Native language issue

[Hayk]

Hi,

I have a tibble with dates in wide format. I transposed this tibble so that I have a column of dates as a variable and a column of values with melt function (reshape package) as well as with pivot_longer function (tidyr package). In both cases the character values which were written in my native language (Armenian) messed up and now there are weird symbols in their place.

[nirgrahamuk]

How do you come to have the tibble in question?
You constructed it by typing it into your code? If so confirm that the script encoding is UTF-8.
If you imported from a file such as a CSV. Confirm that the CSV is in UTF-8

[Hayk]

I have read it from an excel file. okay and how can I confirm script encoding or check it?

[nirgrahamuk]

what R command did you use ?

[Hayk]

R command? R studio? Do you mean that?

[nirgrahamuk]

What code did you write

[Hayk]

Ah yes, sorry. here is my code.

library(readxl)

library(reshape2)

setwd("D:/M&L_Risks/Bank_Performance_Analysis")

excel_sheets("Working_X.xlsx")

df_db <- read_excel("Working_X.xlsx", sheet = "Accounts_DB")
col <- colnames(df_db[,1:18])

m_db <- iconv(df_db,from = "Latin-1", to = "UTF-8")

m_db <- melt(df_db, col, na.rm = TRUE, value.name = "Value" )

[Hayk]

And by the way the code written as
iconv(df_db,from = "Latin-1", to = "UTF-8") doesn't work.
It generates an error - unsupported conversion from Latin-1 to UTF-8

[Hayk]

Encoding to UTF-8

library(readxl) library(reshape2) setwd("C:/R-studio") df_db <- read_excel("Working_X.xlsx", sheet = "Accounts_DB") col <- colnames(df_db[,1:18]) m_db <- melt(df_db, col, na.rm = TRUE, value.name = "Value" ) Created on 2020-05-10 by the reprex package (v0.3.0) After reading an excel file I get a tibble on which I can see the characters written in my native language(Armenian). But after I melt the tibble, the characters become messed. I guess the issue is about encoding, and I need to change encoding to UTF-8, but whatever I searched for and tried I could not manage to conduct the change.

Here is the topic I have created on this issue.

[nirgrahamuk]

iconv() wants to convert character vectors, it wont work for dataframes like you are trying to use it.

Were you trying iconv because you know your excel file is in Latin-1 encoding ? or was that a guess on your part ?
If you have excel perhaps you can open the file and resave it with UTF-8 encoding ?

Perhaps you can raise an issue, or add to an existing issue on readxl github.

Can't read xls with non UTF-8 encoding