R/pnad.R
In ajdamico/lodown: locally download and prepare publicly-available microdata
Defines functions
pnad_remove_uf
lodown_pnad
get_catalog_pnad
get_catalog_pnad <-
function( data_name = "pnad" , output_dir , ... ){
# initiate the full ftp path
year.ftp <- "ftp://ftp.ibge.gov.br/Trabalho_e_Rendimento/Pesquisa_Nacional_por_Amostra_de_Domicilios_anual/microdados/"
# read the text of the microdata ftp into working memory
# download the contents of the ftp directory for all microdata
year.listing <- readLines( textConnection( RCurl::getURL( year.ftp ) ) )
# extract all years
year.lines <- gsub( "(.*)([0-9][0-9][0-9][0-9])" , "\\2" , year.listing )
# available years 2013+
suppressWarnings( year.lines <- year.lines[ !is.na( as.numeric( year.lines ) ) & year.lines >= 2013 ] )
catalog <-
data.frame(
year = c( 1992:1993 , 2001:2009 , 2011:2012 , year.lines ) ,
ftp_folder = paste0( year.ftp , c( 1992:1993 , rep( 'reponderacao_2001_2012' , 11 ) , year.lines ) , '/' ) ,
stringsAsFactors = FALSE
)
# loop through every year
for ( this_entry in seq( nrow( catalog ) ) ){
# find the zipped files in the year-specific folder
ftp.listing <- readLines( textConnection( RCurl::getURL( catalog[ this_entry , 'ftp_folder' ] ) ) )
filenames <- gsub( "(.*) (.*)" , "\\2" , ftp.listing ) ; filenames <- filenames[ filenames != "" ]
if( catalog[ this_entry , 'year' ] >= 2013 ){
catalog[ this_entry , 'full_url' ] <- paste0( catalog[ this_entry , 'ftp_folder' ] , grep( "^dados" , filenames , ignore.case = TRUE , value = TRUE ) )
catalog[ this_entry , 'sas_ri' ] <- paste0( catalog[ this_entry , 'ftp_folder' ] , grep( "^dicionarios" , filenames , ignore.case = TRUE , value = TRUE ) )
} else if( catalog[ this_entry , 'year' ] >= 2001 ) {
catalog[ this_entry , 'full_url' ] <- paste0( catalog[ this_entry , 'ftp_folder' ] , grep( paste0( "pnad_reponderado_" , catalog[ this_entry , 'year' ] ) , filenames , ignore.case = TRUE , value = TRUE ) )
} else {
catalog[ this_entry , 'sas_ri' ] <- paste0( catalog[ this_entry , 'ftp_folder' ] , grep( "^layout\\.zip" , filenames , ignore.case = TRUE , value = TRUE ) )
catalog[ this_entry , 'full_url' ] <- paste0( catalog[ this_entry , 'ftp_folder' ] , "Dados.zip" )
}
}
catalog$output_filename <- paste0( output_dir , "/" , catalog$year , " main.rds" )
catalog
}
lodown_pnad <-
function( data_name = "pnad" , catalog , ... ){
on.exit( print( catalog ) )
tf <- tempfile()
# download and import the tables containing missing codes
household.nr <- read.csv( system.file("extdata", "pnad/household_nr.csv", package = "lodown") , colClasses = 'character' )
person.nr <- read.csv( system.file("extdata", "pnad/person_nr.csv", package = "lodown") , colClasses = 'character' )
# convert these tables to lowercase
names( household.nr ) <- tolower( names( household.nr ) )
names( person.nr ) <- tolower( names( person.nr ) )
# remove all spaces between missing codes
household.nr$code <- gsub( " " , "" , household.nr$code )
person.nr$code <- gsub( " " , "" , person.nr$code )
# convert all code column names to lowercase
household.nr$variable <- tolower( household.nr$variable )
person.nr$variable <- tolower( person.nr$variable )
for ( i in seq_len( nrow( catalog ) ) ){
# download the file
cachaca( catalog[ i , "full_url" ] , tf , mode = 'wb' , filesize_fun = 'unzip_verify' )
unzipped_files <- unzip_warn_fail( tf , exdir = paste0( tempdir() , "/unzips" ) )
# download the sas file
if( !is.na( catalog[ i , 'sas_ri' ] ) ){
cachaca( catalog[ i , "sas_ri" ] , tf , mode = 'wb' , filesize_fun = 'unzip_verify' )
unzipped_files <- c( unzipped_files , unzip_warn_fail( tf , exdir = paste0( tempdir() , "/unzips" ) ) )
