################################################################################
### R BASICS WORKSHOP                                                        ###
### PRESENTATION 9: FLOW CONTROL                                             ###
###                                                                          ###
### Center for Conservation and Sustainable Development                      ###
### Missouri Botanical Garden                                                ###
### Website: rbasicsworkshop.weebly.com                                      ###
################################################################################


################################################################################
### 1. LOOPS: *while* ##########################################################
################################################################################

# *while* is also a very useful way to construct loops. This kind of loop
# repeats a piece of code while a particular condition is TRUE.

# The general structure of a *while* loop is:
#
# while(condition)
# {
#   code
# }

# This translates approximately to:
#
# while condition is TRUE, repeat:
# {
#   this code
# }


####################
## Easy example 1 ##
####################
v <- 1:10

i <- 0

# Version 1
while(i < max(v))
{
	i <- i+1
	print(i)
}


####################
## Easy example 2 ##
####################
Bp <- 0.1
Dp <- 0.12
Np <- 1-Bp-Dp

max.t <- 50
time <- 0
abund <- 10

plot(x=c(0, max.t), y=c(0, 100), type="n", xlab="Time", ylab="Abundance")

while(abund>0 & time<=max.t)
{
	change <- sample(c(-1,0,1), size=abund, prob=c(Dp, Np, Bp), replace=TRUE)

	abund <- abund + sum(change)
	time <- time + 1

	points(time, abund, pch=16, cex=1, col="grey65")
}



################################################################################
### 2. CONDITIONALS: *if* ######################################################
################################################################################

# The conditional *if* allows you to run a piece of code only if a condition is
# TRUE


####################
## Easy example 1 ##
####################
v <- 1:10

for(i in v)
{
	print(i)

	if(i == 5)
	{
		print("Reached 5")
	}
}


####################
## Easy example 2 ##
####################
## THE WHILE DATING LOOP ##

you <- runif(1, 0, 100) # Your personality in a number
your.pickiness.score <- 0.1 # How closely you want your date 
                            # to match your personality

missmatch <- Inf # The initial difference in personality 
date <- 0 # Your date number

while(missmatch > your.pickiness.score)
{
  date <- date + 1

	they <- runif(1, 0, 100)
	missmatch <- abs(you - they)


	if(missmatch <= your.pickiness.score)
	{
	  print("Congratulations, you are getting married!!")
	  print(paste("You've dated", date, "people", sep=" "))
	}

}



################################################################################
### 3. BREAKS: *break* #########################################################
################################################################################

# The function *break* causes a loop to stop. Breaks are frequently used in
# combination with a conditional *if*.

####################
## Easy example 1 ##
####################
v <- 1:10

for(i in v)
{
	print(i)

	if(i == 5) # With more than one command, it is necessary to use *{}*
	{
		print("Reached 5")
		break()
	}
}


####################
## Easy example 2 ##
####################
Bp <- 0.1
Dp <- 0.12
Np <- 1-Bp-Dp

max.t <- 50
time <- 0
abund <- 10

plot(x=c(0, max.t), y=c(0, 100), type="n", xlab="Time", ylab="Abundance")

for(i in 1:max.t)
{
	change <- sample(c(-1,0,1), size=abund, prob=c(Dp, Np, Bp), replace=TRUE)

	abund <- abund + sum(change)
	time <- time + 1

	if(abund<=0)
	{
		warning("Population has gone extinct")
		break()
	}

	if(sum(change) > 0)
	    COL <- "red"

	if(sum(change) < 0)
	    COL <- "blue"

	if(sum(change) == 0)
	    COL <- "grey65"

	points(time, abund, pch=16, cex=1, col=COL)
}



################################################################################
### 4. AVOIDING LOOPS ##########################################################
################################################################################

# Loops are extremely useful constructs in R, however they can be slow. If you
# are having problems with scripts that take too long to run, it is a good idea
# to vectorize: avoid loops and use pre-existing functions in R that work on
# vectors.

# A blog describing vectorization:
# http://www.noamross.net/blog/2014/4/16/vectorization-in-r--why.html


## Easy example 1 ##

x <- seq(1, 10, length.out=10000000)


# Unvectorized version
lsum <- 0
for(i in 1:length(x))
{
	lsum <- lsum + log(x[i])
}
lsum


# Vectorized version
lsum <- sum(log(x))
lsum



## Easy example 2 ##

# Lets assume we have a very large matrix of abundances of 50 species across
# 1,000,000 sites

M <- matrix(rpois(50000000, 10), ncol=50)

dim(M)

head(M)

# How can we calculate the total abundance of individuals at each site?

## Option 1 - a very inefficient loop ##
abund.1 <- numeric() # This creates a numeric vector of length 0

# system.time will help us calculate the time something takes
system.time(
{

  for(i in 1:nrow(M))
  {
	  abund.1 <- c(abund.1, sum(M[i,]))
  }

})


## Option 2 - a slightly better loop by initializing the vector that
## stores the results ##
abund.2 <- rep(NA, nrow(M)) # This creates a vector of the needed length but
							# full of NAs

system.time(
{

  for(i in 1:nrow(M))
  {
	  abund.2[i] <- sum(M[i,])
  }

})


## Option 3 - using the function *apply* hides the loop but does not help ##
system.time(
{
  abund.3 <- apply(M, 1, sum)
})

?apply


## Option 4 - a fully vectorized version ##
system.time(
{
  abund.4 <- rowSums(M)
})