Rosetta Code/Find bare lang tags

Find all <lang> tags without a language specified in the text of a page. Display counts by language section:

Description

<lang>Pseudocode</lang>

=={{header|C}}==
<lang C>printf("Hello world!\n");</lang>

=={{header|Perl}}==
<lang>print "Hello world!\n"</lang>

should display something like

2 bare language tags.

1 in perl
1 in no language

For extra credit, allow multiple files to be read. Summarize all results by language:

5 bare language tags.

2 in c ([[Foo]], [[Bar]])
1 in perl ([[Foo]])
2 in no language ([[Baz]])

For more extra credit, use the Media Wiki API to test actual RC tasks.

AutoHotkey

This code has no syntax highlighting, because Rosetta Code's highlighter fails with code that contains literal </lang> tags.

Stole RegEx Needle from Perl

task =
(
Description

<lang>Pseudocode</lang>

=={{header|C}}==
<lang C>printf("Hello world!\n");</lang>

=={{header|Perl}}==
<lang>print "Hello world!\n"</lang>
)
lang := "no lanugage", out := Object(lang, 0), total := 0
Loop Parse, task, `r`n
	If RegExMatch(A_LoopField, "==\s*{{\s*header\s*\|\s*([^\s\}]+)\s*}}\s*==", $)
		lang := $1, out[lang] := 0
	else if InStr(A_LoopField, "<lang>")
		out[lang]++
For lang, num in Out
	If num
		total++, str .= "`n" num " in " lang
MsgBox % clipboard := total " bare lang tags.`n" . str

Output:

2 bare lang tags.

1 in no lanugage
1 in Perl

Perl

This is a simple implementation that does not attempt either extra credit. <lang perl>my $lang = 'no language'; my $total = 0; my %blanks = (); while (<>) {

 if (m/<lang>/) {
   if (exists $blanks{lc $lang}) {
     $blanks{lc $lang}++
   } else {
     $blanks{lc $lang} = 1
   }
   $total++
 } elsif (m/==\s*Template:\s*header\s*\\s*==/) {
   $lang = lc $1
 }

}

if ($total) { print "$total bare language tag" . ($total > 1 ? 's' : ) . ".\n\n"; while ( my ($k, $v) = each(%blanks) ) { print "$k in $v\n" } }</lang>