}
# manually set the encoding of the unziped files so they don't break things.
if( catalog[ i , 'year' ] > 2012 & .Platform$OS.type != 'windows' ) Encoding( unzipped_files ) <- 'UTF-8' else Encoding( unzipped_files ) <- 'latin1'
dom_uf <- unzipped_files[ grepl( paste0( 'input[^?]dom' , catalog[ i , 'year' ] , '.txt' ) , tolower( unzipped_files ) ) ]
pes_uf <- unzipped_files[ grepl( paste0( 'input[^?]pes' , catalog[ i , 'year' ] , '.txt' ) , tolower( unzipped_files ) ) ]
if( length( dom_uf ) == 0 ) dom_uf <- unzipped_files[ grepl( paste0( 'sas_do(.*)\\.txt' ) , tolower( unzipped_files ) ) ]
if( length( pes_uf ) == 0 ) pes_uf <- unzipped_files[ grepl( paste0( 'sas_pe(.*)\\.txt' ) , tolower( unzipped_files ) ) ]
# remove the UF column and the mistake with "LOCAL ULTIMO FURTO"
# described in the pnad_remove_uf() function that was loaded with source_url as pnad.survey.R
dom.sas <- pnad_remove_uf( dom_uf )
pes.sas <- pnad_remove_uf( pes_uf )
# in 2003 and 2007, the age variable had been read in as a factor variable
# which breaks certain commands by treating the variable incorrectly as a factor
if( catalog[ i , 'year' ] %in% c( 2003 , 2007 ) ){
pes_con <- file( pes.sas , "rb" , encoding = "windows-1252" )
pes_lines <- readLines( pes_con )
close( pes_con )
pes_lines <- iconv( pes_lines , "" , "ASCII//TRANSLIT" )
pes_lines <- gsub( "@00027( *)V8005( *)\\$3\\." , "@00027 V8005 3\\." , pes_lines )
writeLines( pes_lines , pes.sas )
}
# since `files` contains multiple file paths,
# determine the filepath on the local disk to the household (dom) and person (pes) files
dom.fn <- unzipped_files[ grepl( paste0( '/dom' , catalog[ i , 'year' ] ) , tolower( unzipped_files ) ) ]
pes.fn <- unzipped_files[ grepl( paste0( '/pes' , catalog[ i , 'year' ] ) , tolower( unzipped_files ) ) ]
if( length( dom.fn ) == 0 ) dom.fn <- unzipped_files[ grepl( 'dados/d' , tolower( unzipped_files ) ) ]
if( length( pes.fn ) == 0 ) pes.fn <- unzipped_files[ grepl( 'dados/p' , tolower( unzipped_files ) ) ]
pes_df <-
read_SAScii(
pes.fn ,
pes.sas
)
# dump 100% missing columns
pes_df <- pes_df[ !sapply( pes_df , function( w ) all( is.na( w ) ) ) ]
gc()
dom_df <-
read_SAScii(
dom.fn ,
dom.sas
)
# dump 100% missing columns
dom_df <- dom_df[ !sapply( dom_df , function( w ) all( is.na( w ) ) ) ]
gc()
names( dom_df ) <- tolower( names( dom_df ) )
names( pes_df ) <- tolower( names( pes_df ) )
# confirm no fields are in `dom` unless they are in `pes`
b_fields <- c( 'v0101' , if( 'v0102' %in% names( pes_df ) ) 'v0102' else c( 'control' , 'uf' ) , 'v0103' , setdiff( names( pes_df ) , names( dom_df ) ) )
pes_df <- pes_df[ b_fields ] ; gc()
# the ASCII and SAS importation instructions stored in temporary files
# on the local disk are no longer necessary, so delete them.
attempt.one <- try( file.remove( unzipped_files ) , silent = TRUE )
# weird brazilian file encoding operates differently on mac+*nix versus windows, so try both ways.
if( class( attempt.one ) == 'try-error' ) { Encoding( unzipped_files ) <- '' ; file.remove( unzipped_files ) }
