Creating two dfs from the data frame list-CodePudding

I have a list of data frames

a <- data.frame(V1=c("Dzienna",4.12,"czas", "2022-01-01 00:00:00", "2022-01-01 01:00:00"), 
                V2=c(NA,NA,"Moc", 0.13, 1.15)
                )
b <- data.frame(V1=c("Dzienna",5.17,"czas", "2022-02-01 00:00:00", "2022-02-01 01:00:00"), 
                V2=c(NA,NA,"Moc", 0.18, 2.05)
)
L1 <- list(a, b)
L1

I want to get two dataframes that will look like below

  Dzienna.prod       Data
1         4.12 2022-01-01
2         5.17 2022-02-01


                 Czas  Moc
1 2022-01-01 00:00:00 0.13
2 2022-01-01 01:00:00 1.15
3 2022-02-01 00:00:00 0.18
4 2022-02-01 01:00:00 2.05

I am trying to achieve the goal using lapply and do.call, but so far without success

CodePudding user response：

Assuming that every df has this pattern, you can transform each dataframe in the lists to the format you need. Also, from the tidyverse tag I assume I can use dplyr functions.

For df1 the only pattern I see is the numeric value originating from row 2, column 1 and the date originating from row 4 column 1

f1 <- function(df) {
  data.frame(t(df[c(2, 4), 1])) %>% 
    mutate(X2 =  as.Date(X2)) %>% 
    rename(Dzienna.prod = X1, Data = X2)
}

df1 <- bind_rows(lapply(L1, f1))
> df1
  Dzienna.prod       Data
1         4.12 2022-01-01
2         5.17 2022-02-01

For f1 I assume that the column Data is just the date coming from the datetimes in the dataframes.

The only pattern I recognize for the second dataframe is that it comes from the 4th and 5th row of the dataframes in the list.

f2 <- function(df) {
  df[4:nrow(L1[[1]]),] %>% rename(Czas = V1, Moc=V2)
}


df2 <- bind_rows(lapply(L1, f2))
> df2
                 Czas  Moc
1 2022-01-01 00:00:00 0.13
2 2022-01-01 01:00:00 1.15
3 2022-02-01 00:00:00 0.18
4 2022-02-01 01:00:00 2.05