Try to improve unicode string handling for annotations

holgern · May 2, 2017 · ccfc0ad · ccfc0ad
1 parent 54ae3ea
commit ccfc0ad
Show file tree

Hide file tree

Showing 3 changed files with 23 additions and 18 deletions.
diff --git a/pyedflib/edfreader.py b/pyedflib/edfreader.py
@@ -55,12 +55,13 @@ def readAnnotations(self):
             return np.array([]), np.array([]), np.array([])
         ann_time = self._get_float(annot[:, 0])
         ann_text = annot[:, 2]
+        ann_text_out = ["" for x in range(len(annot[:, 1]))]
         for i in np.arange(len(annot[:, 1])):
-            ann_text[i] = self._convert_string(ann_text[i])
+            ann_text_out[i] = self._convert_string(ann_text[i])
             if annot[i, 1] == '':
                 annot[i, 1] = '-1'
         ann_duration = self._get_float(annot[:, 1])
-        return ann_time/10000000, ann_duration, ann_text
+        return ann_time/10000000, ann_duration, np.array(ann_text_out)
 
     def _get_float(self, v):
         result = np.zeros(np.size(v))

diff --git a/pyedflib/edfwriter.py b/pyedflib/edfwriter.py
@@ -34,7 +34,7 @@ def du(x):
             return x
 else:
     def u(x):
-        return x
+        return x.decode("utf-8", "strict")
 
     def du(x):
         if isbytestr(x):

diff --git a/pyedflib/tests/test_edfwriter.py b/pyedflib/tests/test_edfwriter.py
@@ -115,9 +115,10 @@ def test_AnnotationWriting(self):
         f.writePhysicalSamples(data)
         f.writePhysicalSamples(data)
         f.writePhysicalSamples(data)
-        f.writeAnnotation(1.23, 0.2, u"annotation1")
-        f.writeAnnotation(0.25, -1, u"annotation2")
-        f.writeAnnotation(1.25, 0, u"annotation3")
+        f.writeAnnotation(1.23, 0.2, u"annotation1_ä")
+        f.writeAnnotation(0.25, -1, u"annotation2_ü")
+        f.writeAnnotation(1.25, 0, u"annotation3_ö")
+        f.writeAnnotation(1.30, -1, u"annotation4_ß")
         f.close()
         del f
 
@@ -127,13 +128,16 @@ def test_AnnotationWriting(self):
         del f
         np.testing.assert_almost_equal(ann_time[0], 1.23)
         np.testing.assert_almost_equal(ann_duration[0], 0.2)
-        np.testing.assert_equal(ann_text[0], b"annotation1")
+        np.testing.assert_equal(ann_text[0], "annotation1_..")
         np.testing.assert_almost_equal(ann_time[1], 0.25)
         np.testing.assert_almost_equal(ann_duration[1], -1)
-        np.testing.assert_equal(ann_text[1], b"annotation2")
+        np.testing.assert_equal(ann_text[1], "annotation2_..")
         np.testing.assert_almost_equal(ann_time[2], 1.25)
         np.testing.assert_almost_equal(ann_duration[2], 0)
-        np.testing.assert_equal(ann_text[2], b"annotation3")
+        np.testing.assert_equal(ann_text[2], "annotation3_..")
+        np.testing.assert_almost_equal(ann_time[3], 1.30)
+        np.testing.assert_almost_equal(ann_duration[3], -1)
+        np.testing.assert_equal(ann_text[3], "annotation4_..")
 
     def test_AnnotationWritingUTF8(self):
         channel_info = {'label': 'test_label', 'dimension': 'mV', 'sample_rate': 100,
@@ -160,13 +164,13 @@ def test_AnnotationWritingUTF8(self):
         del f
         np.testing.assert_almost_equal(ann_time[0], 1.23)
         np.testing.assert_almost_equal(ann_duration[0], 0.2)
-        np.testing.assert_equal(ann_text[0], b"Z..hne")
+        np.testing.assert_equal(ann_text[0], "Z..hne")
         np.testing.assert_almost_equal(ann_time[1], 0.25)
         np.testing.assert_almost_equal(ann_duration[1], -1)
-        np.testing.assert_equal(ann_text[1], b"Fu..")
+        np.testing.assert_equal(ann_text[1], "Fu..")
         np.testing.assert_almost_equal(ann_time[2], 1.25)
         np.testing.assert_almost_equal(ann_duration[2], 0)
-        np.testing.assert_equal(ann_text[2], b"abc")
+        np.testing.assert_equal(ann_text[2], "abc")
 
     def test_BytesChars(self):
         channel_info = {'label': b'test_label', 'dimension': b'mV', 'sample_rate': 100,
@@ -181,9 +185,9 @@ def test_BytesChars(self):
         f.writePhysicalSamples(data)
         f.writePhysicalSamples(data)
         f.writePhysicalSamples(data)
-        f.writeAnnotation(1.23, 0.2, u"Zähne")
-        f.writeAnnotation(0.25, -1, u"Fuß")
-        f.writeAnnotation(1.25, 0, u"abc")
+        f.writeAnnotation(1.23, 0.2, b'Zaehne')
+        f.writeAnnotation(0.25, -1, b'Fuss')
+        f.writeAnnotation(1.25, 0, b'abc')
         f.close()
         del f
 
@@ -193,13 +197,13 @@ def test_BytesChars(self):
         del f
         np.testing.assert_almost_equal(ann_time[0], 1.23)
         np.testing.assert_almost_equal(ann_duration[0], 0.2)
-        np.testing.assert_equal(ann_text[0], b"Z..hne")
+        np.testing.assert_equal(ann_text[0], "Zaehne")
         np.testing.assert_almost_equal(ann_time[1], 0.25)
         np.testing.assert_almost_equal(ann_duration[1], -1)
-        np.testing.assert_equal(ann_text[1], b"Fu..")
+        np.testing.assert_equal(ann_text[1], "Fuss")
         np.testing.assert_almost_equal(ann_time[2], 1.25)
         np.testing.assert_almost_equal(ann_duration[2], 0)
-        np.testing.assert_equal(ann_text[2], b"abc")
+        np.testing.assert_equal(ann_text[2], "abc")
 
 
 if __name__ == '__main__':