import_slides.utf8

class: center, middle, inverse, title-slide

# <div class="line-block">Getting Started in R<br />
an introduction to data analysis and visualisation</div>
## Import
### Réka Solymosi & Sam Langton
### 2 July 2019

---

class: inverse, center, middle

# Import

---
### Sources of data

- Files stored on your computer
- Web link
- Database
- Web page
- Application Programming Interface (API)

---
### Data formats

.pull-left[
#### Rectangular
- CSV (.csv)
- Excel files (.xls, .xlsx)
- SPSS, Stata, and SAS files (.sav, .dta, .sas)
]

.pull-right[
#### Hierarchical
- eXtensible Markup Language (.xml)
- JavaScript Object Notation (.json)
- Geographic JavaScript Object Notation (.geojson)
]

---
### Reading in data

.pull-left[
#### Rectangular
#### .csv

```r
library(readr)
read_csv()
```

#### .xls, .xlsx

```r
library(readxl)
read_excel()
```

#### .sav, .dta, .sas

```r
library(haven)
read_sav()
read_dta()
read_sas()
```
]

.pull-right[
#### Hierarchical
#### .xml
```r
library(xml2)
read_xml()
```

#### .json
```r
library(jsonlite)
read_json()
```

#### .geojson
```r
library(sf)
st_read()
```
]

---
class: inverse, middle, center

# Parse

---
### Parsing data
The process of ...

1. converting a text file into a matrix of rows and columns and,
2. coercing columns to different data type(s)

---
#### .csv, .xls, .xlsx

```r
library(readr)
read_csv(
"country,continent,year,lifeExp,pop,gdpPercap\
Italy,Europe,2007,80.546,58147733,28569.7197\
Nigeria,Africa,2007,46.859,135031164,2013.977305"
)
```

```
# # A tibble: 2 x 6
#   country continent  year lifeExp       pop gdpPercap
#   <chr>   <chr>     <dbl>   <dbl>     <dbl>     <dbl>
# 1 Italy   Europe     2007    80.5  58147733    28570.
# 2 Nigeria Africa     2007    46.9 135031164     2014.
```

---
#### .xml
.pull-left[

```r
library(xml2)
xml <- 
'<?xml version="1.0" encoding="UTF-8"?>
<root>
  <row>
    <country>Iceland</country>
    <continent>Europe</continent>
    <year>2007</year>
    <lifeExp>81.757</lifeExp>
    <pop>301931</pop>
    <gdpPercap>36180.78919</gdpPercap>
  </row>
  <row>
    <country>Italy</country>
    <continent>Europe</continent>
    <year>2007</year>
    <lifeExp>80.546</lifeExp>
    <pop>58147733</pop>
    <gdpPercap>28569.7197</gdpPercap>
  </row>
</root>'
```
]
.pull-right[

```r
*xml_obj <- read_xml(xml)
*country <- xml_text(xml_find_all(xml_obj, "//country"))
*continent <- xml_text(xml_find_all(xml_obj, "//continent"))
*year <- xml_text(xml_find_all(xml_obj, "//year"))
*lifeExp <- xml_text(xml_find_all(xml_obj, "//lifeExp"))
*pop <- xml_text(xml_find_all(xml_obj, "//pop"))
*gdpPercap <- xml_text(xml_find_all(xml_obj, "//gdpPercap"))
*data.frame(country, continent, year, lifeExp, pop, gdpPercap)
```

```
#   country continent year lifeExp      pop   gdpPercap
# 1 Iceland    Europe 2007  81.757   301931 36180.78919
# 2   Italy    Europe 2007  80.546 58147733  28569.7197
```
]

---
#### .json

```r
library(jsonlite)
json <- 
'[
  {
    "country": "Italy",
    "continent": "Europe",
    "year": 2007,
    "lifeExp": 80.546,
    "pop": 58147733,
    "gdpPercap": 28569.7197
  },
  {
    "country": "Nigeria",
    "continent": "Africa",
    "year": 2007,
    "lifeExp": 46.859,
    "pop": 135031164,
    "gdpPercap": 2013.977305
  }
]'
*fromJSON(json)
```

```
#   country continent year lifeExp       pop gdpPercap
# 1   Italy    Europe 2007  80.546  58147733 28569.720
# 2 Nigeria    Africa 2007  46.859 135031164  2013.977
```

---
#### .geojson
.pull-left[

```r
library(sf)
*italy <- st_read("https://code.highcharts.com/mapdata/countries/it/it-all.geo.json", quiet = TRUE)
as.data.frame(italy)[1:5, 14:15]
```

```
#        name country
# 1    Napoli   Italy
# 2   Trapani   Italy
# 3   Palermo   Italy
# 4   Messina   Italy
# 5 Agrigento   Italy
```

```r
plot(st_geometry(italy))
```

![](import_slides_files/figure-html/unnamed-chunk-5-1.png)
]

.pull-right[

```r
*nigeria <- st_read("https://code.highcharts.com/mapdata/countries/ng/ng-all.geo.json", quiet = TRUE)
as.data.frame(nigeria)[1:5, 14:15]
```

```
#      name country
# 1  Rivers Nigeria
# 2 Katsina Nigeria
# 3  Sokoto Nigeria
# 4 Zamfara Nigeria
# 5    Yobe Nigeria
```

```r
plot(st_geometry(nigeria))
```

![](import_slides_files/figure-html/unnamed-chunk-6-1.png)
]

---
#### .csv, .xls, .xlsx

```r
library(readr)
df <- read_csv(
"country,continent,year,lifeExp,pop,gdpPercap\
Italy,Europe,2007,80.546,58147733,28569.7197\
Nigeria,Africa,2007,46.859,135031164,2013.977305",
*col_types = cols(
* country = col_character(),
* continent = col_character(),
* year = col_integer(),
* lifeExp = col_double(),
* pop = col_integer(),
* gdpPercap = col_double()
* )
)
glimpse(df)
```

```
# Observations: 2
# Variables: 6
# $ country   <chr> "Italy", "Nigeria"
# $ continent <chr> "Europe", "Africa"
# $ year      <int> 2007, 2007
# $ lifeExp   <dbl> 80.546, 46.859
# $ pop       <int> 58147733, 135031164
# $ gdpPercap <dbl> 28569.720, 2013.977
```

---
# Export

---
### Writing out data

.pull-left[
#### Rectangular
#### .csv

```r
library(readr)
write_csv()
```

#### .xls, .xlsx

```r
library(readxl)
write_excel()
```

#### .sav, .dta, .sas

```r
library(haven)
write_sav()
write_dta()
write_sas()
```
]

.pull-right[
#### Hierarchical
#### .xml
```r
library(xml2)
write_xml()
```

#### .json
```r
library(jsonlite)
write_json()
```

#### .geojson
```r
library(sf)
st_write()
```
]

---
### Please note

- R stores factors as integer codes 1, 2, 3 etc. not character strings. So ensure that you set your factor levels.
- Column type information is lost when you export to a CSV. Recommend exporting to .Rds, the native R data format using `write_rds()`

---