factor range as function variable in R

Question

I have code

ggplot(bywells[bywells$Well_N == "KRT3",], aes(x = Date_m)) +
geom_line(aes(y = QOM, colour = "Oil, m3/month"))

which draw chart. This code works without any errors. Now I would like to elaborate it in function where bywells is variable data and "KRT3" is variable wellname, but if I write like this

simple_fun <- function(data, wellname, ...)
{
   require("ggplot2", quietly=TRUE)
   ggplot(data[data$Well_N == "wellname",], aes(x = Date_m)) +
   geom_line(aes(y = QOM, colour = "Oil, m3/month"))
}

after executing function I get error message

Error: Aesthetics must either be length one, or the same length as the dataProblems:QOM, Date_m

when I try

ggplot(data[wellname == Well_N %in% data,], aes(x = Date_m))

I get

Error in match(x, table, nomatch = 0L) : object 'Well_N' not found

Any hints how can I define it as a variable properly?

For example reproducibility I ad small chunk of data sample:

"Well_N";"Date_m";"QOM";"QWM";"QOMT";"BHP";"PRES";"QIW";"THPI";"QFM";"WCT"
"KRT3";2014-06-30;132;525;108;NA;NA;NA;NA;657;79
"KRT3";2014-07-30;36;120;29;NA;NA;NA;NA;156;76
"KRT3";2014-08-30;39;2.6;32.1;NA;NA;NA;NA;41.6;6.25
"KRT3";2014-09-30;211.274;749.362;174.070;NA;NA;NA;NA;960.636;78
"KRT3";2014-10-30;45;45;37.07;NA;NA;NA;NA;90;50
"KRT4";2014-08-30;108.37;1815.358;90.79;NA;NA;NA;NA;1923.73;94
"KRT4";2014-09-30;161.775;202.87;133;NA;NA;NA;NA;364;55
"KRT4";2014-10-30;30;1680;24;NA;NA;NA;NA;1710;98
"KRT4";2014-11-30;31.8;339;26;NA;NA;NA;NA;370.8;91

Type of Well_N is factor, Date_m is POSIXct, others are num.

What is the point of `wellname` argument, if you're not using it? — Roman Luštrik, Aug 06 '14 at 07:22
@RomanLuštrik, maybe I lost somewhere - I'm not a programmer. I just would like to realize that function can bee called as simple_fun(mydataframe, AnyWellName). In my data I have 93 levels of factor Well_N. — dainys, Aug 06 '14 at 07:29
Here are a few hints on how to make your question reproducible. http://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example How do you run the function? — Roman Luštrik, Aug 06 '14 at 07:32
@RomanLuštrik, I added "required". Is it OK now? Function could be run mannually, or called by another function to go through all or range of factors and output saved as separate png or composed pdf file if several factor plots generated. — dainys, Aug 06 '14 at 07:41
Read the thread I posted earlier. There are a lot of tips on how to provide us with data. Take in consideration that we are lazy beings. — Roman Luštrik, Aug 06 '14 at 07:47
Try using aes_string rather than aes for ggplot in function and call function with variable names as strings ("well1"). Google for aes_string and ggplot2 — rnso, Aug 06 '14 at 09:30
There is only one date in example data: 2014-06-30 . One cannot try a graph with that. — rnso, Aug 06 '14 at 09:32
Take out the quotes around `"wellname"`. Otherwise `data[data$Well_N == "wellname",]` returns a `data.frame` of 0 rows. — shadow, Aug 06 '14 at 09:43
@shadow, then I get error `Error in data$Well_N : $ operator is invalid for atomic vectors` @mso, thanks for advice, I'll do some googling and inform on progress. — dainys, Aug 06 '14 at 10:12
How are you calling the function? For me `simple_fun(bywells, "KRT3")` works. Of course `simple_fun("bywells", "KRT3")` would give me the same error you seem to get. — shadow, Aug 06 '14 at 10:21

score 1 · Accepted Answer · answered Aug 07 '14 at 07:56

Here is the reproducible data (I've left only three columns which are needed, use function dput in your future questions):

bywells <-
structure(list(Well_N = structure(c(1L, 1L, 1L, 1L, 1L, 2L, 2L, 
2L, 2L), .Label = c("KRT3", "KRT4"), class = "factor"), Date_m = structure(c(16251, 
16281, 16312, 16343, 16373, 16312, 16343, 16373, 16404), class = "Date"), 
    QOM = c(132, 36, 39, 211.274, 45, 108.37, 161.775, 30, 31.8
    )), class = "data.frame", row.names = c(NA, -9L), .Names = c("Well_N", 
"Date_m", "QOM"))

Here is the call without the function:

library(dplyr)
library(ggplot2)
library(magrittr)
qplot(x=Date_m, y=QOM, data = bywells %>% filter(Well_N == "KRT3"), geom="line")

Here is the function:

pfun <- function(data,wellname) {
    qplot(x=Date_m, y=QOM, data=data %>% filter(Well_N == wellname), geom="line")
}

with the corresponding call:

pfun(bywells,"KRT3")

Your error was to put the quotes on the variable wellname in the condition which filters the data. Also your colour setting is not ok, since there is no such color called "Oil, m3/month". Aim for English words such as "red", "blue", see colors() for more variants. Colour should be set outside aes argument in order for it to work.

factor range as function variable in R

1 Answers1