join: assert that by() vars have same general type (str vs num)

[sergiocorreia]

EG:

key variable id1 is str5 in master but byte in using data
key variable id3 is str12 in master but int in using data
    Each key variable -- the variables on which observations are matched -- must be of the same generic type
    in the master and using datasets.  Same generic type means both numeric or both string.

[luispfonseca]

join seems to require that key variables are of the same generic type. Is this related to the hashing algorithm? It seems to be fine with both being strings or both being numeric but not one of each.

Minimal example:

* generate data
clear

* master dataset
set seed 09092019
set obs 10
gen string_id = "A"
replace string_id = "B" if _n > 5
gen number_id = int(runiform() * 2)
save temp_master_dataset, replace

* using dataset
duplicates drop
gen usingvar = "BLA" * (number_id + 1)
tostring number_id, gen(number_id_tostring)
save temp_using_dataset, replace

use master, clear
* error when joining
join, from(temp_using_dataset) by(string_id number_id)

* case with no error when using only strings
tostring number_id, gen(number_id_tostring)
join usingvar, from(temp_using_dataset) by(string_id number_id_tostring)

Thank you for the package!