv(nm, cp, "")
            names(miss) <- iconv(nm, cp, "")
            for(i in seq_along(miss))
                if(is.character(miss[[i]]$value))
                   miss[[i]]$value <- iconv(miss[[i]]$value, cp, "")
            attr(rval, "missings") <- miss
        }
        if(use.missings)
            for(v in names(rval)) {
                tp <- miss[[v]]$type
                if(tp %in% "none") next
                if(tp %in% c("one", "two", "three")) {
                    xi <- rval[[v]]
                    other <- miss[[v]]$value
                    ## FIXME: do we need to worry about padding for string vals?
                    xi[xi %in% other] <- NA
                    rval[[v]] <- xi
                    ## NB: not much tested from here down
                } else if(tp == "low" || tp == "low+1") {
                    xi <- rval[[v]]
                    z <- miss[[v]]$value
                    if(tp == "low+1") xi[ xi <= z[1L] | xi == z[2L] ] <- NA
                    else  xi[xi <= z[1L]] <- NA
                    rval[[v]] <- xi
                } else if(tp == "high" || tp == "high+1") {
                    xi <- rval[[v]]
                    z <- miss[[v]]$value
                    if(tp == "high+1") xi[ xi >= z[1L] | xi == z[2L] ] <- NA
                    else xi[ xi >= z[1L] ] <- NA
                    rval[[v]] <- xi
                } else if(tp == "range" || tp == "range+1") {
                    xi <- rval[[v]]
                    z <- miss[[v]]$value
                    if(tp == "range+1")
                        xi[ xi >= z[1L] | xi <= z[2L] | xi[xi == z[3L]] ] <- NA
                    else
                        xi[ xi >= z[1L] | xi <= z[2L] ] <- NA
                    rval[[v]] <- xi
                } else
                    warning(gettextf("missingness type %s is not handled", tp),
                            domain = NA)
        }
    } else use.missings <- FALSE

    vl <- attr(rval,"label.table")
    if(reencode) names(vl) <- iconv(names(vl), cp, "")
    has.vl <- which(!sapply(vl, is.null))
    for(v in has.vl) {
	nm <- names(vl)[[v]]
	nvalues <- length(na.omit(unique(rval[[nm]])))
	nlabels <- length(vl[[v]])
        if(reencode && nlabels) {
            nm2 <- names(vl[[v]])
            vl[[v]] <- iconv(vl[[v]], cp, "")
            names(vl[[v]]) <- iconv(nm2, cp, "")
        }
        if(use.missings && !is.null(mv <- miss[[v]]$value))
            vl[[v]] <- vl[[v]][! vl[[v]] %in% mv]
	if (use.value.labels &&
	    (!is.finite(max.value.labels) || nvalues <= max.value.labels) &&
	    nlabels >= nvalues) {
	    rval[[nm]] <- factor(trim(rval[[nm]], trim_values),
                                 levels = rev(trim(vl[[v]], trim_values)),
				 labels = rev(trim(names(vl[[v]]), trim.factor.names)))
        } else
	    attr(rval[[nm]], "value.labels") <- vl[[v]]
    }
    if(reencode) attr(rval, "label.table") <- vl

    if (to.data.frame) {
	varlab <- attr(rval, "variable.labels")
	rval <- as.data.frame(rval)
	attr(rval, "variable.labels") <- varlab
        if(codepage > 500) attr(rval, "codepage") <- codepage
    }
    rval
}