From d06f56fc0d9eb7a3e635b6990b4bff75fa8e1781 Mon Sep 17 00:00:00 2001
From: Ole Laursen <olau@iola.dk>
Date: Thu, 16 Feb 2017 12:32:03 +0000
Subject: [PATCH] Adjust unknown bin in the affiliation statistics too, adjust
 wording  - Legacy-Id: 12859

---
 ietf/stats/views.py                           | 29 ++++++++++---------
 .../document_stats_author_affiliation.html    | 10 ++++---
 .../document_stats_author_continent.html      |  2 +-
 .../stats/document_stats_author_country.html  |  2 +-
 4 files changed, 23 insertions(+), 20 deletions(-)

diff --git a/ietf/stats/views.py b/ietf/stats/views.py
index 3466e329a..b55362162 100644
--- a/ietf/stats/views.py
+++ b/ietf/stats/views.py
@@ -353,6 +353,17 @@ def document_stats(request, stats_type=None):
 
         total_persons = person_qs.distinct().count()
 
+        def prune_unknown_bin_with_known(bins):
+            # remove from the unknown bin all authors within the
+            # named/known bins
+            all_known = set(n for b, names in bins.iteritems() if b for n in names)
+            unknown = []
+            for name in bins[""]:
+                if name not in all_known:
+                    unknown.append(name)
+            bins[""] = unknown
+
+
         if stats_type == "author/documents":
             stats_title = "Number of {}s per author".format(doc_label)
 
@@ -389,6 +400,8 @@ def document_stats(request, stats_type=None):
             for name, affiliation in name_affiliation_set:
                 bins[aliases.get(affiliation, affiliation)].append(name)
 
+            prune_unknown_bin_with_known(bins)
+
             series_data = []
             for affiliation, names in sorted(bins.iteritems(), key=lambda t: t[0].lower()):
                 percentage = len(names) * 100.0 / total_persons
@@ -433,13 +446,7 @@ def document_stats(request, stats_type=None):
                 if c and c.in_eu:
                     bins[eu_name].append(name)
 
-            # remove from the unknown bin all authors with a known country
-            all_known = set(n for b, names in bins.iteritems() if b for n in names)
-            unknown = []
-            for name in bins[""]:
-                if name not in all_known:
-                    unknown.append(name)
-            bins[""] = unknown
+            prune_unknown_bin_with_known(bins)
 
             series_data = []
             for country, names in sorted(bins.iteritems(), key=lambda t: t[0].lower()):
@@ -478,13 +485,7 @@ def document_stats(request, stats_type=None):
                 continent_name = country_to_continent.get(country_name, "")
                 bins[continent_name].append(name)
 
-            # remove from the unknown bin all authors with a known continent
-            all_known = set(n for b, names in bins.iteritems() if b for n in names)
-            unknown = []
-            for name in bins[""]:
-                if name not in all_known:
-                    unknown.append(name)
-            bins[""] = unknown
+            prune_unknown_bin_with_known(bins)
 
             series_data = []
             for continent, names in sorted(bins.iteritems(), key=lambda t: t[0].lower()):
diff --git a/ietf/templates/stats/document_stats_author_affiliation.html b/ietf/templates/stats/document_stats_author_affiliation.html
index e5bcb23c7..6bec8d3c8 100644
--- a/ietf/templates/stats/document_stats_author_affiliation.html
+++ b/ietf/templates/stats/document_stats_author_affiliation.html
@@ -58,15 +58,17 @@
   </tbody>
 </table>
 
-<p>Some authors are authors of multiple documents with different
-  affiliation information associated, so the sum of multiple rows in the
-  table can be more than 100%.</p>
+<p>The statistics are based entirely on the author affiliation
+  provided with each draft. Since this may vary across documents, an
+  author may be counted with more than one affiliation, making the
+  total sum more than 100%.</p>
 
 
 
 <h3>Affiliation Aliases</h3>
 
-<p>In generating the above statistics, some heuristics have been applied to determine the affiliation of each author.</p>
+<p>In generating the above statistics, some heuristics have been
+  applied to determine the affiliations of each author.</p>
 
 {% if request.GET.showaliases %}
   <p><a href="{{ hide_aliases_url }}" class="btn btn-default">Hide generated aliases</a></p>
diff --git a/ietf/templates/stats/document_stats_author_continent.html b/ietf/templates/stats/document_stats_author_continent.html
index b17d49552..0e0a1f849 100644
--- a/ietf/templates/stats/document_stats_author_continent.html
+++ b/ietf/templates/stats/document_stats_author_continent.html
@@ -59,6 +59,6 @@
 </table>
 
 <p>The statistics are based entirely on the author addresses provided
-  in each draft. Since this varies across documents, a travelling
+  with each draft. Since this varies across documents, a travelling
   author may be counted in more than country, making the total sum
   more than 100%.</p>
diff --git a/ietf/templates/stats/document_stats_author_country.html b/ietf/templates/stats/document_stats_author_country.html
index 99c6d0152..93914abcb 100644
--- a/ietf/templates/stats/document_stats_author_country.html
+++ b/ietf/templates/stats/document_stats_author_country.html
@@ -59,7 +59,7 @@
 </table>
 
 <p>The statistics are based entirely on the author addresses provided
-  in each draft. Since this varies across documents, a travelling
+  with each draft. Since this varies across documents, a travelling
   author may be counted in more than country, making the total sum
   more than 100%.</p>