If I have a table of N columns and I want to access individual columns
by name (instead of index), what is the best practice in this case? I
could define the column names alongside the table, like for example
employees ← 0 3 ⍴ ⍬
(nameCol phoneCol addressCol) ← ⍳3
The problem with this approach is that the scope of the column names is
too large; I may have another table in the same scope which also has a
"name" column. To solve this problem I could define a namespace (using
Dyalog APL's function ⎕NS) and define the elements and the column names
inside it, but then accessing the column names is more verbose since
they need to be prefixed with the namespace. Any ideas? How do you write
maintainable APL programs involving tables?
By the way, are there any documents available online with best practices
for programming in the large with APL? All articles I have read about
APL is more or less focused on using it as an advanced calculator.
Regards,
August
|
|