Merge 73ba5d1 into 426b965

analyses/lm_objects/matrix.py

@@ -93,6 +93,60 @@ def load(cls, flo):
                     'Cannot load matrix data from file-like object provided',
                     str(e)))
 
+    # ...........................
+    @classmethod
+    def load_csv(cls, flo, dtype=np.float, num_header_rows=0,
+                 num_header_cols=0):
+        """Attempts to load a Matrix object from a CSV file-like object.
+
+        Args:
+            flo (file-like): A file-like object with matrix data.
+            dtype (:obj:`method`, optional): The data type for the data.  Will
+                be used to cast data when adding to matrix.
+            num_header_rows (:obj:`int`, optional): The number of header rows
+                in the CSV file.
+            num_header_cols (:obj:`int`, optional): The number of header
+                columns in the CSV file.
+
+        Returns:
+            Matrix: The newly loaded Matrix object.
+        """
+        col_headers = []
+        row_headers = []
+        header_lines = []  # Leading rows that are headers
+        data = []
+        i = 0
+        for line in flo:
+            items = line.strip().split(',')
+            # If header row, add to header rows for processing
+            if i < num_header_rows:
+                # Add the headers to header lines for processing
+                header_lines.append(items[num_header_cols:])
+            else:
+                if num_header_cols == 1:
+                    row_headers.append(items[0])
+                elif num_header_cols > 1:
+                    row_headers.append(items[:num_header_cols])
+                data.append([dtype(x) for x in items[num_header_cols:]])
+
+            i += 1
+
+        print(header_lines)
+
+        # Process header columns from header rows
+        if num_header_rows == 1:
+            col_headers = header_lines[0]
+        elif num_header_rows > 1:
+            for j in range(len(header_lines[0])):
+                h = []
+                for x in range(num_header_rows):
+                    h.append(header_lines[x][j])
+                col_headers.append(h)
+
+        data_array = np.array(data)
+
+        return cls(data_array, headers={'0': row_headers, '1': col_headers})
+
     # ...........................
     @classmethod
     def load_new(cls, flo):
@@ -529,11 +583,20 @@ def csv_generator():
 
             # Start with the header row, if we have one
             if '1' in mtx.headers and mtx.headers['1']:
-                # Add a blank entry if we have row headers
-                header_row = ['']*len(listify(
-                                    row_headers[0]) if row_headers else [])
-                header_row.extend(mtx.headers['1'])
-                yield header_row
+                # Make column headers lists of lists
+                if not isinstance(mtx.headers['1'][0], (tuple, list)):
+                    header_row = ['']*len(
+                        listify(row_headers[0]) if row_headers else [])
+                    header_row.extend(mtx.headers['1'])
+                    yield header_row
+                else:
+                    for i in range(len(mtx.headers['1'][0])):
+                        header_row = ['']*len(
+                            listify(row_headers[0]) if row_headers else [])
+                        header_row.extend(
+                            [mtx.headers['1'][j][i] for j in range(
+                                len(mtx.headers['1']))])
+                        yield header_row
             # For each row in the data set
             for i in range(mtx.data.shape[0]):
                 # Add the row headers if exists

analyses/lm_objects/tree.py

@@ -5,6 +5,8 @@
     * Add method to remove annotations.
     * Move label method out of internal functions.
 """
+import os
+
 import dendropy
 import numpy as np
 
@@ -59,6 +61,33 @@ def from_base_tree(cls, tree):
         """
         return cls.get(data=tree.as_string('nexus'), schema='nexus')
 
+    # ..............................
+    @classmethod
+    def from_filename(cls, filename):
+        """Creates a TreeWrapper object by loading a file.
+
+        Args:
+            filename (str): A file path to a tree file that should be loaded.
+
+        Returns:
+            TreeWrapper: The newly loaded tree.
+
+        Raises:
+            IOError: Raised if the tree file cannot be loaded based on the file
+                extension.
+        """
+        _, tree_ext = os.path.splitext(filename)
+        if tree_ext == '.nex':
+            tree_schema = 'nexus'
+        elif tree_ext == '.xml':
+            tree_schema = 'nexml'
+        elif tree_ext == '.tre':
+            tree_schema = 'newick'
+        else:
+            raise IOError(
+                'Cannot handle tree with extension: {}'.format(tree_ext))
+        return cls.get(path=filename, schema=tree_schema)
+
     # ..............................
     def add_node_labels(self, prefix=None, overwrite=False):
         """Add labels to the nodes in the tree.

