Importing Modern Data - useR 2016

importing-modern-data-into-r

---
title: "Importing Modern Data into R"
author: Javier Luraschi
date: June 29, 2016
output: revealjs::revealjs_presentation
---

## Overview

* Importing Data
    + Tabular
    + Hierarchical
    + Relational
* Importing **Modern** Data
    + Services
    + Distributed
    + Binary
    
## Importing Data

## Importing Tabular Data
From Text, Excel, SPSS, SAS and Stata

```{r, eval=FALSE}
# Import from Text
library(readr)
csv <- read_csv("data/Water_Right_Applications.csv")

# Import from Excel
library(readxl)
excel <- read_excel("data/Water_Right_Applications.xls")

# Import from SPSS
library(haven)
sav <- read_sav("data/Child_Data.sav")

# Import fomr SAS
sas <- read_sas("data/iris.sas7bdat")

# Import from STATA
stata <- read_dta("data/Milk_Production.dta")
```

## Importing Hierarchical Data
From XML, HTML and JSON

```{r, eval=FALSE}
# Import from JSON
library(jsonlite)
json <- fromJSON("data/Water_Right_Applications.json")
json[[1]][[1]]

# Import from XML
library(xml2)
xml <- read_xml("data/Water_Right_Applications.xml")
xml_children(xml_children(xml))

# Import from HTML
library(rvest)
html <- read_html("data/CRAN Packages By Name.htm")
table <- xml_find_one(html, "//table")
r_versions <- html_table(table, fill = TRUE)
View(r_versions)
```

## Importing Relational Data
From Postgres, MySQL and SQLite

```{r, eval=FALSE}
library(DBI)
db <- dbConnect(RPostgres::Postgres(), user, pass, ...)

# Connect to MySQL
db <- dbConnect(RMySQL::MySQL(), user, pass, ...)

# Connect to SQLite
db <- dbConnect(RSQLite::SQLite(), dbname = "data/database.sqlite")

# Import data from SQLite
dbListTables(db)
dbGetQuery(db, "SELECT * FROM packages")

# Disconnect
dbDisconnect(db)
```

## Importing Modern Data

## Importing Services Data
From twitter and oauth apis

```{r, eval=FALSE}
library(twitteR)
# Register an application at https://aps.twitter.com
setup_twitter_oauth(
  consumer_key = "X38S4eTmAc9OflRxlx2f48Q3Z", consumer_secret = "C9MczgTnc6FUHJCcsCER4PwloRRSnaiL7WaRIojAoYkQcmZyjq",
  access_token = "19675099-S3fviiHDna9ngxjUMFatu1uhV4Sx1yJDUGtAyXX1v", access_secret = "g3i540XFtPM6UEmE4nxA0DEvAmjfX2X6gZKtD9A3bTYBd")
searchTwitter('@rstudio')

library(httr)
# Register an application at https://developers.facebook.com/apps/
myapp <- oauth_app("facebook", "353609681364760", "1777c63343eba28359537764fab99b9a")
token <- oauth2.0_token(oauth_endpoints("facebook"), myapp, type = "application/x-www-form-urlencoded")

req <- GET("https://graph.facebook.com/me", config(token = token))
str(content(req))
```

## Importing Distributed Data
From Spark

```{r, eval=FALSE}
library(sparklyr)
library(dplyr)

# Install sparklyr
install.packages("devtools")
devtools::install_github("rstudio/sparklyr")
spark_install()

# Connect and load a dataset to Spark
sc <- spark_connect(master = "local")
dataset <- spark_read_parquet(sc, "iris", "data/iris.parquet")

# Sample the dataset back to R
sampled <- dataset %>% sample_n(size = 5) %>% collect
View(sampled)

spark_disconnect(sc)
```

## Importing Binary Data
From Feather

```{r, eval=FALSE}
library(feather)

system.time({
  write_csv(nycflights13::flights, "data/flights.csv")
  read_csv("data/flights.csv")
})

system.time({
  write_feather(nycflights13::flights, "data/flights.feather")
  read_feather("data/flights.feather")
})
```

## Questions

Good one .