Well I never, bless my soul, so it does. (In this case...) I stand corrected!

On Windows, parameters are passed to executables as a single string, so the list forms of system & open etc., join the list into a single string before calling CreateProcess().

There is some intelligence involved in the concatenation, but in many, (by my experience: most) cases, that 'intelligence' flies in the face of expectation, and gets it wrong.

By way of example of the inconsistent nature of the internal handling of the list form:

system 'perl', '-E', 'say 12345;';; ## Works!
12345

system 'perl', '-E', 'say "hello world";';; ## Doesn't work.

system 'perl', '-E', 'say \"hello world\";';; ## Doesn't work.

system 'perl', '-E', '"say \"hello world\";"';; ## Works!
hello world

system 'perl', '-E', '\"say \"hello world\";\"';; ## Doesn't work,
Can't find string terminator '"' anywhere before EOF at -e line 1.
[download]

By way of example of the inconsistent nature of the internal handling of the list form:

Apparently perl isn't to blame, see exec always invokes the shell? win32

Win32::ShellQuote can be a cure

I'm not sure how you draw that conclusion.

1. system is not exec; nor vice versa.

   It is perfectly feasible and simple to invoke a command via system without going through the shell:

   [0] Perl> system q[perl -wE" say for @ARGV; <STDIN>" 1 2 "1 2" "1 \"2\
   +" 3"];;
   1
   2
   1 2
   1 "2" 3
   [download]

2. That thread is very muddled.

   The conclusions drawn do not appear to match up with the code executed.

As for Win32::ShellQuote; I've never found a need for it. I use the scalar form of system & open; quote as I would in the shell, and it just works. The problem with the list form is it does stuff to what you pass; but what it does is undocumented; thus you're always fighting 'intelligence' that isn't very.

---

With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday'

Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.

"Science is about questioning the status quo. Questioning authority".

In the absence of evidence, opinion is indistinguishable from prejudice. Agile (and TDD) debunked
I