Update citation handling to reflect new bibliography formats

Luckily, most of the code to handle this was already written, however we
were targeting the <=0.11 version of Jupyter Book's way of handling
cross references so now we're up-to-date. Basically we check if it's a
bracketed reference (which appears to be always and only the case for
bibliographical citations as far as I can tell) and handle it like we
used to from there.

This commit updated the test for this and the code to pass it.

Update: made the if-check for it being a citation better.

Author: delfanbaum

jupyter_book_to_htmlbook/reference_processing.py

@@ -10,17 +10,21 @@ def process_interal_refs(chapter):
     references to valid htmlbook xrefs. Currently opinionated towards CMS
     author-date.
     """
-    xrefs = chapter.find_all(class_='internal')
+    xrefs = chapter.find_all("a", class_='internal')
     for ref in xrefs:
-        # handle bib references, be opinionated!
-        if ref['href'].find('references.html') > -1:
+        # handle bib references
+        if (
+                ref.parent.name == "span" and
+                ref.parent.contents[0] == "[" and
+                ref.parent.contents[-1] == "]"
+           ):
             ref.name = 'span'
             del ref['href']
             # remove any internal tags
             inner_str = ''
             for part in ref.contents:
                 inner_str += part.string
-            # remove last comma per CMS
+            # remove last comma (before year/date) per CMS
             inner_str = ','.join(inner_str.split(',')[0:-1]) + \
                         inner_str.split(',')[-1]
             ref.string = f'({inner_str})'

tests/test_bibliography.py

@@ -6,21 +6,15 @@
     )
 from jupyter_book_to_htmlbook.reference_processing import process_citations
 
-"""
-[[bibliography]]
-[appendix]
-[role="bibliography"]
-== Bibliography
-
-Bibliography content starts here.
-"""
-
 
 class TestBibliography:
     """
     Tests targeting "bibliography" content included in the book, including:
     * End of book "bibliography" appendix
     * End-of-file / chapter bibliographies
+
+    NOTE: Tests against citations/references should be found in the
+    test_references.py file.
     """
 
     def test_bib_appx_datatype(self, tmp_path):

tests/test_reference_processing.py

@@ -20,15 +20,18 @@ def test_process_internal_refs_bibliograpy(self):
         Should convert bib refs to spans containing CMS author-date style
         citations
         """
-        text = '<p>For a more detailed breakdown, see ' + \
-               '<span id="1">[<a class="reference internal" href="../../' + \
-               'references.html#id35">Leek and Peng, 2015</a>]</span>.</p>' + \
-               '<p>In the next section, we’ll talk about...</p>'
+        text = """
+<p>Finally, here is a citation <span id="id1">[<a class="reference internal"
+href="#id6" title="Terry Baruch...">Baruch, 1993</a>]</span>.
+And then some others <span id="id2">[<a class="reference internal"
+href="#id5" title="Terry Aadams...">Aadams, 1993</a>]</span>,
+<span id="id3">[<a class="reference internal" href="#id7"
+title="Terry Carver...">Carver, 1993</a>]</span>.</p>
+"""
         chapter = BeautifulSoup(text, 'html.parser')
         result = process_interal_refs(chapter)
-        assert str(result) == ("<p>For a more detailed breakdown, see <span>" +
-                               "(Leek and Peng 2015)</span>.</p><p>In the " +
-                               "next section, we’ll talk about...</p>")
+        assert not result.find("a")
+        assert "(Baruch 1993)" in result.find("span").contents
 
     def test_alert_on_external_images(self, caplog):
         """