added get_sample_names() method that's replicate compliant (#143)

sequence_processing_pipeline/Pipeline.py

@@ -637,6 +637,35 @@ def _get_sample_names_from_sample_sheet(self, project_name):
 
             return results
 
+    def get_orig_names_from_sheet(self, project_name):
+        if project_name is None:
+            results = [x.orig_name for x in self.sample_sheet.samples]
+            # eliminate inavoidable duplicates and sort.
+            return sorted(set(results))
+        else:
+            # Since the project-name is stored in an internal variable
+            # in a third-party library, convert the data structure to
+            # JSON using the exposed method and obtain from the result.
+            jsn = json_loads(self.sample_sheet.to_json())
+
+            results = []
+
+            for sample in jsn['Data']:
+                # handle case where project_name includes an appended qiita-id.
+                if sample['Sample_Project'] == project_name:
+                    results.append(sample['orig_name'])
+                    continue
+
+                # handle case where project_name does not include a qiita-id.
+                # exact matching is required for cases where one project name
+                # in a sheet is a superset of another project in the same
+                # sheet.
+                m = search(r'^(.+)_(\d+)$', sample['Sample_Project'])
+                if m[1] == project_name:
+                    results.append(sample['orig_name'])
+
+            return sorted(set(results))
+
     def _get_sample_names_from_mapping_file(self, project_name):
         if project_name is None:
             return list(self.mapping_file.sample_name)

sequence_processing_pipeline/tests/test_NuQCJob.py

@@ -1282,7 +1282,6 @@ def test_regular_expressions(self):
         self._helper(job.json_regex, good_names, bad_names)
 
     def test_generate_mmi_filter_cmds(self):
-        # CHARLIE
         double_db_paths = ["db_path/mmi_1.db", "db_path/mmi_2.db"]
         job = NuQCJob(self.fastq_root_path, self.output_path,
                       self.good_sample_sheet_path, double_db_paths,

sequence_processing_pipeline/tests/test_Pipeline.py

@@ -36,6 +36,7 @@ def setUp(self):
         self.good_run_dir = self.path(self.good_run_id)
         self.runinfo_file = self.path(self.good_run_id, 'RunInfo.xml')
         self.rtacomplete_file = self.path(self.good_run_id, 'RTAComplete.txt')
+        self.good_sheet_w_replicates = self.path('good_sheet_w_replicates.csv')
 
         # most of the tests here were written with the assumption that these
         # files already exist.
@@ -152,6 +153,20 @@ def test_get_sample_names_from_sample_sheet(self):
             obs = set(pipeline._get_sample_names_from_sample_sheet(param))
             self.assertEqual(obs, exp)
 
+    def test_get_orig_names_from_sheet_with_replicates(self):
+        pipeline = Pipeline(self.good_config_file, self.good_run_id,
+                            self.good_sheet_w_replicates, None,
+                            self.output_file_path, self.qiita_id,
+                            Pipeline.METAGENOMIC_PTYPE)
+
+        obs = pipeline.get_orig_names_from_sheet('Feist_11661')
+        exp = {'BLANK.43.12G', 'BLANK.43.12H', 'JBI.KHP.HGL.021',
+               'JBI.KHP.HGL.022', 'JBI.KHP.HGL.023', 'JBI.KHP.HGL.024',
+               'RMA.KHP.rpoS.Mage.Q97D', 'RMA.KHP.rpoS.Mage.Q97E',
+               'RMA.KHP.rpoS.Mage.Q97L', 'RMA.KHP.rpoS.Mage.Q97N'}
+
+        self.assertEqual(set(obs), exp)
+
     def test_required_file_checks(self):
         # begin this test by deleting the RunInfo.txt file and verifying that
         # Pipeline object will raise an Error.