# add 4617 and 4618 to 2001 file
if( catalog[ i , 'year' ] == 2001 ){
dom_df$v4617 <- dom_df$strat
dom_df$v4618 <- dom_df$psu
}
# missing level blank-outs #
# this section loops through the non-response values & variables for all years
# and sets those variables to NULL.
# cat( 'non-response variable blanking-out only occurs on numeric variables\n' )
# cat( 'categorical variable blanks are usually 9 in the pnad\n' )
# cat( 'thanks for listening\n' )
# loop through each row in the missing household-level codes table
for ( curRow in seq( nrow( household.nr ) ) ){
# if the variable is in the current table..
if( household.nr[ curRow , 'variable' ] %in% names( dom_df ) ){
# ..and the variable should be recoded for that year
if( catalog[ i , 'year' ] %in% eval( parse( text = household.nr[ curRow , 'year' ] ) ) ){
dom_df[ dom_df[ , household.nr[ curRow , 'variable' ] ] %in% household.nr[ curRow , 'code' ] , household.nr[ curRow , 'variable' ] ] <- NA
gc()
}
}
}
# loop through each row in the missing person-level codes table
for ( curRow in seq( nrow( person.nr ) ) ){
# if the variable is in the current table..
if( person.nr[ curRow , 'variable' ] %in% names( pes_df ) ){
# ..and the variable should be recoded for that year
if( catalog[ i , 'year' ] %in% eval( parse( text = person.nr[ curRow , 'year' ] ) ) ){
pes_df[ pes_df[ , person.nr[ curRow , 'variable' ] ] %in% person.nr[ curRow , 'code' ] , person.nr[ curRow , 'variable' ] ] <- NA
gc()
}
}
}
if( catalog[ i , 'year' ] >= 2001 ){
pes_df$uf <- substr( pes_df$v0102 , 1 , 2 )
pes_df$region <- substr( pes_df$v0102 , 1 , 1 )
}
pes_df$one <- 1
x <- merge( dom_df , pes_df )
stopifnot( nrow( x ) == nrow( pes_df ) ) ; rm( pes_df , dom_df ) ; gc()
# determine if the table contains a `v4619` variable.
# v4619 is the factor of subsampling used to compensate the loss of units in some states
# for 2012, the variable v4619 is one and so it is not needed.
# if it does not, create it.
if( !( 'v4619' %in% names( x ) ) ) x$v4619 <- 1
x$v4609 <- as.numeric( x$v4609 )
if( catalog[ i , 'year' ] < 2004 ){
x$pre_wgt <- as.numeric( x$v4610 )
} else {
x$pre_wgt <- as.numeric( x$v4619 ) * as.numeric( x$v4610 )
}
catalog[ i , 'case_count' ] <- nrow( x )
saveRDS( x , file = catalog[ i , 'output_filename' ] ) ; rm( x ) ; gc()
# delete the temporary files
suppressWarnings( file.remove( tf , unzipped_files ) )
cat( paste0( data_name , " catalog entry " , i , " of " , nrow( catalog ) , " stored at '" , catalog[ i , 'output_filename' ] , "'\r\n\n" ) )
}
on.exit()
catalog
}
# initiate a function that removes the "UF" field from the PNAD SAS importation script,
# because the R `SAScii` package does not currently handle overlapping fields and this UF column always overlaps
pnad_remove_uf <-
function( sasfile ){
# read the SAS import file into R
##does not work on linux sascon <- file( sasfile , "rb" , blocking = FALSE , encoding = "windows-1252" )
sas_lines <- readLines( sasfile , encoding="windows-1252")
sas_lines <- iconv( sas_lines , "windows-1252" , "ASCII//TRANSLIT" )
if (is.na(sas_lines[1])) {
sas_lines <- readLines( sasfile , encoding="latin1")
sas_lines <- iconv( sas_lines , "latin1" , "ASCII//TRANSLIT" )
}
##close( sascon )
# remove any TAB characters, replace them with two spaces
sas_lines <- gsub( "\t" , " " , sas_lines )
# throw out any lines that contain the UF line
sas_lines <- sas_lines[ !grepl( "@00005[ ]+UF[ ]" , sas_lines ) ]
# fix the "$1 ." difference seen in a number of PNAD SAS importation scripts
# by simply switching the space and the period
sas_lines <-
gsub(
"@00840 V2913 $1 ." ,
"@00840 V2913 $1. " ,
sas_lines ,
fixed = TRUE
)
# fix the duplicate column names in 2007
sas_lines <-
gsub(
"00965 V9993" ,
"00965 V9993A" ,
sas_lines ,
)
# create a temporary file
tf <- tempfile()
