| 1 |
98 |
ahitrov |
package Contenido::Parser::Util; |
| 2 |
|
|
|
| 3 |
|
|
use strict; |
| 4 |
|
|
|
| 5 |
|
|
sub clean_invalid_chars { # http://www.w3.org/TR/REC-xml/#NT-Char |
| 6 |
|
|
my ($cont_ref) = shift; |
| 7 |
|
|
$$cont_ref =~ s/[\x0-\x8|\xB\xC|\xE-\x1F|\x{d800}-\x{dfff}|\x{fffe}\x{ffff}]//sgi; |
| 8 |
|
|
} |
| 9 |
|
|
|
| 10 |
|
|
sub text_cleanup { |
| 11 |
|
|
my $text = shift; |
| 12 |
|
|
my $delim = shift || "\n\n"; |
| 13 |
|
|
|
| 14 |
|
|
$text =~ s/^\s+//; $text =~ s/\s+$//; |
| 15 |
|
|
$text =~ s/\r\n/\n/g; |
| 16 |
|
|
|
| 17 |
|
|
my @paragfs = $text =~ /\n{2,}/ ? # is paragraphs detected? |
| 18 |
|
|
split /\n{2,}/, $text : # - by paragraphs only |
| 19 |
|
|
split /\n+/, $text; # - by any newline |
| 20 |
|
|
|
| 21 |
|
|
for (@paragfs) { |
| 22 |
|
|
s/^\s+//mg; s/\s+$//mg; # trim whitespace |
| 23 |
|
|
s/[[:blank:]]+/ /g; # collapse spaces |
| 24 |
|
|
} |
| 25 |
|
|
|
| 26 |
|
|
return join "\n\n", grep length $_, @paragfs; |
| 27 |
|
|
} |
| 28 |
|
|
|
| 29 |
|
|
1; |