API planned updates/upgrades for 2018 about comtradr HOT 7 CLOSED

FYI @jugdemon , API is back up and running.

from comtradr.

Good to know that it is not just me. In the meantime I used the manual interface of the worldbank to comtrade to download excel files with the necessary information. Apparently, they still have access (or more likely they run a replica). It looks like a cool package and I will use it the next time I have to access comtrade.

from comtradr.

A quick question, could it be, that the upgrade has rendered comtradr not working? When I make any queries, they always return empty (also on the comtrade.un.org).

from comtradr.

Ah, yeah that's interesting, I had no idea the API was down, thank you for bringing it to my attention. I'm not sure what's causing it. It's odd that their online data interface doesn't indicate the service being down or anything like that.

I'll keep my eyes on this, if the outage persist for more than today I will send them a bug report email asking what's going on.

Thanks!

from comtradr.

Hello Chris Muir,

I did not find a specific topic for my question. I would like to know if there is any command in comtradr to use country codes instead of the country names in the ct_search command.

Thank you for understanding

from comtradr.

Hello @wallacemp , nope the query function ct_search() does not support passing country codes as input. You can, however, create your own wrapper function, like I've demoed below, that will allow you to pass in country codes, and will convert the inputs to their character country names prior to calling ct_search(). Be aware, this is a super hacky fix, it uses comtradr:::get_country_db(), which is an undocumented non-exported function. Also, there are no checks to ensure the input ints are valid country codes, although you could easily add those checks to the wrapper function below. Hope this is helpful!

ct_search2 <- function(reporters, partners, ...) {
  
  if (!is.character(reporters)) {
    reporters <- as.character(reporters)
  }
  if (!is.character(partners)) {
    partners <- as.character(partners)
  }
  
  country_df <- comtradr:::get_country_db()
  
  reporters <- vapply(reporters, function(x) {
    country_df[country_df$code == x &
                 country_df$reporter == TRUE, ]$country_name
  }, character(1), USE.NAMES = FALSE)
  
  partners <- vapply(partners, function(x) {
    country_df[country_df$code == x &
                 country_df$partner == TRUE, ]$country_name
  }, character(1), USE.NAMES = FALSE)
  
  return(comtradr::ct_search(reporters, partners, ...))
}

df <- ct_search2(reporters = 156, 
                 partners = c(276, 288, 392, 484), 
                 trade_direction = "imports", 
                 start_date = 2010, 
                 end_date = 2014)

dim(df)
#> [1] 20 35

The example above is equivalent to doing:

df <- ct_search(reporters = "China", 
                partners = c("Germany", "Ghana", "Japan", "Mexico"), 
                trade_direction = "imports", 
                start_date = 2010, 
                end_date = 2014)

from comtradr.

Closing this issue, as changes will be adressed in planned relaunch. #45

from comtradr.