In enpjp/PrepareDataForETL: Prepares Data for Extract Transform and Load

# Set code chunk options for all chunks
# These can be overridden at the chunk level, but setting global options 
# ensures consistency of chunk behaviour.
# To print a version of this document without code set echo = FALSE. 
#Include false will ignore the code chunk completely!

knitr::opts_chunk$set(include = TRUE, 
                      echo = FALSE, 
                      warning=FALSE, 
                      message=FALSE, 
                      error=FALSE, 
                      results='asis',
                      out.width='\\textwidth') 
# Globally set the figure width here in inches.


library(rgdal)
library(ggmap)
library(rgeos)
library(maptools)
library(readxl)
#library(devtools)
#library(roxygen2)
library(tidyverse)
library(PrepareDataForETL)

Load data

Start by loading the package data:

# Uncomment this line to load the package default data
my.data <- data("Green.Carpet.Data")
# Adapt this by using the import function in R Studio and cut and paste
# The exact command from the console. Use the interactive menu to set data types
# such as dates.
# my.data <- read_excel("Data.xlsx") 
# my.data <- GreenCarpetDerbys <- read_excel("GreenCarpetDerbys.xlsx", 
#      col_types = c("date", "numeric", "text", 
#          "text", "text", "text", "text", "text", 
#         "numeric"))

summary(my.data)
# Use view to check the data
# Rename any columns using this type of syntax
colnames(my.data)[5] <- c("Grid.Reference")
# Keep the named columns
temp.keep <- c("Date", "Year", "Grid.Reference")
# Do it this way so you do not make a mistake
temp.my.data <- my.data[,temp.keep]
my.data <- temp.my.data

\newpage

Clean the dates

PrepareDataForETL has some helper functions: Clean.Dates takes in a variety of date formats, including Excel, and produces a standard set of ISO format dates with exactly the same number of rows:

my.clean.dates <- Clean.Dates(my.data$Date) %>% as.data.frame()
# Now check which rows have valid dates
rows.with.dates <- grepl("\\d{4}-\\d{2}-\\d{2}",my.clean.dates$YYMMDD )
# Make a list of those rows
rows.to.keep <- which(rows.with.dates)
# Bind columns of darta
my.data <- cbind(my.data, my.clean.dates)

# Only keep those rows with data.
my.data <- my.data[rows.to.keep,]

\newpage

Including Plots

We can now produce a phenology plot of all the records using the day of year DOY

 temp.my.plot <- 
  ggplot(my.data,aes(x=as.numeric(my.data$DOY) ) ) + 
  geom_density(fill="Red", bw = 4) +
   labs(title="Phenology",x="Day of Year", y = "Density") +
   xlim(1, 365) # Set the x-axis limits.

temp.my.plot

# This line saves the plot in an images subdirectory
dir.create("images")
ggsave("images/Phenology-density.pdf", device = "pdf")

# Clean up
rm(list = ls(pattern = "temp*"))

Now to plot a map of all the records

library(rgeos)
# The current function for getting the gridreferences as co-ordinates is fussy
# so suppress the warnings.
# we need to supply an outline map. This is one I prepared earlier
data("leics")
# Check the documentation for the source of the map data

Plot.Maps("test name", my.data$Grid.Reference, leics)