pg_select fails to fully unset the data in arrayVar for each loop

[nugget]

pg_select fails to cleanly unset the data array on each iteration of the code block. It only reliably clears out array elements that exist in the SELECT statement, but neglects to clear any array elements that were added during the execution of the code block.

This can be reproduced using this code:

#!/usr/bin/env tclsh
package require Pgtcl
set db [pg_connect -connlist {dbname nugget}]
pg_select $db "SELECT generate_series(1,10) as id" buf {
if {$buf(id) == 5} {
    set buf(interloper) "this field should only be in record number 5"
}
puts [array get buf]
}

Which runs with this output:
.tupno 0 id 1 .headers id .numcols 1
.tupno 1 id 2 .headers id .numcols 1
.tupno 2 id 3 .headers id .numcols 1
.tupno 3 id 4 .headers id .numcols 1
interloper {this field should only be in record number 5} .tupno 4 id 5 .headers id .numcols 1
interloper {this field should only be in record number 5} .tupno 5 id 6 .headers id .numcols 1
interloper {this field should only be in record number 5} .tupno 6 id 7 .headers id .numcols 1
interloper {this field should only be in record number 5} .tupno 7 id 8 .headers id .numcols 1
interloper {this field should only be in record number 5} .tupno 8 id 9 .headers id .numcols 1
interloper {this field should only be in record number 5} .tupno 9 id 10 .headers id .numcols 1

I expected the buf array to be reset completely each time the code block is evaluated.

Notably, the buf array is properly unset (all values) at the conclusion of the pg_select operation.

[resuna]

This behavior is not technically wrong, and I'm not sure it's even unexpected.

[bovine]

A related bug was fixed by commit 3521d22, which could cause a crash if you happened to try to actually unset the array variable within the pg_select code body.