GitHub/pbot
3
0
Fork 0
mirror of https://github.com/pragma-/pbot.git synced 2024-11-23 04:19:27 +01:00
Code Issues Projects Releases Wiki Activity

Spinach: Improve normalization, and laxen validation

Browse Source
This commit is contained in:
Pragmatic Software 2018-02-11 20:44:14 -08:00
parent 16e4a0d1e4
commit e2f9810d91
1 changed files with 18 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

28
PBot/Plugins/Spinach.pm
View File

@ -1048,29 +1048,35 @@ sub normalize_text {
if ($word =~ m/^\d{4}$/ and $word >= 1700 and $word <= 2100) { if ($word =~ m/^\d{4}$/ and $word >= 1700 and $word <= 2100) {
$newword = year2en($word); $newword = year2en($word);
} elsif ($word =~ m/^\d+$/) { } elsif ($word =~ m/^-?\d+$/) {
$newword = num2en($word); $newword = num2en($word);
if (defined $punct and $punct eq '%') { if (defined $punct and $punct eq '%') {
$newword .= " percent"; $newword .= " percent";
$punct = undef; $punct = undef;
} }
} elsif ($word =~ m/^(\d+)(?:st|nd|rd|th)$/i) { } elsif ($word =~ m/^(-?\d+)(?:st|nd|rd|th)$/i) {
$newword = num2en_ordinal($1); $newword = num2en_ordinal($1);
} elsif ($word =~ m/^\$(\d+)(\.\d+)?$/i) { } elsif ($word =~ m/^(-)?\$(\d+)?(\.\d+)?$/i) {
my ($dollars, $cents) = ($1, $2); my ($neg, $dollars, $cents) = ($1, $2, $3);
$word = num2en($dollars); $newword = '';
$newword = "$word " . ($dollars == 1 ? "dollar" : "dollars"); $dollars = "$neg$dollars" if defined $neg and defined $dollars;
if (defined $dollars) {
$word = num2en($dollars);
$newword = "$word " . (abs $dollars == 1 ? "dollar" : "dollars");
}
if (defined $cents) { if (defined $cents) {
$cents =~ s/^\.0*//; $cents =~ s/^\.0*//;
$cents = "$neg$cents" if defined $neg and not defined $dollars;
$word = num2en($cents); $word = num2en($cents);
$newword .= " and $word cent" if $cents == 1; $newword .= " and " if defined $dollars;
$newword .= " and $word cents" if $cents > 1; $newword .= (abs $cents == 1 ? "$word cent" : "$word cents");
} }
} elsif ($word =~ m/^(\d+\.\d+)(?:st|nd|rd|th)?$/i) { } elsif ($word =~ m/^(-?\d*\.\d+)(?:st|nd|rd|th)?$/i) {
$newword = num2en($1); $newword = num2en($1);
} elsif ($word =~ m{^(\d+\s*/\s*\d+)(?:st|nd|rd|th)?$}i) { } elsif ($word =~ m{^(-?\d+\s*/\s*-?\d+)(?:st|nd|rd|th)?$}i) {
$newword = fraction2words($1); $newword = fraction2words($1);
} }
@ -1102,6 +1108,7 @@ sub validate_lie {
return 0; return 0;
} }
=cut
$count = 0; $count = 0;
foreach my $word (keys %lie_words) { foreach my $word (keys %lie_words) {
if (exists $truth_words{$word}) { if (exists $truth_words{$word}) {
@ -1112,6 +1119,7 @@ sub validate_lie {
if ($count == $lie_word_count) { if ($count == $lie_word_count) {
return 0; return 0;
} }
=cut
my $stripped_truth = $truth; my $stripped_truth = $truth;
$stripped_truth =~ s/(?:\s|\p{PosixPunct})+//g; $stripped_truth =~ s/(?:\s|\p{PosixPunct})+//g;