Merge pull request #3190 from antgonza/fix-3158

fix #3158

Author: charles-cowart

qiita_db/metadata_template/prep_template.py

@@ -858,12 +858,20 @@ def _get_predecessors(workflow, node):
                     pdp = pnode.default_parameter
                     pdp_cmd = pdp.command
                     params = pdp.values.copy()
-                    reqp = {x: y[1][0]
-                            for x, y in pdp_cmd.required_parameters.items()}
+                    # verifying that the workflow.artifact_type is included
+                    # in the command input types or raise an error
+                    wkartifact_type = wk.artifact_type
+                    reqp = dict()
+                    for x, y in pdp_cmd.required_parameters.items():
+                        if wkartifact_type not in y[1]:
+                            raise ValueError(f'{wkartifact_type} is not part '
+                                             'of this preparation and cannot '
+                                             'be applied')
+                        reqp[x] = wkartifact_type
+
                     cmds_to_create.append([pdp_cmd, params, reqp])
 
-                    init_artifacts = {
-                        self.artifact.artifact_type: self.artifact.id}
+                    init_artifacts = {wkartifact_type: self.artifact.id}
 
                 cmds_to_create.reverse()
                 current_job = None
@@ -874,8 +882,9 @@ def _get_predecessors(workflow, node):
                         for iname, dname in rp.items():
                             if dname not in init_artifacts:
                                 msg = (f'Missing Artifact type: "{dname}" in '
-                                       'this preparation; are you missing a '
-                                       'step to start?')
+                                       'this preparation; this might be due '
+                                       'to missing steps or not having the '
+                                       'correct raw data.')
                                 # raises option c.
                                 raise ValueError(msg)
                             req_params[iname] = init_artifacts[dname]

qiita_db/metadata_template/test/test_prep_template.py

@@ -1412,13 +1412,13 @@ def test_artifact_setter(self):
             pt.add_default_workflow(qdb.user.User('test@foo.bar'))
 
         # now let's test that an error is raised when there is no valid initial
-        # input data; this moves the data type from FASTQ to taxa_summary
+        # input data; this moves the data type from FASTQ to taxa_summary for
+        # the default_workflow_id = 1
         qdb.sql_connection.perform_as_transaction(
-            'UPDATE qiita.artifact SET artifact_type_id = 10 WHERE '
-            f'artifact_id = {pt.artifact.id}')
-        with self.assertRaisesRegex(ValueError, 'Missing Artifact type: '
-                                    '"FASTQ" in this preparation; are you '
-                                    'missing a step to start?'):
+            'UPDATE qiita.default_workflow SET artifact_type_id = 10 WHERE '
+            'default_workflow_id = 1')
+        with self.assertRaisesRegex(ValueError, 'taxa_summary is not part of '
+                                    'this preparation and cannot be applied'):
             pt.add_default_workflow(qdb.user.User('test@foo.bar'))
 
         # cleaning

qiita_db/software.py

@@ -1946,6 +1946,23 @@ def data_type(self):
             qdb.sql_connection.TRN.add(sql, [self.id])
             return qdb.sql_connection.TRN.execute_fetchflatten()
 
+    @property
+    def artifact_type(self):
+        """Retrieves artifact_type that the workflow can be applied to
+
+        Returns
+        ----------
+        str
+            The name of the artifact type this workflow can be applied to
+        """
+        with qdb.sql_connection.TRN:
+            sql = """SELECT artifact_type
+                     FROM qiita.artifact_type
+                     LEFT JOIN qiita.default_workflow USING (artifact_type_id)
+                     WHERE default_workflow_id = %s"""
+            qdb.sql_connection.TRN.add(sql, [self.id])
+            return qdb.sql_connection.TRN.execute_fetchflatten()[0]
+
     @property
     def graph(self):
         """Returns the graph that represents the workflow

