Attempt to fix travis issues with #46

rankratioviz/generate.py

@@ -83,15 +83,15 @@ def gen_rank_plot(V):
 
     Returns:
 
-    altair.Chart object for the rank plot.
+    JSON describing altair.Chart for the rank plot.
     """
 
     # Get stuff ready for the rank plot
     # First off, convert all rank column IDs to strings (since Altair gets
     # angry if you pass in ints as column IDs). This is a problem with
     # OrdinationResults files, since just getting the raw column IDs gives int
     # values (0 for the first column, 1 for the second column, etc.)
-    V.columns = [str(c) for c in V.columns]
+    V.columns = ["Rank " + str(c) for c in V.columns]
 
     # The default rank column is just whatever the first rank is. This is what
     # the rank plot will use when it's first drawn.
@@ -150,7 +150,11 @@ def gen_rank_plot(V):
         # lines
         gridOpacity=0.35
     ).interactive()
-    return rank_chart
+
+    rank_chart_json = rank_chart.to_dict()
+    rank_ordering = "rankratioviz_rank_ordering"
+    rank_chart_json["datasets"][rank_ordering] = list(V.columns)
+    return rank_chart_json
 
 
 def gen_sample_plot(table, metadata):
@@ -163,7 +167,7 @@ def gen_sample_plot(table, metadata):
 
     Returns:
 
-    altair.Chart object for the sample scatterplot.
+    JSON describing altair.Chart for the sample plot.
     """
 
     # Used to set x-axis and color
@@ -272,7 +276,7 @@ def gen_visualization(V, processed_table, df_sample_metadata, output_dir):
        index_path: a path to the index.html file for the output visualization.
                    This is needed when calling q2templates.render().
     """
-    rank_plot_chart = gen_rank_plot(V)
+    rank_plot_json = gen_rank_plot(V)
     sample_plot_json = gen_sample_plot(processed_table, df_sample_metadata)
     os.makedirs(output_dir, exist_ok=True)
     # copy files for the visualization
@@ -307,8 +311,9 @@ def gen_visualization(V, processed_table, df_sample_metadata, output_dir):
     # write new files
     rank_plot_loc = os.path.join(output_dir, 'rank_plot.json')
     sample_plot_loc = os.path.join(output_dir, 'sample_plot.json')
-    rank_plot_chart.save(rank_plot_loc)
     # For reference: https://stackoverflow.com/a/12309296
-    with open(sample_plot_loc, "w") as jfile:
-        json.dump(sample_plot_json, jfile)
+    with open(rank_plot_loc, "w") as jf:
+        json.dump(rank_plot_json, jf)
+    with open(sample_plot_loc, "w") as jf2:
+        json.dump(sample_plot_json, jf2)
     return index_path

rankratioviz/tests/testing_utilities.py

@@ -101,12 +101,15 @@ def validate_rank_plot_json(input_ranks_loc, rank_json_loc):
         assert rank_plot["title"] == "Ranks"
         basic_vegalite_json_validation(rank_plot)
         dn = rank_plot["data"]["name"]
-        # Check that we have the same count of ranks as in the input ranks file
+        # Check that we have the same count of ranked features as in the
+        # input ranks file: this assumes that every ranked feature has an entry
+        # in the BIOM table, which is a reasonable assumption
         assert len(rank_plot["datasets"][dn]) == len(ranked_features)
         # Loop over every rank included in this JSON file:
-        prev_coefs_val = float("-inf")
+        rank_ordering = rank_plot["datasets"]["rankratioviz_rank_ordering"]
+        prev_rank_0_val = float("-inf")
         prev_x_val = -1
-        for rank in rank_plot["datasets"][dn]:
+        for feature in rank_plot["datasets"][dn]:
             # Check that we're using the correct "coefs" value
             # We use pytest's approx class to get past floating point
             # imprecisions. Note that we just leave this at the default for
@@ -116,28 +119,26 @@ def validate_rank_plot_json(input_ranks_loc, rank_json_loc):
                 # NOTE Based on how we construct feature labels from DEICODE
                 # input. If that changes, this will need to change or this
                 # will break.
-                feature_id = rank["Feature ID"].split("|")[2]
+                feature_id = feature["Feature ID"].split("|")[2]
             else:
-                feature_id = rank["Feature ID"]
-            # NOTE This assumes that the 0-th rank (i.e. the first) is the one
-            # stored in the rank plot. When we eventually update this so that
-            # an arbitrary number of ranks for each feature are stored in the
-            # rank plot, we'll need to ensure that there's a one-to-one
-            # correspondence between the various "coefs" and
-            # ranked_features[feature_id].
-            assert ranked_features[feature_id][0] == approx(rank["coefs"])
-            # Check that the ranks are in order (i.e. their "coefs" vals are
-            # monotonically increasing)
+                feature_id = feature["Feature ID"]
+
+            # Check that each ranked feature matches
+            for r in range(len(rank_ordering)):
+                actual_rank_val = ranked_features[feature_id][r]
+                assert actual_rank_val == approx(feature[rank_ordering[r]])
+            # Check that the ranks are in order (i.e. their initial rank vals
+            # are monotonically increasing)
             # (If this rank is approximately equal to the previous rank, then
             # don't bother with the comparison -- but still update
-            # prev_coefs_val, of course.)
-            if rank["coefs"] != approx(prev_coefs_val):
-                assert rank["coefs"] >= prev_coefs_val
+            # prev_rank_0_val, of course.)
+            if feature[rank_ordering[0]] != approx(prev_rank_0_val):
+                assert feature[rank_ordering[0]] >= prev_rank_0_val
             # Check that x values are also in order
-            assert rank["x"] == prev_x_val + 1
+            assert feature["x"] == prev_x_val + 1
             # Update prev_ things for the next iteration of the loop
-            prev_coefs_val = rank["coefs"]
-            prev_x_val = rank["x"]
+            prev_rank_0_val = feature[rank_ordering[0]]
+            prev_x_val = feature["x"]
 
 
 def validate_sample_plot_json(biom_table_loc, metadata_loc, sample_json_loc):