Allow MP names that differ only in honourifics.

This means if WriteToThem has Sir Malcolm Rifkind but TheyWorkForYou only has Malcolm Rifkind, we still consider that a match. This might well apply to all types this script handles, but will restrict just to MPs for now, seems to be most likely place it could happen.
mysociety · Nov 18, 2014 · 145d367 · 145d367
1 parent 0516f4a
commit 145d367
Show file tree

Hide file tree

Showing 2 changed files with 20 additions and 3 deletions.
diff --git a/services/mapit-dadem-loading/CouncilMatch.pm b/services/mapit-dadem-loading/CouncilMatch.pm
@@ -384,6 +384,9 @@ sub canonicalise_council_name ($) {
     return $_;
 }
 
+my $titles = "Cllr|Councillor|Dr|Hon|hon|rah|rh|Mrs|Ms|Mr|Miss|Rt Hon|Reverend|The Rev|The Reverend|Sir|Dame|Rev|Prof";
+my $honourifics = "MP|CBE|OBE|MBE|QC|BEM|rh|RH|Esq|QPM|JP|FSA|Bt|BEd|Hons|TD|MA|QHP|DL|CMG|BB|AKC|Bsc|Econ|LLB|GBE|QSO|BA|FRSA|FCA|DD|KBE|PhD";
+
 # canonicalise_person_name NAME
 # Convert name from various formats "Fred Smith", "Smith, Fred",
 # "Fred R Smith", "Smith, Fred RK" to uniform one "fred smith".  Removes
@@ -393,8 +396,6 @@ sub canonicalise_person_name ($) {
     ($_) = @_;
 
     # Remove fancy words
-    my $titles = "Cllr|Councillor|Dr|Hon|hon|rah|rh|Mrs|Ms|Mr|Miss|Rt Hon|Reverend|The Rev|The Reverend|Sir|Dame|Rev|Prof";
-    my $honourifics = "MP|CBE|OBE|MBE|QC|BEM|rh|RH|Esq|QPM|JP|FSA|Bt|BEd|Hons|TD|MA|QHP|DL|CMG|BB|AKC|Bsc|Econ|LLB|GBE|QSO|BA|FRSA|FCA|DD|KBE|PhD";
     while (s#(\b(?:$titles)\b)##) {};
     while (s#(\b(?:$honourifics)\b)##) {};
 
@@ -420,6 +421,14 @@ sub canonicalise_person_name ($) {
     return $_;
 }
 
+# remove_honourifics NAME
+sub remove_honourifics ($) {
+    ($_) = @_;
+    while (s#(\b(?:$titles)\b)##) {};
+    while (s#(\b(?:$honourifics)\b)##) {};
+    return $_;
+}
+
 # Internal use
 # get_extradata COUNCIL_ID 
 # Checks we have the councillor names webpage URL, and any other needed data.

diff --git a/services/mapit-dadem-loading/dadem_csv_load b/services/mapit-dadem-loading/dadem_csv_load
@@ -262,10 +262,18 @@ while (my $line = <MAIN>) {
             $updates_required = 1;
             $conflict .= "Party changed ";
         }
+
         # ... changing name is also (even though it must have matched canonical name)
-        if ($cur_repinfo->{name} ne $finalname) {
+        if ($kind ne 'mp' && $cur_repinfo->{name} ne $finalname) {
             $updates_required = 1;
             $conflict .= "Name changed ";
+        } elsif ($kind eq 'mp') {
+            my $curname_nohonourifics = CouncilMatch::remove_honourifics($cur_repinfo->{name});
+            my $finalname_nohonourifics = CouncilMatch::remove_honourifics($finalname);
+            if ($curname_nohonourifics ne $finalname_nohonourifics) {
+                $updates_required = 1;
+                $conflict .= "Name changed ";
+            }
         }
 
         # Use old fax/email if we don't have a new one