Merge branch 'develop'

Author: machinelit

.travis.yml

@@ -1,6 +1,8 @@
+sudo: false
 language: python
+cache: pip
 python:
-- '3.6'
+  - 3.6
 before_install:
 - bash -x get_test_data.sh
 install:

isatools/convert/isatab2w4m.py

@@ -14,6 +14,7 @@
 
 
 from isatools import isatab as ISATAB
+from isatools.utils import utf8_text_file_open
 
 # original from https://github.com/workflow4metabolomics/mtbls-dwnld/blob/develop/isatab2w4m.py
 __author__ = 'pkrog (Pierrick Roger)'
@@ -334,7 +335,7 @@ def get_investigation_file(input_dir):
 ################################################################
 
 def load_investigation(investigation_file):
-    f = open(investigation_file, 'r')
+    f = utf8_text_file_open(investigation_file)
     investigation = ISATAB.load(f)
     return investigation
 
@@ -364,8 +365,10 @@ def get_sample_names(assay_df, measures_df):
 
 def make_sample_metadata(study_df, assay_df, sample_names, normalize=True):
     # Normalize column names
-    study_df.set_axis(axis=1, labels=make_names(study_df.axes[1].tolist()))
-    assay_df.set_axis(axis=1, labels=make_names(assay_df.axes[1].tolist()))
+    study_df.set_axis(
+        inplace=True, axis=1, labels=make_names(study_df.axes[1].tolist()))
+    assay_df.set_axis(
+        inplace=True, axis=1, labels=make_names(assay_df.axes[1].tolist()))
 
     # Merge data frames
     sample_metadata = assay_df.merge(study_df, on='Sample.Name', sort=False)
@@ -374,7 +377,7 @@ def make_sample_metadata(study_df, assay_df, sample_names, normalize=True):
     if normalize:
         norm_sample_names = make_names(sample_names, uniq=True)
         sample_metadata.insert(0, 'sample.name', norm_sample_names)
-        sample_metadata.set_axis(axis=1, labels=make_names(
+        sample_metadata.set_axis(inplace=True, axis=1, labels=make_names(
             sample_metadata.axes[1].tolist(), uniq=True))
 
     return sample_metadata
@@ -395,7 +398,7 @@ def make_variable_metadata(measures_df, sample_names, variable_names,
 
     # Normalize
     if normalize:
-        variable_metadata.set_axis(axis=1, labels=make_names(
+        variable_metadata.set_axis(inplace=True, axis=1, labels=make_names(
             variable_metadata.axes[1].tolist(), uniq=True))
 
     return variable_metadata
@@ -422,7 +425,8 @@ def make_matrix(measures_df, sample_names, variable_names, normalize=True):
     if normalize:
         norm_sample_names = make_names(sample_names, uniq=True)
         norm_sample_names.insert(0, 'variable.name')
-        sample_variable_matrix.set_axis(axis=1, labels=norm_sample_names)
+        sample_variable_matrix.set_axis(
+            inplace=True, axis=1, labels=norm_sample_names)
 
     return sample_variable_matrix
 

isatools/isatab.py

@@ -8,6 +8,7 @@
 import uuid
 from functools import reduce
 from zipfile import ZipFile
+import sys
 
 
 from isatools import isatab
@@ -866,3 +867,11 @@ def remove_unused_protocols(self):
             investigation, output_path=os.path.dirname(self.path),
             i_file_name='{filename}.fix'.format(
                 filename=os.path.basename(self.path)), skip_dump_tables=True)
+
+
+def utf8_text_file_open(path):
+    if sys.version_info[0] < 3:
+        fp = open(path, 'rb')
+    else:
+        fp = open(path, 'r', newline='', encoding='utf8')
+    return fp

isatools/utils.py

@@ -1,6 +1,6 @@
 numpy
 jsonschema
-pandas==0.20.*
+pandas
 networkx
 behave
 httpretty

requirements-tests.txt

@@ -1,6 +1,6 @@
 numpy
 jsonschema
-pandas==0.20.*
+pandas
 networkx
 lxml
 requests

requirements.txt

@@ -4,7 +4,7 @@
 
 setup(
     name='isatools',
-    version='0.10.2',
+    version='0.10.3',
     packages=['isatools',
               'isatools.convert',
               'isatools.create',
@@ -48,7 +48,7 @@
     install_requires=[
         'numpy',
         'jsonschema',
-        'pandas==0.20.*',
+        'pandas',
         'networkx',
         'lxml',
         'requests',

setup.py

@@ -150,7 +150,6 @@ def test_serialize_ms_assay_topology_modifiers(self):
                 json.dumps(top_mods, cls=SampleAssayPlanEncoder)
             )
         )
-        print(json.dumps(top_mods, cls=SampleAssayPlanEncoder, indent=4))
         self.assertTrue(expected == actual)
 
