0

I've searched a lot everywhere, but could not find a clear solution, despite this is, I thihk, a basic simple question. Finally wanna make clear this. :p

The basic problem is how to get datatype or convert,maybe.

ex. Type[String] -> String

Ok, the behind is.... I am trying to move a data from A_table to B_table. For preparating B_table, have to add 'name' column that is the same type with A_table as you know. Then I do

dtype = LibPQ.columntypes(execute(conn,"name"))           <-1.get the data type
alterstr = """alter talbe .... add column name $dtype"""  <-2.create add sentence
execute(conn,alterstr)                                    <-3.then done

The 1 return 'dtype=Type[String]', as you know this does not fit on SQL. I have an idea that load the A_table to DataFrames(DF), then rely on DF column data type insted of 1 process. Indeed it can be, but I believe it can be simplifed without DF. The tips in LibPQ are OK as well.(^_-)

Teach me the basic on. Many thanks.

onoke
  • 79
  • 6

1 Answers1

0

There are many ways to do the basic conversion Type{T} -> T. For example accessing the parameters field:

julia> Type{String}.parameters[1]
String

or lifting the value into the type to dispatch to a method returning the type:

type_param(t::DataType) = type_param(Val(t))
type_param(::Val{Type{T}}) where T = T

julia> type_param(Type{String})
String

this approach makes it easy to define special cases:

type_param_name(t::DataType) = type_param_name(Val(t))
type_param_name(::Val{Type{T}}) where T = string(T)
type_param_name(::Val{Type{String}}) = "text"


julia> type_param_name(Type{Int})
"Int64"

julia> type_param_name(Type{String})
"text"
ahnlabb
  • 2,017
  • 1
  • 6
  • 16
  • Thank you, ahnlabb. I could find,at least, the datatype conversion in Julias is mysterious. :) And I got the solution by your tip. Thank you again. – onoke Jul 03 '23 at 00:49