qiita_db/support_files/patches/85.sql

@@ -10,3 +10,14 @@ BEGIN
     INSERT INTO qiita.command_parameter (command_id, parameter_name, parameter_type, required, default_value)
         VALUES (cmd_id, 'categories', 'mchoice', True, NULL);
 END $do$;
+
+-- Feb 28, 2022
+-- adding a new column to the default_workflow table to keep track of the
+-- artifact type that is expecting vs. "guessing"
+
+ALTER TABLE qiita.default_workflow ADD artifact_type_id BIGINT NOT NULL DEFAULT 3;
+ALTER TABLE qiita.default_workflow
+  ADD CONSTRAINT fk_artifact_type_id
+    FOREIGN KEY (artifact_type_id)
+    REFERENCES qiita.artifact_type(artifact_type_id)
+    ON UPDATE CASCADE;

qiita_db/test/test_software.py

@@ -632,6 +632,9 @@ def test_default_workflows(self):
         obs = list(qdb.software.DefaultWorkflow.iter(False))
         self.assertEqual(obs, exp)
 
+        self.assertEqual(
+            qdb.software.DefaultWorkflow(1).artifact_type, 'FASTQ')
+
         qdb.software.DefaultWorkflow(1).active = False
         obs = list(qdb.software.DefaultWorkflow.iter(False))
         self.assertEqual(obs, exp)

qiita_pet/handlers/software.py

@@ -72,13 +72,23 @@ def _default_parameters_parsing(node):
         #                   output_type: output_node_name}, ...}
         # for easy look up and merge of output_names
         main_nodes = dict()
-        for x, y in graph.edges:
+        for i, (x, y) in enumerate(graph.edges):
             connections = []
             for a, _, c in graph[x][y]['connections'].connections:
                 connections.append("%s | %s" % (a, c))
 
             vals_x, input_x, output_x = _default_parameters_parsing(x)
             vals_y, input_y, output_y = _default_parameters_parsing(y)
+
+            if i == 0:
+                # we are in the first element so we can specifically select
+                # the type we are looking for
+                at = w.artifact_type
+                if at in input_x[0][1]:
+                    input_x[0][1] = at
+                else:
+                    input_x[0][1] = '** WARNING, NOT DEFINED **'
+
             name_x = vals_x[0]
             name_y = vals_y[0]
             if vals_x not in (nodes):