     @unittest.skip(

tests/test_create_models_json.py

@@ -867,9 +867,7 @@ def test_study_from_2_level_factorial_plan(self):
         study = IsaModelObjectFactory(study_design).create_assays_from_plan()
         self.assertEqual(len(study.assays), 6)
         self.assertEqual(len(study.protocols), 4)
-        study.filename = 's_study.txt'
-        from isatools import isatab
-        print(isatab.dumps(Investigation(studies=[study])))
+
 
     def test_study_from_2_by_3_by_2_factorial_plan(self):
         factor1 = StudyFactor(name='1')

tests/test_create_models_study_design.py

@@ -13,6 +13,7 @@
 from isatools.model import *
 from isatools.tests.utils import assert_tab_content_equal
 from isatools.tests import utils
+from isatools.isatab import IsaTabDataFrame
 
 
 def setUpModule():
@@ -911,8 +912,7 @@ def test_source_protocol_ref_sample(self):
         factory = ProcessSequenceFactory(study_protocols=[Protocol(name="sample collection")])
         table_to_load = """Source Name\tProtocol REF\tSample Name
 source1\tsample collection\tsample1"""
-        DF = pd.read_csv(StringIO(table_to_load), sep='\t')
-        DF.isatab_header = ["Source Name", "Protocol REF", "Sample Name"]
+        DF = IsaTabDataFrame(pd.read_csv(StringIO(table_to_load), sep='\t'))
         so, sa, om, d, pr, _, __ = factory.create_from_df(DF)
         self.assertEqual(len(so), 1)
         self.assertEqual(len(sa), 1)
@@ -925,8 +925,7 @@ def test_source_protocol_ref_sample_x2(self):
         table_to_load = """Source Name\tProtocol REF\tSample Name
 source1\tsample collection\tsample1
 source2\tsample collection\tsample2"""
-        DF = pd.read_csv(StringIO(table_to_load), sep='\t')
-        DF.isatab_header = ["Source Name", "Protocol REF", "Sample Name"]
+        DF = IsaTabDataFrame(pd.read_csv(StringIO(table_to_load), sep='\t'))
         so, sa, om, d, pr, _, __ = factory.create_from_df(DF)
         self.assertEqual(len(so), 2)
         self.assertEqual(len(sa), 2)
@@ -939,8 +938,7 @@ def test_source_protocol_ref_split_sample(self):
         table_to_load = """Source Name\tProtocol REF\tSample Name
 source1\tsample collection\tsample1
 source1\tsample collection\tsample2"""
-        DF = pd.read_csv(StringIO(table_to_load), sep='\t')
-        DF.isatab_header = ["Source Name", "Protocol REF", "Sample Name"]
+        DF = IsaTabDataFrame(pd.read_csv(StringIO(table_to_load), sep='\t'))
         so, sa, om, d, pr, _, __ = factory.create_from_df(DF)
         self.assertEqual(len(so), 1)
         self.assertEqual(len(sa), 2)
@@ -953,8 +951,7 @@ def test_source_protocol_ref_pool_sample(self):
         table_to_load = """Source Name\tProtocol REF\tSample Name
 source1\tsample collection\tsample1
 source2\tsample collection\tsample1"""
-        DF = pd.read_csv(StringIO(table_to_load), sep='\t')
-        DF.isatab_header = ["Source Name", "Protocol REF", "Sample Name"]
+        DF = IsaTabDataFrame(pd.read_csv(StringIO(table_to_load), sep='\t'))
         so, sa, om, d, pr, _, __ = factory.create_from_df(DF)
         self.assertEqual(len(so), 2)
         self.assertEqual(len(sa), 1)
@@ -969,8 +966,7 @@ def test_sample_protocol_ref_split_extract_protocol_ref_data(self):
         table_to_load = """Sample Name\tProtocol REF\tExtract Name\tProtocol REF\tRaw Data File
 sample1\textraction\te1\tscanning\td1
 sample1\textraction\te2\tscanning\td2"""
-        DF = pd.read_csv(StringIO(table_to_load), sep='\t')
-        DF.isatab_header = ["Source Name", "Protocol REF", "Extract Name", "Protocol REF", "Raw Data File"]
+        DF = IsaTabDataFrame(pd.read_csv(StringIO(table_to_load), sep='\t'))
         so, sa, om, d, pr, _, __ = factory.create_from_df(DF)
         self.assertEqual(len(so), 0)
         self.assertEqual(len(sa), 1)