Fix warnings, add warnings

Author: Sparks29032

news/docs_patch.rst

@@ -1,10 +1,12 @@
 **Added:**
 
 * New documentation build.
+* Quickstart tutorial for parsers.
 
 **Changed:**
 
-* <news item>
+* Theme changed from `sphinx_py3doc_enhanced_theme` to `sphinx_rtd_theme`.
+* User now warned when data_table data overwrites hdata (header) data.
 
 **Deprecated:**
 

rever.xsh

@@ -1,8 +1,8 @@
 $PROJECT = 'diffpy.utils'
 $ACTIVITIES = [
-              'changelog', # Create a CHANGELOG.rst using news directory
               'tag',  # Creates a tag for the new version number
               'push_tag',  # Pushes the tag up to the $TAG_REMOTE
+              'changelog', # Create a CHANGELOG.rst using news directory
               'ghrelease',  # Creates a Github release entry for the new tag
               'pypi',  # Sends the package to pypi
                ]

src/diffpy/utils/parsers/loaddata.py

@@ -88,7 +88,7 @@ def countcolumnsvalues(line):
             if usecols is not None:
                 nv = len([float(words[i]) for i in usecols])
             else:
-                nv = len([float(w) for w in words])
+                nv = len([  float(w) for w in words])
         except (IndexError, ValueError):
             nc = nv = 0
         return nc, nv

src/diffpy/utils/parsers/resample.py

@@ -26,7 +26,7 @@ def wsinterp(x, xp, fp, left=None, right=None):
     This uses the Whittaker-Shannon interpolation formula to interpolate the value of fp (array), which is defined over
     xp (array), at x (array or float).
 
-    Paramaters
+    Parameters
     ----------
     x: ndarray
         Desired range for interpolation.
@@ -72,7 +72,7 @@ def wsinterp(x, xp, fp, left=None, right=None):
 
     # Return a float if we got a float
     if scalar:
-        return float(fp_at_x)
+        return float(fp_at_x[0])
 
     return fp_at_x
 
@@ -97,7 +97,7 @@ def resample(r, s, dr):
     Returns resampled (r, s).
     """
 
-    dr0 = r[1] - r[0]
+    dr0 = r[1] - r[0]  # Constant timestep
 
     if dr0 < dr:
         rnew = numpy.arange(r[0], r[-1]+0.5*dr, dr)

src/diffpy/utils/parsers/serialization.py

@@ -17,6 +17,7 @@
 import json
 
 from .custom_exceptions import UnsupportedTypeError, ImproperSizeError
+import warnings
 
 # FIXME: add support for yaml, xml
 supported_formats = ['.json']
@@ -64,7 +65,11 @@ def serialize_data(filename, hdata: dict, data_table: list, show_path=True, dt_c
     # title the entry with name of file (taken from end of path)
     title = abs_path.name
 
-    # first add named columns in dt_cols
+    # first add data in hddata dict
+    data.update(hdata)
+
+    # second add named columns in dt_cols
+    # performed second to prioritize overwriting hdata entries with data_table column entries
     named_columns = 0  # initial value
     max_columns = 1  # higher than named_columns to trigger 'data table' entry
     if dt_colnames is not None:
@@ -77,21 +82,18 @@ def serialize_data(filename, hdata: dict, data_table: list, show_path=True, dt_c
         for idx in range(num_col_names):
             colname = dt_colnames[idx]
             if colname is not None:
+                if colname in hdata.keys():
+                    warnings.warn(f'Entry \'{colname}\' in hdata has been overwritten by a data_table entry.',
+                                  RuntimeWarning)
                 data.update({colname: list(data_table[:, idx])})
                 named_columns += 1
 
-    # second add data in hddata dict
-    data.update(hdata)
-
     # finally add data_table as an entry named 'data table' if not all columns were parsed
     if named_columns < max_columns:
-        if 'data table' not in data.keys():
-            data.update({'data table': data_table})
-        else:  # if 'data table' is already a key, keep adding primes to the end
-            dt_name = 'data table'
-            while dt_name in data.keys():
-                dt_name += " prime"
-            data.update({dt_name: data_table})
+        if 'data table' in data.keys():
+            warnings.warn('Entry \'data table\' in hdata has been overwritten by data_table.',
+                          RuntimeWarning)
+        data.update({'data table': data_table})
 
     # parse name using pathlib and generate dictionary entry
     entry = {title: data}

src/diffpy/utils/tests/test_resample.py

@@ -12,7 +12,7 @@ def test_wsinterp():
     # Sampling rate
     ssr = 44100**-1  # Standard sampling rate for human-hearable frequencies
     t = ssr
-    n = 5  # Do 6 samples
+    n = 5
     xp = np.array([i*t for i in range(-n, n+1, 1)])
     x = np.array([i*t for i in range(-n-1, n+2, 1)])
 

src/diffpy/utils/tests/test_serialization.py

@@ -14,6 +14,7 @@
 schemaname = datafile('strumining.json')
 wrongtype = datafile('wrong.type')
 loadfile = datafile('loadfile.txt')
+warningfile = datafile('generatewarnings.txt')
 nodt = datafile('loaddatawithheaders.txt')
 
 
@@ -69,9 +70,7 @@ def test_exceptions():
     # check equivalence
     assert missing_parameter == empty_parameter
     assert missing_parameter == none_entry_parameter
-    print(data_table)
-    print(missing_parameter[data_name]['data table prime'])
-    assert numpy.allclose(missing_parameter[data_name]['data table prime'], data_table)
+    assert numpy.allclose(missing_parameter[data_name]['data table'], data_table)
     # extract a single column
     r_extract = serialize_data(loadfile, hdata, data_table, show_path=False, dt_colnames=['r'])
     gr_extract = serialize_data(loadfile, hdata, data_table, show_path=False, dt_colnames=[None, 'gr'])
@@ -89,3 +88,12 @@ def test_exceptions():
     no_dt = serialize_data(nodt, nodt_hdata, nodt_dt, show_path=False)
     nodt_data_name = list(no_dt.keys())[0]
     assert numpy.allclose(no_dt[nodt_data_name]['data table'], nodt_dt)
+
+    # ensure user is warned when columns are overwritten
+    hdata = loadData(warningfile, headers=True)
+    data_table = loadData(warningfile)
+    with pytest.warns(RuntimeWarning) as record:
+        data = serialize_data(warningfile, hdata, data_table, show_path=False, dt_colnames=['c1', 'c2', 'c3'])
+    assert len(record) == 4
+    for msg in record:
+        assert 'overwritten' in msg.message.args[0]

src/diffpy/utils/tests/testdata/generatewarnings.txt

@@ -0,0 +1,11 @@
+data table = generate data table
+c1 = time
+c2 = position
+c3 = velocity
+
+# label: [c1] [c2] [c3]
+0.0 0.0 0.0 1.0
+1.0 0.5 1.0 1.0
+2.0 2.0 2.0 1.0
+3.0 4.5 3.0 1.0
+4.0 8.0 4.0 1.0

src/diffpy/utils/tests/testdata/loadfile.txt

@@ -15,7 +15,6 @@ qbroad = 0.2
 qdamp = 0.0001
 stype = Neutron
 stru_str = baddata
-data table = no_table
 
 0 0
 1 0.00001