qiita_pet/test/test_software.py

@@ -144,6 +144,8 @@ def test_retrive_workflows(self):
                 'rev_comp_mapping_barcodes': 'True', 'rev_comp': 'False',
                 'phred_quality_threshold': '3', 'barcode_type': 'golay_12',
                 'max_barcode_errors': '1.5', 'phred_offset': 'auto'}],
+            ['input_params_7_FASTQ | per_sample_FASTQ', 1,
+             'FASTQ | per_sample_FASTQ'],
             ['output_params_7_demultiplexed | Demultiplexed', 1,
              'demultiplexed | Demultiplexed'],
             ['params_8', 3, 'Pick closed-reference OTUs', 'Defaults', {
@@ -152,7 +154,7 @@ def test_retrive_workflows(self):
                 'sortmerna_coverage': '0.97', 'threads': '1'}],
             ['output_params_8_OTU table | BIOM', 3, 'OTU table | BIOM']])
         exp[0]['edges'].extend([
-            ['input_params_1_FASTQ | per_sample_FASTQ', 'params_7'],
+            ['input_params_7_FASTQ | per_sample_FASTQ', 'params_7'],
             ['params_7', 'output_params_7_demultiplexed | Demultiplexed'],
             ['output_params_7_demultiplexed | Demultiplexed', 'params_8'],
             ['params_8', 'output_params_8_OTU table | BIOM']])
@@ -180,8 +182,8 @@ def test_retrive_workflows(self):
             'rev_comp_mapping_barcodes': 'False', 'rev_comp': 'False',
             'phred_quality_threshold': '3', 'barcode_type': 'golay_12',
             'max_barcode_errors': '1.5', 'phred_offset': 'auto'}],
-        ['input_params_1_FASTQ | per_sample_FASTQ', 1,
-         'FASTQ | per_sample_FASTQ'],
+        ['input_params_1_FASTQ', 1,
+         'FASTQ'],
         ['output_params_1_demultiplexed | Demultiplexed', 1,
          'demultiplexed | Demultiplexed'],
         ['params_2', 3, 'Pick closed-reference OTUs', 'Defaults', {
@@ -190,7 +192,7 @@ def test_retrive_workflows(self):
             'sortmerna_coverage': '0.97', 'threads': '1'}],
         ['output_params_2_OTU table | BIOM', 3, 'OTU table | BIOM']],
      'edges': [
-        ['input_params_1_FASTQ | per_sample_FASTQ', 'params_1'],
+        ['input_params_1_FASTQ', 'params_1'],
         ['params_1', 'output_params_1_demultiplexed | Demultiplexed'],
         ['output_params_1_demultiplexed | Demultiplexed', 'params_2'],
         ['params_2', 'output_params_2_OTU table | BIOM']]},
@@ -206,8 +208,8 @@ def test_retrive_workflows(self):
             'qual_score_window': '0', 'disable_primers': 'False',
             'reverse_primers': 'disable', 'reverse_primer_mismatches': '0',
             'truncate_ambi_bases': 'False'}],
-        ['input_params_3_FASTA | FASTA_Sanger | SFF', 2,
-         'FASTA | FASTA_Sanger | SFF'],
+        ['input_params_3_** WARNING, NOT DEFINED **', 2,
+         '** WARNING, NOT DEFINED **'],
         ['output_params_3_demultiplexed | Demultiplexed', 2,
          'demultiplexed | Demultiplexed'],
         ['params_4', 3, 'Pick closed-reference OTUs', 'Defaults', {
@@ -216,7 +218,7 @@ def test_retrive_workflows(self):
             'sortmerna_coverage': '0.97', 'threads': '1'}],
         ['output_params_4_OTU table | BIOM', 3, 'OTU table | BIOM']],
      'edges': [
-        ['input_params_3_FASTA | FASTA_Sanger | SFF', 'params_3'],
+        ['input_params_3_** WARNING, NOT DEFINED **', 'params_3'],
         ['params_3', 'output_params_3_demultiplexed | Demultiplexed'],
         ['output_params_3_demultiplexed | Demultiplexed', 'params_4'],
         ['params_4', 'output_params_4_OTU table | BIOM']]},
@@ -229,8 +231,8 @@ def test_retrive_workflows(self):
             'rev_comp_mapping_barcodes': 'False', 'rev_comp': 'False',
             'phred_quality_threshold': '3', 'barcode_type': 'not-barcoded',
             'max_barcode_errors': '1.5', 'phred_offset': 'auto'}],
-        ['input_params_5_FASTQ | per_sample_FASTQ', 1,
-         'FASTQ | per_sample_FASTQ'],
+        ['input_params_5_FASTQ', 1,
+         'FASTQ'],
         ['output_params_5_demultiplexed | Demultiplexed', 1,
          'demultiplexed | Demultiplexed'],
         ['params_6', 3, 'Pick closed-reference OTUs', 'Defaults', {
@@ -239,7 +241,7 @@ def test_retrive_workflows(self):
             'sortmerna_coverage': '0.97', 'threads': '1'}],
         ['output_params_6_OTU table | BIOM', 3, 'OTU table | BIOM']],
      'edges': [
-        ['input_params_5_FASTQ | per_sample_FASTQ', 'params_5'],
+        ['input_params_5_FASTQ', 'params_5'],
         ['params_5', 'output_params_5_demultiplexed | Demultiplexed'],
         ['output_params_5_demultiplexed | Demultiplexed', 'params_6'],
         ['params_6', 'output_params_6_OTU table | BIOM']]}]