Splitting on non-initial uppercase without split

loris has asked for the wisdom of the Perl Monks concerning the following question:

Replies are listed 'Best First'.
Re: Splitting on non-initial uppercase without split by edan (Curate) on Jan 26, 2005 at 10:46 UTC
This works, but I have the distinct feeling there is a more straightforward way to do it... `$s =~ s/(?<!^)(?=[A-Z])/ /g;` [download] -- edan	[reply] [d/l]
Re^2: Splitting on non-initial uppercamse without split by ambrus (Abbot) on Jan 26, 2005 at 20:09 UTC
What about this: `$s =~ s/\B(?=[[:upper:]])/ /g;` [download] or this `$s =~ s/\B([[:upper:]])/ $1/g;` [download]	[reply] [d/l] [select]
Re^2: Splitting on non-initial uppercase without split by tphyahoo (Vicar) on Jan 26, 2005 at 12:03 UTC
How about: `$s =~ s/([a-z])([A-Z])/$1 $2/ig;` [download] (untested)	[reply] [d/l]
Re^3: Splitting on non-initial uppercase without split by si_lence (Deacon) on Jan 26, 2005 at 12:38 UTC
I guess you would want to make your regex case sensitive: `$s =~ s/([a-z])([A-Z])/$1 $2/g;` [download] But then I think it still would not fit the description as it woud not split something like "Test1Test2" into "Test1 Test2" So following your line of thought it should be something like `$s =~ s/([^A-Z])([A-Z])/$1 $2/g;` [download] si_lence	[reply] [d/l] [select]
Re: Splitting on non-initial uppercase without split by si_lence (Deacon) on Jan 26, 2005 at 10:22 UTC
Hi If I get you right you want to have a blank before each capital letter except the first one in the string. Right? If yes use `$s =~ s/.(?=[A-Z])/ /g;` [download] si_lence update: The comment of edan is right (sorry to mess it up) but this should work `$s =~ s/(.)(?=[A-Z])/$1 /g;` [download] si_lence	[reply] [d/l] [select]
Re^2: Splitting on non-initial uppercase without split by edan (Curate) on Jan 26, 2005 at 10:43 UTC
Erm, no. That's going to turn "LikeThis" into "Lik This" -- edan	[reply]
Re: Splitting on non-initial uppercase without split by dimar (Curate) on Jan 26, 2005 at 20:15 UTC
Since no one else has mentioned it yet, here is an example that uses the SlashG anchor. (SlashG is typed like that just to make it search-engine-distinguishable). `$_ = "ThisIsACamelCaseWorld"; my @aTemp; @aTemp = m/\G[A-Z]?[a-z]*/g; ## same thing print join " ", @aTemp; __END__ This Is A Camel Case World` [download]	[reply] [d/l]
Re: Splitting on non-initial uppercase without split by Anonymous Monk on Jan 26, 2005 at 12:31 UTC
`substr ($s, 1) =~ s/([A-Z])/ $1/g;` [download]	[reply] [d/l]
Re^2: Splitting on non-initial uppercase without split by Aristotle (Chancellor) on Jan 26, 2005 at 14:29 UTC
Yep. Or per the OP's substitution, which I prefer: `substr( $s, 1 ) =~ s/(?=[A-Z])/ /g;` [download] Makeshifts last the longest.	[reply] [d/l]
Re: Splitting on non-initial uppercase without split by ambrus (Abbot) on Jan 26, 2005 at 14:52 UTC
I don't quite undrestand your question, but `$w = "ThisIsTheHouseThatJackBuilt"; $w = join " ", $w =~ /[[:upper:]][ +^[:upper:]]*/g; print $w, $/;` [download] gives `This Is The House That Jack Built` [download]	[reply] [d/l] [select]