# write the updated sas input file to the temporary file
writeLines( sas_lines , tf )
# return the filepath to the temporary file containing the updated sas input script
tf
}
ajdamico/lodown documentation built on Oct. 21, 2024, 4:53 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Defines functions pnad_remove_uf lodown_pnad get_catalog_pnad
get_catalog_pnad <-
function( data_name = "pnad" , output_dir , ... ){
# initiate the full ftp path
year.ftp <- "ftp://ftp.ibge.gov.br/Trabalho_e_Rendimento/Pesquisa_Nacional_por_Amostra_de_Domicilios_anual/microdados/"
# read the text of the microdata ftp into working memory
# download the contents of the ftp directory for all microdata
year.listing <- readLines( textConnection( RCurl::getURL( year.ftp ) ) )
# extract all years
year.lines <- gsub( "(.*)([0-9][0-9][0-9][0-9])" , "\\2" , year.listing )
# available years 2013+
suppressWarnings( year.lines <- year.lines[ !is.na( as.numeric( year.lines ) ) & year.lines >= 2013 ] )
catalog <-
data.frame(
year = c( 1992:1993 , 2001:2009 , 2011:2012 , year.lines ) ,
ftp_folder = paste0( year.ftp , c( 1992:1993 , rep( 'reponderacao_2001_2012' , 11 ) , year.lines ) , '/' ) ,
stringsAsFactors = FALSE
)
# loop through every year
for ( this_entry in seq( nrow( catalog ) ) ){
# find the zipped files in the year-specific folder
ftp.listing <- readLines( textConnection( RCurl::getURL( catalog[ this_entry , 'ftp_folder' ] ) ) )
filenames <- gsub( "(.*) (.*)" , "\\2" , ftp.listing ) ; filenames <- filenames[ filenames != "" ]
if( catalog[ this_entry , 'year' ] >= 2013 ){
catalog[ this_entry , 'full_url' ] <- paste0( catalog[ this_entry , 'ftp_folder' ] , grep( "^dados" , filenames , ignore.case = TRUE , value = TRUE ) )
catalog[ this_entry , 'sas_ri' ] <- paste0( catalog[ this_entry , 'ftp_folder' ] , grep( "^dicionarios" , filenames , ignore.case = TRUE , value = TRUE ) )
} else if( catalog[ this_entry , 'year' ] >= 2001 ) {
catalog[ this_entry , 'full_url' ] <- paste0( catalog[ this_entry , 'ftp_folder' ] , grep( paste0( "pnad_reponderado_" , catalog[ this_entry , 'year' ] ) , filenames , ignore.case = TRUE , value = TRUE ) )
} else {
catalog[ this_entry , 'sas_ri' ] <- paste0( catalog[ this_entry , 'ftp_folder' ] , grep( "^layout\\.zip" , filenames , ignore.case = TRUE , value = TRUE ) )
catalog[ this_entry , 'full_url' ] <- paste0( catalog[ this_entry , 'ftp_folder' ] , "Dados.zip" )
}
}
catalog$output_filename <- paste0( output_dir , "/" , catalog$year , " main.rds" )
catalog
}
lodown_pnad <-
function( data_name = "pnad" , catalog , ... ){
on.exit( print( catalog ) )
tf <- tempfile()
# download and import the tables containing missing codes
household.nr <- read.csv( system.file("extdata", "pnad/household_nr.csv", package = "lodown") , colClasses = 'character' )
person.nr <- read.csv( system.file("extdata", "pnad/person_nr.csv", package = "lodown") , colClasses = 'character' )
# convert these tables to lowercase
names( household.nr ) <- tolower( names( household.nr ) )
names( person.nr ) <- tolower( names( person.nr ) )
# remove all spaces between missing codes
household.nr$code <- gsub( " " , "" , household.nr$code )
person.nr$code <- gsub( " " , "" , person.nr$code )
# convert all code column names to lowercase
household.nr$variable <- tolower( household.nr$variable )
person.nr$variable <- tolower( person.nr$variable )
for ( i in seq_len( nrow( catalog ) ) ){
# download the file
cachaca( catalog[ i , "full_url" ] , tf , mode = 'wb' , filesize_fun = 'unzip_verify' )
unzipped_files <- unzip_warn_fail( tf , exdir = paste0( tempdir() , "/unzips" ) )
# download the sas file
if( !is.na( catalog[ i , 'sas_ri' ] ) ){
cachaca( catalog[ i , "sas_ri" ] , tf , mode = 'wb' , filesize_fun = 'unzip_verify' )
unzipped_files <- c( unzipped_files , unzip_warn_fail( tf , exdir = paste0( tempdir() , "/unzips" ) ) )
}
# manually set the encoding of the unziped files so they don't break things.
if( catalog[ i , 'year' ] > 2012 & .Platform$OS.type != 'windows' ) Encoding( unzipped_files ) <- 'UTF-8' else Encoding( unzipped_files ) <- 'latin1'
dom_uf <- unzipped_files[ grepl( paste0( 'input[^?]dom' , catalog[ i , 'year' ] , '.txt' ) , tolower( unzipped_files ) ) ]
pes_uf <- unzipped_files[ grepl( paste0( 'input[^?]pes' , catalog[ i , 'year' ] , '.txt' ) , tolower( unzipped_files ) ) ]
if( length( dom_uf ) == 0 ) dom_uf <- unzipped_files[ grepl( paste0( 'sas_do(.*)\\.txt' ) , tolower( unzipped_files ) ) ]
if( length( pes_uf ) == 0 ) pes_uf <- unzipped_files[ grepl( paste0( 'sas_pe(.*)\\.txt' ) , tolower( unzipped_files ) ) ]
# remove the UF column and the mistake with "LOCAL ULTIMO FURTO"
# described in the pnad_remove_uf() function that was loaded with source_url as pnad.survey.R
dom.sas <- pnad_remove_uf( dom_uf )
pes.sas <- pnad_remove_uf( pes_uf )
# in 2003 and 2007, the age variable had been read in as a factor variable
# which breaks certain commands by treating the variable incorrectly as a factor
if( catalog[ i , 'year' ] %in% c( 2003 , 2007 ) ){
pes_con <- file( pes.sas , "rb" , encoding = "windows-1252" )
pes_lines <- readLines( pes_con )
close( pes_con )
pes_lines <- iconv( pes_lines , "" , "ASCII//TRANSLIT" )
pes_lines <- gsub( "@00027( *)V8005( *)\\$3\\." , "@00027 V8005 3\\." , pes_lines )
writeLines( pes_lines , pes.sas )
}
# since `files` contains multiple file paths,
# determine the filepath on the local disk to the household (dom) and person (pes) files
dom.fn <- unzipped_files[ grepl( paste0( '/dom' , catalog[ i , 'year' ] ) , tolower( unzipped_files ) ) ]
pes.fn <- unzipped_files[ grepl( paste0( '/pes' , catalog[ i , 'year' ] ) , tolower( unzipped_files ) ) ]
if( length( dom.fn ) == 0 ) dom.fn <- unzipped_files[ grepl( 'dados/d' , tolower( unzipped_files ) ) ]
if( length( pes.fn ) == 0 ) pes.fn <- unzipped_files[ grepl( 'dados/p' , tolower( unzipped_files ) ) ]
pes_df <-
read_SAScii(
pes.fn ,
pes.sas
)
# dump 100% missing columns
pes_df <- pes_df[ !sapply( pes_df , function( w ) all( is.na( w ) ) ) ]
gc()
dom_df <-
read_SAScii(
dom.fn ,
dom.sas
)
# dump 100% missing columns
dom_df <- dom_df[ !sapply( dom_df , function( w ) all( is.na( w ) ) ) ]
gc()
names( dom_df ) <- tolower( names( dom_df ) )
names( pes_df ) <- tolower( names( pes_df ) )
# confirm no fields are in `dom` unless they are in `pes`
b_fields <- c( 'v0101' , if( 'v0102' %in% names( pes_df ) ) 'v0102' else c( 'control' , 'uf' ) , 'v0103' , setdiff( names( pes_df ) , names( dom_df ) ) )
pes_df <- pes_df[ b_fields ] ; gc()
# the ASCII and SAS importation instructions stored in temporary files
# on the local disk are no longer necessary, so delete them.
attempt.one <- try( file.remove( unzipped_files ) , silent = TRUE )
# weird brazilian file encoding operates differently on mac+*nix versus windows, so try both ways.
if( class( attempt.one ) == 'try-error' ) { Encoding( unzipped_files ) <- '' ; file.remove( unzipped_files ) }
# add 4617 and 4618 to 2001 file
if( catalog[ i , 'year' ] == 2001 ){
dom_df$v4617 <- dom_df$strat
dom_df$v4618 <- dom_df$psu
}
# missing level blank-outs #
# this section loops through the non-response values & variables for all years
# and sets those variables to NULL.
# cat( 'non-response variable blanking-out only occurs on numeric variables\n' )
# cat( 'categorical variable blanks are usually 9 in the pnad\n' )
# cat( 'thanks for listening\n' )
# loop through each row in the missing household-level codes table
for ( curRow in seq( nrow( household.nr ) ) ){
# if the variable is in the current table..
if( household.nr[ curRow , 'variable' ] %in% names( dom_df ) ){
# ..and the variable should be recoded for that year
if( catalog[ i , 'year' ] %in% eval( parse( text = household.nr[ curRow , 'year' ] ) ) ){
dom_df[ dom_df[ , household.nr[ curRow , 'variable' ] ] %in% household.nr[ curRow , 'code' ] , household.nr[ curRow , 'variable' ] ] <- NA
gc()
}
}
}
# loop through each row in the missing person-level codes table
for ( curRow in seq( nrow( person.nr ) ) ){
# if the variable is in the current table..
if( person.nr[ curRow , 'variable' ] %in% names( pes_df ) ){
# ..and the variable should be recoded for that year
if( catalog[ i , 'year' ] %in% eval( parse( text = person.nr[ curRow , 'year' ] ) ) ){
pes_df[ pes_df[ , person.nr[ curRow , 'variable' ] ] %in% person.nr[ curRow , 'code' ] , person.nr[ curRow , 'variable' ] ] <- NA
gc()
}
}
}
if( catalog[ i , 'year' ] >= 2001 ){
pes_df$uf <- substr( pes_df$v0102 , 1 , 2 )
pes_df$region <- substr( pes_df$v0102 , 1 , 1 )
}
pes_df$one <- 1
x <- merge( dom_df , pes_df )
stopifnot( nrow( x ) == nrow( pes_df ) ) ; rm( pes_df , dom_df ) ; gc()
# determine if the table contains a `v4619` variable.
# v4619 is the factor of subsampling used to compensate the loss of units in some states
# for 2012, the variable v4619 is one and so it is not needed.
# if it does not, create it.
if( !( 'v4619' %in% names( x ) ) ) x$v4619 <- 1
x$v4609 <- as.numeric( x$v4609 )
if( catalog[ i , 'year' ] < 2004 ){
x$pre_wgt <- as.numeric( x$v4610 )
} else {
x$pre_wgt <- as.numeric( x$v4619 ) * as.numeric( x$v4610 )
}
catalog[ i , 'case_count' ] <- nrow( x )
saveRDS( x , file = catalog[ i , 'output_filename' ] ) ; rm( x ) ; gc()
# delete the temporary files
suppressWarnings( file.remove( tf , unzipped_files ) )
cat( paste0( data_name , " catalog entry " , i , " of " , nrow( catalog ) , " stored at '" , catalog[ i , 'output_filename' ] , "'\r\n\n" ) )
}
on.exit()
catalog
}
# initiate a function that removes the "UF" field from the PNAD SAS importation script,
# because the R `SAScii` package does not currently handle overlapping fields and this UF column always overlaps
pnad_remove_uf <-
function( sasfile ){
# read the SAS import file into R
##does not work on linux sascon <- file( sasfile , "rb" , blocking = FALSE , encoding = "windows-1252" )
sas_lines <- readLines( sasfile , encoding="windows-1252")
sas_lines <- iconv( sas_lines , "windows-1252" , "ASCII//TRANSLIT" )
if (is.na(sas_lines[1])) {
sas_lines <- readLines( sasfile , encoding="latin1")
sas_lines <- iconv( sas_lines , "latin1" , "ASCII//TRANSLIT" )
}
##close( sascon )
# remove any TAB characters, replace them with two spaces
sas_lines <- gsub( "\t" , " " , sas_lines )
# throw out any lines that contain the UF line
sas_lines <- sas_lines[ !grepl( "@00005[ ]+UF[ ]" , sas_lines ) ]
# fix the "$1 ." difference seen in a number of PNAD SAS importation scripts
# by simply switching the space and the period
sas_lines <-
gsub(
"@00840 V2913 $1 ." ,
"@00840 V2913 $1. " ,
sas_lines ,
fixed = TRUE
)
# fix the duplicate column names in 2007
sas_lines <-
gsub(
"00965 V9993" ,
"00965 V9993A" ,
sas_lines ,
)
# create a temporary file
tf <- tempfile()
# write the updated sas input file to the temporary file
writeLines( sas_lines , tf )
# return the filepath to the temporary file containing the updated sas input script
tf
}
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.