DateRole: Handling 'of' in textual date formats (#3982)

* DateRole: Adding failing test for `$nth of $month $year` * DateRole: Now parses "7th of march 2017" * DateRole: Adding failing test for '5 of jan 1993' (no suffix and short month) * DateRole: Handles 5 of march * DateRole: Now also handle "march the 30th, 2017"
duckduckgo · Apr 11, 2017 · 264282b · 264282b
1 parent 011c78e
commit 264282b
Show file tree

Hide file tree

Showing 2 changed files with 6 additions and 2 deletions.
diff --git a/lib/DDG/GoodieRole/Dates.pm b/lib/DDG/GoodieRole/Dates.pm
@@ -347,12 +347,12 @@ sub build_datestring_regex {
     # month-first date formats
     push @regexes, qr#$date_number$date_delim$short_month$date_delim$full_year#i;
     push @regexes, qr#$date_number$date_delim$full_month$date_delim$full_year#i;
-    push @regexes, qr#(?:$short_month|$full_month) $date_number(?: ?$number_suffixes)?[,]? $full_year#i;
+    push @regexes, qr#(?:$short_month|$full_month) (?:the )?$date_number(?: ?$number_suffixes)?[,]? $full_year#i;
 
     # day-first date formats
     push @regexes, qr#$short_month$date_delim$date_number$date_delim$full_year#i;
     push @regexes, qr#$full_month$date_delim$date_number$date_delim$full_year#i;
-    push @regexes, qr#$date_number[,]?(?: ?$number_suffixes)? (?:$short_month|$full_month)[,]? $full_year#i;
+    push @regexes, qr#$date_number[,]?(?: ?$number_suffixes)? (?:of )?(?:$short_month|$full_month)[,]? $full_year#i;
 
     ## Ambiguous, but potentially valid date formats
     push @regexes, $ambiguous_dates;
@@ -391,6 +391,7 @@ sub parse_formatted_datestring_to_date {
     }
 
     $d =~ s/(\d+)\s?$number_suffixes/$1/i;                                       # Strip ordinal text.
+    $d =~ s/(\sof\s)|(\sthe\s)/ /i;                                                          # Strip "of" for "4th of march" and "the" for "march the 4th"
     $d =~ s/,//i;                                                                # Strip any random commas.
     $d =~ s/($full_month)/$full_month_to_short{lc $1}/i;                         # Parser deals better with the shorter month names.
     $d =~ s/^($short_month)$date_delim(\d{1,2})/$2-$short_month_fix{lc $1}/i;    # Switching Jun-01-2012 to 01 Jun 2012

diff --git a/t/00-roles.t b/t/00-roles.t
@@ -138,6 +138,9 @@ subtest 'Dates' => sub {
             '29 feb, 2012'      => 1330473600,
             '2038-01-20'        => 2147558400,     # 32-bit signed int UNIX epoch ends 2038-01-19
             '1780-01-20'        => -5994172800,    # Way before 32-bit signed int epoch
+            '5th of january 1993' => 726192000,
+            '5 of jan 1993'     => 726192000,
+            'june the 1st 2012' => 1338508800,
         );
 
         foreach my $test_date (sort keys %dates_to_match) {