Merge pull request #671 from jairideout/convert-finalization

biom convert finalization

biom/cli/__init__.py

@@ -13,33 +13,8 @@
 from .metadata_adder import add_metadata
 from .table_validator import validate_table
 from .table_converter import convert
-
-import biom.parse
-import biom.util
+from .util import write_biom_table
 
 __all__ = ['validate_table', 'summarize_table', 'add_metadata',
            'show_install_info', 'normalize_table', 'subset_table',
-           'convert']
-
-
-def write_biom_table(table, fmt, filepath):
-    """Write table in specified format to filepath"""
-
-    if fmt not in ['hdf5', 'json', 'tsv']:
-        raise ValueError("Unknown file format")
-
-    if fmt == 'hdf5' and not biom.util.HAVE_H5PY:
-        fmt = 'json'
-
-    if fmt == 'json':
-        with open(filepath, 'w') as f:
-            f.write(table.to_json(biom.parse.generatedby()))
-    elif fmt == 'tsv':
-        with open(filepath, 'w') as f:
-            f.write(table)
-            f.write('\n')
-    else:
-        import h5py
-
-        with h5py.File(filepath, 'w') as f:
-            table.to_hdf5(f, biom.parse.generatedby())
+           'convert', 'write_biom_table']

biom/cli/table_converter.py

@@ -8,6 +8,8 @@
 
 from __future__ import division
 
+from biom.cli.util import write_biom_table
+
 table_types = ["OTU table",
                "Pathway table",
                "Function table",
@@ -30,17 +32,26 @@
 }
 
 
-def convert(table, sample_metadata, observation_metadata,
-            to_json, to_hdf5, to_tsv, collapsed_samples,
-            collapsed_observations, header_key, output_metadata_id, table_type,
-            process_obs_metadata, tsv_metadata_formatter):
+def convert(table, output_filepath, sample_metadata=None,
+            observation_metadata=None,
+            to_json=False, to_hdf5=False, to_tsv=False,
+            collapsed_samples=False, collapsed_observations=False,
+            header_key=None, output_metadata_id=None, table_type=None,
+            process_obs_metadata=None, tsv_metadata_formatter='sc_separated'):
 
     if sum([to_tsv, to_hdf5, to_json]) == 0:
         raise ValueError("Must specify an output format")
     elif sum([to_tsv, to_hdf5, to_json]) > 1:
         raise ValueError("Can only specify a single output format")
 
-    table.type = table_type
+    if table_type is None:
+        if table.type in [None, "None"]:
+            table.type = "Table"
+        else:
+            pass
+    else:
+        table.type = table_type
+
     if tsv_metadata_formatter is not None:
         obs_md_fmt_f = observation_metadata_formatters[tsv_metadata_formatter]
 
@@ -75,9 +86,14 @@ def convert(table, sample_metadata, observation_metadata,
         result = table.to_tsv(header_key=header_key,
                               header_value=output_metadata_id,
                               metadata_formatter=obs_md_fmt_f)
+        with open(output_filepath, 'w') as f:
+            f.write(result)
+        return
     elif to_json:
+        fmt = 'json'
         result = table
     elif to_hdf5:
+        fmt = 'hdf5'
         result = table
         if collapsed_observations:
             metadata = [{'collapsed_ids': md.keys()}
@@ -91,5 +107,6 @@ def convert(table, sample_metadata, observation_metadata,
             # We have changed the metadata, it is safer to make sure that
             # it is correct
             result._cast_metadata()
+    write_biom_table(result, fmt, output_filepath)
 
-    return result
+    return

biom/cli/util.py

@@ -0,0 +1,33 @@
+# ----------------------------------------------------------------------------
+# Copyright (c) 2011-2015, The BIOM Format Development Team.
+#
+# Distributed under the terms of the Modified BSD License.
+#
+# The full license is in the file COPYING.txt, distributed with this software.
+# ----------------------------------------------------------------------------
+
+import biom.util
+import biom.parse
+
+
+def write_biom_table(table, fmt, filepath):
+    """Write table in specified format to filepath"""
+
+    if fmt not in ['hdf5', 'json', 'tsv']:
+        raise ValueError("Unknown file format")
+
+    if fmt == 'hdf5' and not biom.util.HAVE_H5PY:
+        fmt = 'json'
+
+    if fmt == 'json':
+        with open(filepath, 'w') as f:
+            f.write(table.to_json(biom.parse.generatedby()))
+    elif fmt == 'tsv':
+        with open(filepath, 'w') as f:
+            f.write(table)
+            f.write('\n')
+    else:
+        import h5py
+
+        with h5py.File(filepath, 'w') as f:
+            table.to_hdf5(f, biom.parse.generatedby())

scripts/biom

@@ -199,7 +199,7 @@ def add_metadata(input_fp, output_fp, sample_metadata_fp,
               help='The name to be given to the observation metadata '
                    'column when creating a tsv table file if the column '
                    'should be renamed.')
-@click.option('--table-type', required=False, default="Table",
+@click.option('--table-type', required=False,
               type=click.Choice(biom.cli.table_converter.table_types),
               help='The type of the table.')
 @click.option('--process-obs-metadata', required=False,
@@ -253,13 +253,12 @@ def convert(input_fp, output_fp, sample_metadata_fp, observation_metadata_fp,
     else:
         observation_metadata_f = None
 
-    result = biom.cli.convert(table, sample_metadata_f, observation_metadata_f,
-                              to_json, to_hdf5, to_tsv, collapsed_samples,
-                              collapsed_observations, header_key,
-                              output_metadata_id, table_type,
-                              process_obs_metadata, tsv_metadata_formatter)
-
-    biom.cli.write_biom_table(result, fmt, output_fp)
+    biom.cli.convert(table, output_fp, sample_metadata_f,
+                     observation_metadata_f,
+                     to_json, to_hdf5, to_tsv, collapsed_samples,
+                     collapsed_observations, header_key,
+                     output_metadata_id, table_type,
+                     process_obs_metadata, tsv_metadata_formatter)
 
 @cli.command(name='show-install-info')
 def show_install_info():