najsaqib / lfsdataviewer Goto Github PK

Need to add stuff so that all parts of the document that rely on CANSIM tables can be viewed in the app.

In the current set-up, what is effectively going on is that once the data is imported, the character variable regarding the date is turned into an explicit date variable through:

df$Date <- as.Date(as.yearmon(df$t, format="%Y/%m"))

Once all the operations are done that depend on there being an explicit date variable (i.e. creating a new variable for each observation that contains the value 1 month, or 12 month ago), the date variable is being turned back into character through this:

# Round, and then suppress scientific notation
df <- format(df, big.mark=",", scientific=FALSE, TRIM=TRUE) 

This allows us to display statements by using the following code in the Shiny app:

paste0("Employment for men increased by: ", df %>%
        filter(Date == input$date) %>%
        filter(i==input$region) %>%
        select(EmpMen))

However, if the fields are not turned back into character, we get the following error message when trying to execute the above code:

Adding missing grouping variables: i

Need to figure out why that is so that we know whether to fix this issue, or continue to work around it by converting the data frame to character before feeding into the Shiny app.