Merge time series data with different length (gaps) - r

Question

I have two water flow measurement devices which give a value every minute. Now i need to merge both time series. My problem: The devices produce every couple of hour some failures. Thus, the two time series have a different length. I need to fill the gaps first. This could be done with a NA, zero value or with the leading value before the gap.
I can easily define the required time vector tseq by min and max values of the time series:
from <- as.POSIXct(min(Measurement1[[1]], Measurement1[[1]]))
to <- as.POSIXct(max(Measurement1[[1]], Measurement1[[1]]))
tseq <- as.data.frame(seq.POSIXt(from = from, to = to, by = deltaT, tz=UTC))
Then i tried to complete the two lists Measurement1 and Measurement2 with the zoo function as follows:
Measurement1Zoo <- as.data.frame(zoo(x=Measurement1, tseq[[1]]))
This leads to a df with the same length than tseq, but zoo just adds some values at the end of the vector.
I'm a bit confused how zoo works. I just want to add the missing time stamps in the two time series and complete it with NA (or another value). How could this be done? You can find two example files here:
Example time series
Thank you!

Answer 1

You can use dplyr to do an outerjoin (i.e. full_join):
library(data.table)
m1 <- fread(file = "/Measurement1.CSV", sep = ";", header = TRUE)
m1$Date <- as.POSIXct(m1$Date,format="%d.%m.%Y %H:%M",tz=Sys.timezone())
m2 <- fread(file = "/Measurement2.CSV", sep = ";", header = TRUE)
m2$Date <- as.POSIXct(m2$Date,format="%d.%m.%Y %H:%M",tz=Sys.timezone())
names(m2)[2] <- "Value 5"
min(m1$Date) == min(m2$Date) #TRUE
max(m1$Date) == max(m2$Date) #TRUE
library(dplyr)
m_all <- full_join(x = m1, y = m2, by = "Date")
nrow(m1) #11517
nrow(m2) #11520
nrow(m_all) #11520
head(m_all)
# Date Value 1 Value 2 Value 3 Value 4 Value 5
#1 2015-07-24 00:00:00 28 2 0 26 92
#2 2015-07-24 00:01:00 28 2 0 26 95
#3 2015-07-24 00:02:00 28 2 0 26 90
#4 2015-07-24 00:03:00 28 2 0 26 89
#5 2015-07-24 00:04:00 28 2 0 26 94
#6 2015-07-24 00:05:00 27 1 0 26 95
#checking NA's
sum(is.na(m1$`Value 1`)) #0
sum(is.na(m1$`Value 2`)) #0
sum(is.na(m1$`Value 3`)) #3
sum(is.na(m1$`Value 4`))#0
sum(is.na(m2$`Value 5`)) #42
sum(is.na(m_all$`Value 1`)) #3
sum(is.na(m_all$`Value 2`)) #3
sum(is.na(m_all$`Value 3`)) #6
sum(is.na(m_all$`Value 4`)) #3
sum(is.na(m_all$`Value 5`)) #42