[WIP] CI move to pytest

.travis.yml

@@ -1,13 +1,12 @@
 language: python
 python:
   - "2.7"
-  - "3.2"
-  - "3.3"
   - "3.4"
   - "3.5"
-  - "3.5-dev" # 3.5 development branch
-  - "nightly" # currently points to 3.6-dev
+  - "3.6"
 # command to install dependencies
-install: "pip install -r requirements.txt"
+install:
+  - pip install -r requirements.txt
+  - pip install pytest pytest-cov pandas
 # command to run tests
-script: ./test.py
+script: py.test --cov=spc tests

tests/test_spc.py

@@ -0,0 +1,108 @@
+import os
+from os.path import join, dirname
+
+import numpy as np
+import pandas as pd
+from numpy.testing import assert_allclose
+
+import spc
+
+
+def _list_all_spc_files():
+    """List the spc files in the data folders."""
+    path_data = join(dirname(__file__), 'data')
+    filenames = [filename
+                 for filename in os.listdir(path_data)
+                 if filename.lower().endswith('spc')]
+    return filenames
+
+
+def _split_data(spc_file):
+    """Reorganize into numpy format the data depending of the format.
+
+    Parameters
+    ----------
+    spc_file : spc.File
+        The spc.File object to be reorganized to ease the testing.
+
+    Returns
+    -------
+    X : ndarray, shape (n_spectra, n_frequency) or (n_frequency,)
+        The x axis.
+
+    Y : ndarray shape (n_spectra, n_frequency)
+        The y axis.
+
+    """
+    if (spc_file.dat_fmt == 'gx-y' or
+            spc_file.dat_fmt == 'x-y'):
+        X = spc_file.x
+    elif spc_file.dat_fmt == '-xy':
+        X = np.array([f.x for f in spc_file.sub])
+    Y = np.array([f.y for f in spc_file.sub])
+
+    return X, Y
+
+
+def _read_expected_file(spc_filename, spc_file_format):
+    """Reorganize into numpy format the data depending of the format.
+
+    Parameters
+    ----------
+    spc_filename : str
+        The filename of the SPC file which will be checked.
+
+    spc_file_format : str
+        The format of the SPC file: '-xy', 'x-y', or 'gx-y'.
+
+    Returns
+    -------
+    X : ndarray, shape (n_spectra, n_frequency) or (n_frequency,)
+        The x axis.
+
+    Y : ndarray shape (n_spectra, n_frequency)
+        The y axis.
+
+    """
+    path_data = join(dirname(__file__), 'data', 'txt2')
+    data_expected = pd.read_csv(join(path_data, spc_filename + '.txt'),
+                                sep='\t', header=None, skip_blank_lines=False)
+    if (spc_file_format == 'gx-y' or
+            spc_file_format == 'x-y'):
+        X = data_expected.iloc[:, 0].values.T
+        Y = data_expected.iloc[:, 1:].values.T
+    elif spc_file_format == '-xy':
+        data_expected['group_no'] = data_expected.isnull().all(axis=1).cumsum()
+        # split into a list of DataFrames
+        data_expected = [
+            (data_expected.loc[data_expected['group_no'] == i,
+                               [0, 1]]).dropna()
+            for i in range(data_expected['group_no'].iloc[-1])]
+        X, Y = [], []
+        for df in data_expected:
+            X.append(df.iloc[:, 0].values.T)
+            Y.append(df.iloc[:, 1].values.T)
+        X = np.array(X)
+        Y = np.array(Y)
+
+    return X, Y
+
+
+def test_spc():
+    filenames_spc = _list_all_spc_files()
+
+    for filename in filenames_spc:
+        path_data = join(dirname(__file__), 'data')
+        file_spc = spc.File(join(path_data, filename))
+        X, Y = _split_data(file_spc)
+        X_expected, Y_expected = _read_expected_file(filename,
+                                                     file_spc.dat_fmt)
+        if file_spc.dat_fmt == '-xy':
+            for x, y, x_expected, y_expected in zip(X, Y,
+                                                    X_expected,
+                                                    Y_expected):
+                assert_allclose(x, x_expected)
+                assert_allclose(y, y_expected)
+        else:
+            assert_allclose(X, X_expected)
+            assert_allclose(Y, Y_expected)