Importación de datos tabulares: resolver problemas de importación usando type_convert, preservar datos con write_feather para facilitar su intercambiarlos, obtener datos de internet con download.file y unzip, o obtenerlos de base de datos con DBI.

importar.R

if(!require(tidyverse)) install.packages("tidyverse")
if(!require(feather)) install.packages("feather")
if(!require(RSQLite)) install.packages("RSQLite")
library(tidyverse)
library(hms)
library(haven)
library(feather)
library(readxl)
library(DBI)

# Análisis sintáctico de vectores
str(parse_guess(c("2010-10-10","2021-10-10")))
x <- parse_logical(c("TRUE", "FALSE", "N"))
problems(x)
str(parse_logical(c("TRUE", "FALSE", "N"), na = c("","N")))


# Análisis sintáctico de números
parse_double("3333.99")
parse_number("Precio: 3.333,99€", locale = locale(decimal_mark = ",", grouping_mark = "."))


# Análisis sintáctico de cadenas de carácteres
x1 <- "El niño come galletas"
guess_encoding(charToRaw(x1)) # UTF-8
parse_character(x1)
parse_character(x1, locale = locale(encoding = "Latin1"))

x2 <- "El ni\xf1o come galletas"
guess_encoding(charToRaw(x2)) # ISO-8859-1
parse_character(x2)
parse_character(x2, locale = locale(encoding = "Latin1"))

# Análisis sintáctico de fechas y horas
parse_date("2020-01-13")
mon <- c("Jänner", "Februar", "März", "April", "Mai", "Juni", "Juli", "August", "September", "Oktober", "November", "Dezember")
monab <- c("Jän", "Feb", "Mrz", "Apr", "Mai", "Jun", "Jul", "Aug", "Sep", "Okt", "Nov", "Dez")
day <- c("Sonntag", "Montag", "Dienstag", "Mittwoch", "Donnerstag", "Freitag", "Samstag")
dayab <- c("So", "Mo", "Di", "Mi", "Do", "Fr", "Sa")
deAT <- date_names(mon, monab, day, dayab)
parse_date("13 Jänner 2020", "%d %B %Y", locale = locale(date_names = deAT))
parse_time("19:30")
parse_time("7:30 pm")
parse_datetime("2020-01-13T19:30:30Z")
parse_datetime("2020-01-13T19:30:30Z", "%Y-%m-%dT%H:%M:%SZ")
parse_datetime("2020-01-13T20:30:30+01:00")
parse_datetime("2020-01-13T20:30:30+01:00", "%Y-%m-%dT%H:%M:%S%z")


# Encontrando problemas
challenge <- read_csv(readr_example("challenge.csv"))
problems(challenge)

# Resolviendo prolemas
challenge2 <- read_csv(readr_example("challenge.csv"),  col_types = cols(.default = col_character()))
type_convert(challenge2)

# Problema resuelto
challenge <- read_csv(
  readr_example("challenge.csv"), 
      col_types = cols(
        x = col_double(),
        y = col_date() # En vez de col_logical()
      )
)
stop_for_problems(challenge)

  
# Guardando los datos
challenge <- read_csv(
  readr_example("challenge.csv"), 
      col_types = cols(
        x = col_double(),
        y = col_date() # En vez de col_logical()
      )
)
write_csv(challenge, "challenge.csv")
read_csv("challenge.csv")
write_feather(challenge, "challenge.feather")
read_feather("challenge.feather")


# Leer un fichero .xlsx desde Internet
download.file("https://mossos.gencat.cat/.content/home/01_els_mossos_desquadra/indicadors_i_qualitat/Dades_obertes/fitxers/Fets_2020.xls","~/R/data/Fets_2020.xls")
delitos <- read_excel("~/R/data/Fets_2020.xls", skip = 3)
glimpse(delitos)

# Leer microdatos .sav desde .zip
codigo <- 3283
zip_file <- paste("~/R/data/cis/MD",codigo,".zip", sep="")
sav_file <- paste(codigo,".sav", sep="")
url <- paste("http://datos.cis.es/webFtp/redireccionador.jsp?dwld=/Microdatos/MD",codigo,".zip&Email=domain@domain.com&Terminos=1", sep="")
download.file(url, zip_file)
sav_filepath <- unzip(zipfile=zip_file, files = sav_file, exdir=tempdir())
ceo <- read_sav(sav_filepath)
glimpse(ceo)

# Trabajar con SQLLite
con <- dbConnect(RSQLite::SQLite(), ":memory:")
dbWriteTable(con, "mtcars", mtcars)
dbListTables(con)
dbListFields(con, "mtcars")
res <- dbSendQuery(con, "SELECT * FROM mtcars WHERE cyl = 4")
dbFetch(res)
dbClearResult(res)
dbDisconnect(con)