Printing in Tabular format in TCL/PERL - file

Question

I have a script in tcl in which a variable gets a collection of data in every loop and appends in a file. Suppose in loop1 ,
$var = {xy} {ty} {po} {iu} {ii}
and in loop2
$var = {a} {b} {c} {d1} {d2} {e3}
Now in a file f.txt the variable in dumped. Like puts $file $var. And in file it comes like this:
Line number 1: {xy} {ty} {po} {iu} {ii}
Line number 2: {a} {b} {c} {d1} {d2}
I want to print them finally in a file in tabular format. Like below:
xy a
ty b
po c
iu d1
ii d2

Answer 1

First, read the file in and extract the words on the first two lines:
set f [open "f.txt"]
set words1 [regexp -all -inline {\S+} [gets $f]]
set words2 [regexp -all -inline {\S+} [gets $f]]
close $f
The trick here is that regexp -all -inline returns all matching substrings, and \S+ selects non-whitespace character sequences.
Then, because we're producing tabular output, we need to measure the maximum size of the items in the first list. We might as well measure the second list at the same time.
set len1 [tcl::mathfunc::max {*}[lmap w $words1 {string length $w}]]
set len2 [tcl::mathfunc::max {*}[lmap w $words2 {string length $w}]]
The lmap applies a string length to each word, and then we find the maximum of them. {*} substitutes the list (of word lengths) as multiple arguments.
Now, we can iterate over the two lists and produce formatted output:
foreach w1 $words1 w2 $words2 {
puts [format "%-*s %-*s" $len1 $w1 $len2 $w2]
}
The format sequence %-*s consumes two arguments, one is the length of the field, and the other is the string to put in that field. It left-aligns the value within the field, and pads on the right with spaces. Without the - it would right-align; that's more useful for integers. You could instead use tab characters to separate, which usually works well if the words are short, but isn't so good once you get a wider mix of lengths.
If you're looking to produce an actual Tab-Separated Values file, the csv package in Tcllib will generate those fine with the right (obvious!) options.

Answer 2

Try this:
$ perl -anE 'push #{$vars[$_]}, ($F[$_] =~ s/^[{]|[}]$//gr) for 0.. $#F; END {say join "\t", #$_ for #vars}' f.txt
xy a
ty b
po c
iu d1
ii d2
command line switches:
-a : Turn on autosplit on white space to #F array.
-n : Loop over lines in input file, setting the #F array to the words on the current line.
-E : Execute the following argument as a one-liner
Removing surrounding braces from each words:
$F[$_] =~ s/^[{]|[}]$//gr
g : global substitution (we want to remove both { and })
r : non destructive operation, returns the result of the substitution instead of modifying #F