R use gsub as substr - regex

Question

I'm using H2O for some distributed computing work (via the h2o package in R). Many of the base R functions are present but I'm unable to find a suitable substitute for the substr function. I do have access to the sub and gsub functions and was hoping to possibly use some form of regex as a workaround.
I'm using the following code but not having any luck:
df1 <- data.frame(id = 1:10, var1 = seq(14102201,14103200, 100))
df1$var2 <- substr(df1$var1, 1,6)
df1$var3 <- gsub('\\d{1,8}','\\d{1,6}', df1$var1)
df1
The output in df1$var2 is what I'm looking for. Any suggestions?
EDIT:
Running this code:
library(h2o)
localH2O = h2o.init(nthreads = 2)
df1 <- data.frame(id = 1:10, var1 = seq(14102201,14103200, 100))
df1.hex <- as.h2o(localH2O , df1)
df1.hex$var2 <- substr(df1.hex$var1, 1, 6)
Gets this message:
> df1.hex$var2 <- substr(df1.hex$var1, 1, 6)
Error in as.character.default(x) :
no method for coercing this S4 class to a vector

Answer 1

Use capture groups:
gsub('(.+)..','\\1', df1$var1)
This regex matches (.+).. with df1$var1, and replace it with the substring that matches the first capture group (.+). Since there is .. at the end of the regex, the last two characters are not matched with the .+, thus they are not in the result.

Answer 2

Capture the first 6 value like so using a pattern that matches the whole sting
gsub('^(.{6}).*$','\\1', df1$var1)
A slightly more general replacement for substr(x,start,stop) is
if(start > 1)
gsub('^(.{*start-1*})(.{*stop-start+1*})).*$','\\1', 'asdfhjkl')
else
gsub('^(.{*stop*})).*$','\\1', 'asdfhjkl')
where the values between the * characters are the actual integer values of the expression. (although you'll have to make sure that nchar(x)is less than stop, otherwise the patterns won't match b/c the string is too short.)

Answer 3

The regex (?<=^.{6}).*$ matches al characters after the first 6 ones. If you want to replace substr(df1$var1, 1, 6) with sub, you can use this command:
sub('(?<=^.{6}).*$', '', df1$var1, perl = TRUE)
# [1] "141022" "141023" "141024" "141025" "141026" "141027" "141028" "141029"
# [9] "141030" "141031"
This command replaces all digits after the first 6 ones with the empty string.