Sort documentation page in canonical order

backend/addcorpus/models.py

@@ -431,11 +431,12 @@ def clean(self):
                     e
                 ])
 
+
 class CorpusDocumentationPage(models.Model):
     class PageType(models.TextChoices):
         GENERAL = ('general', 'General information')
         CITATION = ('citation', 'Citation')
-        LICENSE = ('license', 'Licence')
+        LICENSE = ('license', 'License')
         TERMS_OF_SERVICE = ('terms_of_service', 'Terms of service')
         WORDMODELS = ('wordmodels', 'Word models')
 
@@ -455,6 +456,9 @@ class PageType(models.TextChoices):
         help_text='markdown contents of the documentation'
     )
 
+    def __str__(self):
+        return f'{self.corpus_configuration.corpus.name} - {self.type}'
+
     class Meta:
         constraints = [
             UniqueConstraint(

backend/addcorpus/tests/test_corpus_views.py

@@ -19,9 +19,14 @@ def test_no_corpora(db, settings, admin_client):
 def test_corpus_documentation_view(admin_client, basic_mock_corpus, settings):
     response = admin_client.get(f'/api/corpus/documentation/{basic_mock_corpus}/')
     assert response.status_code == 200
+    pages = response.data
+
+    # check that the pages are sorted in canonical order
+    page_types = [page['type'] for page in pages]
+    assert page_types == ['General information', 'Citation', 'License']
 
     # should contain citation guidelines
-    citation_page = next(page for page in response.data if page['type'] == 'Citation')
+    citation_page = next(page for page in pages if page['type'] == 'Citation')
 
     # check that the page template is rendered with context
     content = citation_page['content']

backend/addcorpus/views.py

@@ -42,22 +42,30 @@ def send_corpus_file(corpus='', subdir='', filename=''):
 
     return FileResponse(open(path, 'rb'))
 
+
 class CorpusDocumentationPageViewset(viewsets.ModelViewSet):
     permission_classes = [IsAuthenticatedOrReadOnly, CorpusAccessPermission]
     serializer_class = CorpusDocumentationPageSerializer
 
-    def get_queryset(self):
-        corpus_name = corpus_name_from_request(self.request)
-        pages = CorpusDocumentationPage.objects.filter(corpus_configuration__corpus__name=corpus_name)
-
+    @staticmethod
+    def get_relevant_pages(pages, corpus_name):
         # only include wordmodels documentation if models are present
         if Corpus.objects.get(name=corpus_name).has_python_definition:
             definition = load_corpus_definition(corpus_name)
             if definition.word_models_present:
                 return pages
-
         return pages.exclude(type=CorpusDocumentationPage.PageType.WORDMODELS)
 
+    def get_queryset(self):
+        corpus_name = corpus_name_from_request(self.request)
+        pages = CorpusDocumentationPage.objects.filter(
+            corpus_configuration__corpus__name=corpus_name)
+        relevant_pages = self.get_relevant_pages(pages, corpus_name)
+        canonical_order = [e.value for e in CorpusDocumentationPage.PageType]
+
+        return sorted(
+            relevant_pages, key=lambda p: canonical_order.index(p.type))
+
 
 class CorpusImageView(APIView):
     '''

backend/corpora_test/basic/license/license.md

@@ -0,0 +1 @@
+Do whatever you please.

backend/corpora_test/basic/mock_csv_corpus.py

@@ -25,6 +25,8 @@ class MockCSVCorpus(CSVCorpusDefinition):
     max_date = datetime.datetime(year=2022, month=12, day=31)
     data_directory = os.path.join(here, 'source_data')
     citation_page = 'citation.md'
+    license_page = 'license.md'
+    description_page = 'mock-csv-corpus.md'
 
     languages = ['en']
     category = 'book'