fix-1591 (#2370)

* fix-1591

* removing warning ATTN @josenavas, fix tests

Author: antgonza

qiita_db/metadata_template/test/test_prep_template.py

@@ -887,9 +887,7 @@ def test_create(self):
     def test_create_already_prefixed_samples(self):
         """Creates a new PrepTemplate"""
         fp_count = qdb.util.get_count('qiita.filepath')
-        pt = npt.assert_warns(
-            qdb.exceptions.QiitaDBWarning,
-            qdb.metadata_template.prep_template.PrepTemplate.create,
+        pt = qdb.metadata_template.prep_template.PrepTemplate.create(
             self.metadata_prefixed, self.test_study, self.data_type)
         self._common_creation_checks(pt, fp_count)
 

qiita_db/metadata_template/test/test_sample_template.py

@@ -1084,9 +1084,7 @@ def test_create_str_prefixes(self):
 
     def test_create_already_prefixed_samples(self):
         """Creates a new SampleTemplate with the samples already prefixed"""
-        st = npt.assert_warns(
-            qdb.exceptions.QiitaDBWarning,
-            qdb.metadata_template.sample_template.SampleTemplate.create,
+        st = qdb.metadata_template.sample_template.SampleTemplate.create(
             self.metadata_prefixed, self.new_study)
         new_id = self.new_study.id
         # The returned object has the correct id

qiita_db/metadata_template/test/test_util.py

@@ -8,6 +8,7 @@
 
 from six import StringIO
 from unittest import TestCase, main
+import warnings
 
 import numpy.testing as npt
 import pandas as pd
@@ -36,12 +37,31 @@ def test_prefix_sample_names_with_id(self):
         }
         exp_df = pd.DataFrame.from_dict(exp_metadata_dict, orient='index',
                                         dtype=str)
-        qdb.metadata_template.util.prefix_sample_names_with_id(
-            self.metadata_map, 1)
+        with warnings.catch_warnings(record=True) as warn:
+            qdb.metadata_template.util.prefix_sample_names_with_id(
+                self.metadata_map, 1)
+            self.assertEqual(len(warn), 0)
         self.metadata_map.sort_index(inplace=True)
         exp_df.sort_index(inplace=True)
         assert_frame_equal(self.metadata_map, exp_df)
 
+        # test that it only prefixes the samples that are needed
+        metadata_dict = {
+            'Sample1': {'int_col': 1, 'float_col': 2.1, 'str_col': 'str1'},
+            '1.Sample2': {'int_col': 2, 'float_col': 3.1, 'str_col': '200'},
+            'Sample3': {'int_col': 3, 'float_col': 3, 'str_col': 'string30'},
+        }
+        metadata_map = pd.DataFrame.from_dict(
+            metadata_dict, orient='index', dtype=str)
+        with warnings.catch_warnings(record=True) as warn:
+            qdb.metadata_template.util.prefix_sample_names_with_id(
+                metadata_map, 1)
+            self.assertEqual(len(warn), 1)
+            self.assertEqual(str(warn[0].message), 'Some of the samples were '
+                             'already prefixed with the study id.')
+        metadata_map.sort_index(inplace=True)
+        assert_frame_equal(metadata_map, exp_df)
+
     def test_load_template_to_dataframe(self):
         obs = qdb.metadata_template.util.load_template_to_dataframe(
             StringIO(EXP_SAMPLE_TEMPLATE))

qiita_db/metadata_template/util.py

@@ -34,29 +34,25 @@ def prefix_sample_names_with_id(md_template, study_id):
     study_id : int
         The study to which the metadata belongs to
     """
-    # Get all the prefixes of the index, defined as any string before a '.'
-    prefixes = {idx.split('.', 1)[0] for idx in md_template.index}
-    # If the samples have been already prefixed with the study id, the prefixes
-    # set will contain only one element and it will be the str representation
-    # of the study id
-    if len(prefixes) == 1 and prefixes.pop() == str(study_id):
-        # The samples were already prefixed with the study id
-        warnings.warn("Sample names were already prefixed with the study id.",
-                      qdb.exceptions.QiitaDBWarning)
-    else:
-        # Create a new pandas series in which all the values are the study_id
-        # and it is indexed as the metadata template
-        study_ids = pd.Series([str(study_id)] * len(md_template.index),
-                              index=md_template.index)
-        # Create a new column on the metadata template that includes the
-        # metadata template indexes prefixed with the study id
-        md_template['sample_name_with_id'] = (study_ids + '.' +
-                                              md_template.index.values)
-        md_template.index = md_template.sample_name_with_id
-        del md_template['sample_name_with_id']
-        # The original metadata template had the index column unnamed - remove
-        # the name of the index for consistency
-        md_template.index.name = None
+    # loop over the samples and prefix those that aren't prefixed
+    md_template['qiita_sample_name_with_id'] = pd.Series(
+        [idx if idx.split('.', 1)[0] == str(study_id)
+         else '%d.%s' % (study_id, idx)
+         for idx in md_template.index], index=md_template.index)
+
+    # get the rows that are gonna change
+    changes = len(md_template.index[
+        md_template['qiita_sample_name_with_id'] != md_template.index])
+    if changes != 0 and changes != len(md_template.index):
+        warnings.warn(
+            "Some of the samples were already prefixed with the study id.",
+            qdb.exceptions.QiitaDBWarning)
+
+    md_template.index = md_template.qiita_sample_name_with_id
+    del md_template['qiita_sample_name_with_id']
+    # The original metadata template had the index column unnamed -> remove
+    # the name of the index for consistency
+    md_template.index.name = None
 
 
 def load_template_to_dataframe(fn, index='sample_name'):

qiita_pet/handlers/api_proxy/tests/test_prep_template.py

@@ -380,7 +380,6 @@ def test_prep_template_post_req(self):
                                      '16S')
         exp = {'status': 'warning',
                'message': [
-                    'Sample names were already prefixed with the study id.',
                     ('Some columns required to generate a QIIME-compliant '
                      'mapping file are not present in the template. A '
                      'placeholder value (XXQIITAXX) has been used to populate '