Refactor some IO to io mod #69

tests/io/test_helpers.py

@@ -0,0 +1,19 @@
+""" Tests for ``yatsm.io.helpers``
+"""
+import pytest
+
+from yatsm.io import helpers
+
+
+def test_mkdir_p_success(tmpdir):
+    helpers.mkdir_p(tmpdir.join('test').strpath)
+
+
+def test_mkdir_p_succcess_exists(tmpdir):
+    helpers.mkdir_p(tmpdir.join('test').strpath)
+    helpers.mkdir_p(tmpdir.join('test').strpath)
+
+
+def test_mkdir_p_failure_permission(tmpdir):
+    with pytest.raises(OSError):
+        helpers.mkdir_p('/asdf')

yatsm/cli/line.py

@@ -10,6 +10,7 @@
 from yatsm.cli import options
 from yatsm.config_parser import parse_config_file
 from yatsm.errors import TSLengthException
+from yatsm.io.helpers import mkdir_p
 from yatsm.utils import (distribute_jobs, get_output_name, get_image_IDs,
                          csvfile_to_dataframe)
 from yatsm.reader import get_image_attribute, read_line
@@ -52,20 +53,13 @@ def line(ctx, config, job_number, total_jobs,
     # Make sure output directory exists and is writable
     output_dir = cfg['dataset']['output']
     try:
-        os.makedirs(output_dir)
-    except OSError as e:
-        # File exists
-        if e.errno == 17:
-            pass
-        elif e.errno == 13:
-            click.secho('Cannot create output directory %s' % output_dir,
-                        fg='red')
-            raise click.Abort()
-
+        mkdir_p(output_dir)
+    except OSError as err:
+        raise click.ClickException('Cannot create output directory %s (%s)' %
+                                   (output_dir, str(err)))
     if not os.access(output_dir, os.W_OK):
-        click.secho('Cannot write to output directory %s' % output_dir,
-                    fg='red')
-        raise click.Abort()
+        raise click.ClickException('Cannot write to output directory %s' %
+                                   output_dir)
 
     # Test existence of cache directory
     read_cache, write_cache = test_cache(cfg['dataset'])

yatsm/io/__init__.py

@@ -0,0 +1,9 @@
+""" YATSM IO module
+
+TODO: include result file IO abstraction (issue #69)
+
+Contents:
+
+    * ``helpers``: Collection of helper functions that ease common filesystem
+      operations
+"""

yatsm/io/helpers.py

@@ -0,0 +1,24 @@
+""" Collection of helper functions that ease common filesystem operations
+"""
+import errno
+import os
+
+
+def mkdir_p(d):
+    """ Make a directory, ignoring error if it exists (i.e., ``mkdir -p``)
+
+    Args:
+        d (str): directory path to create
+
+    Raises:
+        OSError: Raise OSError if cannot create directory for reasons other
+            than it existing already (errno 13 "EEXIST")
+    """
+    try:
+        os.makedirs(d)
+    except OSError as err:
+        # File exists
+        if err.errno == errno.EEXIST:
+            pass
+        else:
+            raise err

yatsm/utils.py

@@ -21,18 +21,18 @@ def distribute_jobs(job_number, total_jobs, n, interlaced=True):
     """ Assign `job_number` out of `total_jobs` a subset of `n` tasks
 
     Args:
-      job_number (int): 0-indexed processor to distribute jobs to
-      total_jobs (int): total number of processors running jobs
-      n (int): number of tasks (e.g., lines in image, regions in segment)
-      interlaced (bool, optional): interlace job assignment (default: True)
+        job_number (int): 0-indexed processor to distribute jobs to
+        total_jobs (int): total number of processors running jobs
+        n (int): number of tasks (e.g., lines in image, regions in segment)
+        interlaced (bool, optional): interlace job assignment (default: True)
 
     Returns:
-      np.ndarray: np.ndarray of task IDs to be processed
+        np.ndarray: np.ndarray of task IDs to be processed
 
     Raises:
-      ValueError: raise error if `job_number` and `total_jobs` specified
-        result in no jobs being assinged (happens if `job_number` and
-        `total_jobs` are both 1)
+        ValueError: raise error if `job_number` and `total_jobs` specified
+            result in no jobs being assinged (happens if `job_number` and
+            `total_jobs` are both 1)
 
     """
     if interlaced: