Remove extraneous str casts (#159)

* Remove extraneous str casts

* Edge case checking for empty tables

* imports and test was doing unnecessary things:

* kick tests

* kick tests

* the value of a local env

unifrac/_methods.py

@@ -50,6 +50,16 @@ def is_biom_v210(f, ids=None):
     return True
 
 
+def has_samples_biom_v210(f):
+    # assumes this is already checked to be biom v210
+    import h5py
+    with h5py.File(f, 'r') as fp:
+        if len(fp['sample/ids']) == 0:
+            return False
+
+    return True
+
+
 def is_newick(f):
     sniffer = skbio.io.format.newick.newick.sniffer_function
     return sniffer(f)[0]
@@ -58,12 +68,16 @@ def is_newick(f):
 def _validate(table, phylogeny, ids=None):
     if not is_biom_v210(table, ids):
         raise ValueError("Table does not appear to be a BIOM-Format v2.1")
+    if not has_samples_biom_v210(table):
+        raise ValueError("Table does not contain any samples")
     if not is_newick(phylogeny):
         raise ValueError("The phylogeny does not appear to be newick")
 
 
 def _call_ssu(table, phylogeny, *args):
     if isinstance(table, Table) and isinstance(phylogeny, (TreeNode, BP)):
+        if table.is_empty():
+            raise ValueError("Table does not contain any samples")
         return qsu.ssu_inmem(table, phylogeny, *args)
     elif isinstance(table, str) and isinstance(phylogeny, str):
         ids = []
@@ -356,7 +370,7 @@ def weighted_normalized(table: Union[str, Table],
        powerful beta diversity measure for comparing communities based on
        phylogeny. BMC Bioinformatics 12:118 (2011).
     """
-    return _call_ssu(str(table), str(phylogeny), 'weighted_normalized',
+    return _call_ssu(table, phylogeny, 'weighted_normalized',
                      variance_adjusted, 1.0, bypass_tips, n_substeps)
 
 
@@ -423,7 +437,7 @@ def weighted_normalized_fp64(table: Union[str, Table],
        powerful beta diversity measure for comparing communities based on
        phylogeny. BMC Bioinformatics 12:118 (2011).
     """
-    return _call_ssu(str(table), str(phylogeny), 'weighted_normalized_fp64',
+    return _call_ssu(table, phylogeny, 'weighted_normalized_fp64',
                      variance_adjusted, 1.0, bypass_tips, n_substeps)
 
 
@@ -490,7 +504,7 @@ def weighted_normalized_fp32(table: Union[str, Table],
        powerful beta diversity measure for comparing communities based on
        phylogeny. BMC Bioinformatics 12:118 (2011).
     """
-    return _call_ssu(str(table), str(phylogeny), 'weighted_normalized_fp32',
+    return _call_ssu(table, phylogeny, 'weighted_normalized_fp32',
                      variance_adjusted, 1.0, bypass_tips, n_substeps)
 
 
@@ -557,7 +571,7 @@ def weighted_unnormalized(table: Union[str, Table],
        powerful beta diversity measure for comparing communities based on
        phylogeny. BMC Bioinformatics 12:118 (2011).
     """
-    return _call_ssu(str(table), str(phylogeny), 'weighted_unnormalized',
+    return _call_ssu(table, phylogeny, 'weighted_unnormalized',
                      variance_adjusted, 1.0, bypass_tips, n_substeps)
 
 
@@ -625,7 +639,7 @@ def weighted_unnormalized_fp64(table: Union[str, Table],
        powerful beta diversity measure for comparing communities based on
        phylogeny. BMC Bioinformatics 12:118 (2011).
     """
-    return _call_ssu(str(table), str(phylogeny), 'weighted_unnormalized_fp64',
+    return _call_ssu(table, phylogeny, 'weighted_unnormalized_fp64',
                      variance_adjusted, 1.0, bypass_tips, n_substeps)
 
 
@@ -693,7 +707,7 @@ def weighted_unnormalized_fp32(table: Union[str, Table],
        powerful beta diversity measure for comparing communities based on
        phylogeny. BMC Bioinformatics 12:118 (2011).
     """
-    return _call_ssu(str(table), str(phylogeny), 'weighted_unnormalized_fp32',
+    return _call_ssu(table, phylogeny, 'weighted_unnormalized_fp32',
                      variance_adjusted, 1.0, bypass_tips, n_substeps)
 
 
@@ -779,7 +793,7 @@ def generalized(table: Union[str, Table],
         return weighted_normalized(table, phylogeny, threads,
                                    variance_adjusted, bypass_tips, n_substeps)
     else:
-        return _call_ssu(str(table), str(phylogeny), 'generalized',
+        return _call_ssu(table, phylogeny, 'generalized',
                          variance_adjusted, alpha, bypass_tips, n_substeps)
 
 
@@ -866,7 +880,7 @@ def generalized_fp64(table: Union[str, Table],
                                         variance_adjusted, bypass_tips,
                                         n_substeps)
     else:
-        return _call_ssu(str(table), str(phylogeny), 'generalized_fp64',
+        return _call_ssu(table, phylogeny, 'generalized_fp64',
                          variance_adjusted, alpha, bypass_tips, n_substeps)
 
 
@@ -953,7 +967,7 @@ def generalized_fp32(table: Union[str, Table],
                                         variance_adjusted, bypass_tips,
                                         n_substeps)
     else:
-        return _call_ssu(str(table), str(phylogeny), 'generalized_fp32',
+        return _call_ssu(table, phylogeny, 'generalized_fp32',
                          variance_adjusted, alpha, bypass_tips, n_substeps)
 
 

unifrac/tests/test_methods.py

@@ -7,11 +7,16 @@
 # ----------------------------------------------------------------------------
 import unittest
 import pkg_resources
+import os
 
+import h5py
+import biom
+import skbio
 import numpy as np
 import numpy.testing as npt
 
 from unifrac import meta
+from unifrac._methods import _call_ssu, has_samples_biom_v210
 
 
 class StateUnifracTests(unittest.TestCase):
@@ -107,6 +112,26 @@ def test_meta_unifrac_alpha_not_generalized(self):
             meta((self.table1, ), (self.tree1, ), method='unweighted',
                  alpha=1, consolidation='skipping_missing_matrices')
 
+    def test_has_samples_biom_v210(self):
+        fp = self.get_data_path('crawford.biom')
+        self.assertTrue(has_samples_biom_v210(fp))
+
+        tmpfile = '/tmp/fake.biom'
+        empty = biom.Table([], [], [])
+        try:
+            with h5py.File(tmpfile, 'w') as fp:
+                empty.to_hdf5(fp, 'asd')
+            fp = self.get_data_path(tmpfile)
+            self.assertFalse(has_samples_biom_v210(tmpfile))
+        finally:
+            os.unlink(tmpfile)
+
+    def test_call_ssu_empty_biom(self):
+        empty = biom.Table([], [], [])
+        tre = skbio.TreeNode()
+        with self.assertRaisesRegex(ValueError, "contain any samples"):
+            _call_ssu(empty, tre)
+
 
 if __name__ == "__main__":
     unittest.main()