tests/conftest.py

@@ -110,6 +110,8 @@ def _get_format_extension(self, fmt):
             return '.tre'
         elif fmt.lower() == 'nexus':
             return '.nex'
+        elif fmt.lower() == 'nexml':
+            return '.xml'
         elif fmt.lower() == 'phylip':
             return '.phylip'
         elif fmt.lower() == 'table':
@@ -212,6 +214,7 @@ def pytest_generate_tests(metafunc):
         ('valid_csv_alignment', df.get_alignments('csv', True)),
         ('valid_json_alignment', df.get_alignments('json', True)),
         ('valid_newick_tree', df.get_trees('newick', True)),
+        ('valid_nexml_tree', df.get_trees('nexml', True)),
         ('valid_nexus_tree', df.get_trees('nexus', True)),
         ('valid_phylip_alignment', df.get_alignments('phylip', True)),
         ('valid_table_alignment', df.get_alignments('table', True))

tests/data_dir/trees/valid_tree.xml

@@ -0,0 +1,50 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<nex:nexml
+    version="0.9"
+    xsi:schemaLocation="http://www.nexml.org/2009 ../xsd/nexml.xsd"
+    xmlns="http://www.nexml.org/2009"
+    xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+    xmlns:xml="http://www.w3.org/XML/1998/namespace"
+    xmlns:nex="http://www.nexml.org/2009"
+    xmlns:xsd="http://www.w3.org/2001/XMLSchema#"
+>
+    <otus id="d0">
+        <otu id="d1" label="A" />
+        <otu id="d2" label="B" />
+        <otu id="d3" label="C" />
+        <otu id="d4" label="G" />
+        <otu id="d5" label="D" />
+        <otu id="d6" label="E" />
+        <otu id="d7" label="F" />
+    </otus>
+    <trees id="d8" otus="d0">
+        <tree id="d9" xsi:type="nex:FloatTree">
+            <node id="d10" />
+            <node id="d11" otu="d1" />
+            <node id="d12" />
+            <node id="d13" />
+            <node id="d14" otu="d2" />
+            <node id="d15" otu="d3" />
+            <node id="d16" />
+            <node id="d17" otu="d4" />
+            <node id="d18" />
+            <node id="d19" otu="d5" />
+            <node id="d20" />
+            <node id="d21" otu="d6" />
+            <node id="d22" otu="d7" />
+            <rootedge id="d23" target="d10" />
+            <edge id="d24" source="d10" target="d11" length="2.9999" />
+            <edge id="d25" source="d10" target="d12" length="0.1" />
+            <edge id="d26" source="d12" target="d13" length="0.1" />
+            <edge id="d27" source="d13" target="d14" length="0.1" />
+            <edge id="d28" source="d13" target="d15" length="0.1" />
+            <edge id="d29" source="d12" target="d16" length="0.1" />
+            <edge id="d30" source="d16" target="d17" length="0.2" />
+            <edge id="d31" source="d16" target="d18" length="0.1" />
+            <edge id="d32" source="d18" target="d19" length="0.1" />
+            <edge id="d33" source="d18" target="d20" length="0.1" />
+            <edge id="d34" source="d20" target="d21" length="0.1" />
+            <edge id="d35" source="d20" target="d22" length="0.1" />
+        </tree>
+    </trees>
+</nex:nexml>

tests/test_lm_objects/test_matrix.py

@@ -65,6 +65,63 @@ def test_load(self):
         with pytest.raises(IOError):
             mtx = matrix.Matrix.load(io.BytesIO())
 
+    # .....................................
+    def test_load_csv(self):
+        """Test the load_csv method.
+        """
+        orig_mtx = get_random_matrix(5, 5)
+
+        with io.StringIO() as out_str:
+            orig_mtx.write_csv(out_str)
+            out_str.seek(0)
+
+            # Attempt to load matrix
+            loaded_mtx = matrix.Matrix.load_csv(
+                out_str, num_header_rows=1, num_header_cols=1)
+
+        print(loaded_mtx.get_headers())
+        print(orig_mtx.get_headers())
+
+        # Verify data and headers are the same
+        assert np.allclose(loaded_mtx.data, orig_mtx.data)
+        assert loaded_mtx.get_headers() == orig_mtx.get_headers()
+
+    # .....................................
+    def test_load_csv_multi_headers(self):
+        """Test the load_csv method.
+        """
+        orig_mtx = get_random_matrix(5, 5)
+        new_row_headers = []
+        new_col_headers = []
+        orig_row_headers = orig_mtx.get_row_headers()
+        orig_col_headers = orig_mtx.get_column_headers()
+
+        for h in orig_row_headers:
+            new_row_headers.append([h, '{}-2'.format(h)])
+        orig_mtx.set_row_headers(new_row_headers)
+        for h in orig_col_headers:
+            new_col_headers.append([h, '{}-2'.format(h)])
+        orig_mtx.set_column_headers(new_col_headers)
+
+        print(orig_mtx.get_headers())
+
+        with io.StringIO() as out_str:
+            orig_mtx.write_csv(out_str)
+            out_str.seek(0)
+            print(out_str.getvalue())
+            out_str.seek(0)
+
+            # Attempt to load matrix
+            loaded_mtx = matrix.Matrix.load_csv(
+                out_str, num_header_rows=2, num_header_cols=2)
+
+        print(loaded_mtx.get_headers())
+        print(orig_mtx.get_headers())
+
+        # Verify data and headers are the same
+        assert np.allclose(loaded_mtx.data, orig_mtx.data)
+        assert loaded_mtx.get_headers() == orig_mtx.get_headers()
+
     # .....................................
     def test_load_new(self):
         """Test the load_new method.

tests/test_lm_objects/test_tree.py

@@ -82,6 +82,58 @@ def test_from_base_nexus_tree(self, valid_nexus_tree):
         wrapped_tree = tree.TreeWrapper.from_base_tree(dendropy_tree)
         assert isinstance(wrapped_tree, tree.TreeWrapper)
 
+    # .....................................
+    def test_from_filename_newick_tree(self, valid_newick_tree):
+        """Attempt to get a tree using a newick file.
+
+        Args:
+            valid_newick_tree (pytest.fixture): A parameterized pytest fixture
+                that provides valid newick trees, one at a time, to this test
+                function.
+        """
+        loaded_tree = tree.TreeWrapper.from_filename(valid_newick_tree)
+        assert isinstance(loaded_tree, tree.TreeWrapper)
+
+    # .....................................
+    def test_from_filename_nexml_tree(self, valid_nexml_tree):
+        """Attempt to get a tree using a nexml file.
+
+        Args:
+            valid_nexml_tree (pytest.fixture): A parameterized pytest fixture
+                that provides valid nexml trees, one at a time, to this test
+                function.
+        """
+        loaded_tree = tree.TreeWrapper.from_filename(valid_nexml_tree)
+        assert isinstance(loaded_tree, tree.TreeWrapper)
+
+    # .....................................
+    def test_from_filename_nexus_tree(self, valid_nexus_tree):
+        """Attempt to get a tree using a nexus file.
+
+        Args:
+            valid_nexus_tree (pytest.fixture): A parameterized pytest fixture
+                that provides valid nexus trees, one at a time, to this test
+                function.
+        """
+        loaded_tree = tree.TreeWrapper.from_filename(valid_nexus_tree)
+        assert isinstance(loaded_tree, tree.TreeWrapper)
+
+    # .....................................
+    def test_from_filename_invalid_file(self, invalid_csv_alignment):
+        """Attempt to get a tree using a nexus file.
+
+        Args:
+            invalid_csv_alignment (pytest.fixture): A parameterized pytest
+                fixture that provides a CSV filename, one at a time, to test
+                this function.  Trees cannot be loaded from CSV, so this should
+                raise an IOError.
+            valid_nexus_tree (pytest.fixture): A parameterized pytest fixture
+                that provides valid nexus trees, one at a time, to this test
+                function.
+        """
+        with pytest.raises(IOError):
+            tree.TreeWrapper.from_filename(invalid_csv_alignment)
+
     # .....................................
     def test_add_node_labels_no_prefix_no_overwrite(self):
         """Test that node labels are added correctly to a tree.