Converting scripts to one liners: https://github.com/masak/workshop/blob/master/ONELINER.md

Oneliners.txt

# Converting to both perl5 and perl6 to see differences and similarities.  
# Note that if you have a perl > 5.14.* the -E switch enables
# say, unicode, etc. which makes perl5 a bit more like perl6 
# for oneliners anyway. Is using the -p switch cheating?

#------------------------------------------------------------
# 1. Print the third column of the line in a file.

# a. 
perl6 -n -e '.split(" ")[2].say' file.txt

# How to get this to print only non Nil values; or to print "" 
# when the 3rd column is empty; or print "No third column" 
# instead of Nil ...

# This prints column three if it exists:
perl6 -n -e '.say if .split("")[2]' file.txt

# This tells us what happened:
perl6 -n -e 'if .split("")[2] {.say} else {say "No third column"}

# a ternary construct could be used instead:
perl6 -n -e '.split("")[2] ?? say $_ !! say "" '

#DRYer:
perl6 -n -e 'say .split("")[2] ?? $_ !! "" ' file.txt

# ...

#------------------------------------------------------------
# 2. Sum up the lines in a file:
# a. They teach you this in beginner perl books!
perl5 -E 'say scalar(()=<>)' file.txt  

# b. This is quite different can I make it more similar?
perl6 -e 'say $*ARGFILES.lines.elems' file.txt

# c. It seems if file.txt presents as input (i.e. "<") and not 
# just an ARG, then $*IN works too.
perl6 -e 'say $*IN.lines.elems' < file.txt 

# d. OTOH if you call lines() as a function instead of as a 
# method on a file object you get this nice correct "idiom":

perl6 -e 'say lines().elems' file.txt

# Since running scalar sub on an empty list scalar(()=<>)  
# is a bit of a "trick" in perl5 syntax, it's more fair to 
# compare something like:

perl5 -nE '}{say $.' file.txt

# Is the butterly operator ("}{") a perl5-ism?


#------------------------------------------------------------
# 3. Print an empty line after each line
# a. First the obvious way in perl5 then a perl6ish way in perl5 ...
perl5 -ne 'print "\n" '  file.txt
perl5 -nE 'say "$_"' file.txt

# b. Since the -p option prints lines and "say" adds a newline
@ this works. Why can't -p and -e be combined with perl6? 
perl6 -p -e 'say ' file.txt  


#------------------------------------------------------------
# 4. Number the lines
# a. They teach you this in beginner perl books too!!
perl5 -pe '$_ = "$. $_"' file.txt 

# b. Wanted to ++ the variable and use "state" to "localize" it in one pass.
# without using "my $num; ".  But I took too long and then saw grondilu's 
# solution. I've mostly copied it :-\
perl6 -e 'say (state $num)++, " $_" for $*IN.lines' < file.txt

# The one I like best was from PerlJam on IRC:

perl6 -e 'say (my $)++ ~ " $_" for lines()' < file.txt

# This one is very close to perl5
perl6 -p -e '$_ = "{(state $)++} $_"' file.txt

# The variable can be nameless and the match matches everything
# coming from lines().

#------------------------------------------------------------
# 5. Print everything between two lines:
# a. 
perl5 -ne 'print if /START/../END/'  file.txt

# b. 
perl5 -nE 'say if /START/../END/'  file.txt 

# c. 
perl6 -e 'say $_ if /^ START/ ff /^END/ for lines' file.txt

# d. 
perl6 -e '.say if /START/ ff /END/ for lines' file.txt

# e.
perl6 -e '.say if /START/ ff /END/ for lines' file.txt

# I like that b. and  d. are similar since I may remember how and 
# why they differ.  In perl6, "say" wants to refer to something
# explicitly  (even if it is $_) or to be called as a method 
# on the results (as in e. where ".say" is invoked on 
# the results of the 'if /START/ ff /END/ for lines' statement.