Remove Python 2.7 leftovers

Functions `u()` and `du()` are not required anymore in Python 3. All variables are `str`.

doc/source/dev/building_extension.rst

@@ -34,7 +34,7 @@ repository or use the upstream repository to get the source code::
 
     git clone https://github.com/holgern/pyedflib.git pyedflib
 
-Install Microsoft Visual C++ Compiler for Python 2.7 from https://www.microsoft.com/en-us/download/details.aspx?id=44266
+Install Microsoft Visual C++ Compiler from https://visualstudio.microsoft.com/fr/downloads/
 
 Activate your Python virtual environment, go to the cloned source directory
 and type the following commands to build and install the package::

doc/source/dev/how_to_release.rst

@@ -41,7 +41,7 @@ Register all files with
 
 and upload with
 
-```twine upload  dist\filename_which_should_uploaded.whl```
+```twine upload dist\filename_which_should_uploaded.whl```
 
 Prepare for continued development
 ---------------------------------

doc/source/dev/index.rst

@@ -22,6 +22,6 @@ Something not working?
 ----------------------
 
 If these instructions are not clear or you need help setting up your
-development environment, go ahead and  open a ticket on GitHub_.
+development environment, go ahead and open a ticket on GitHub_.
 
-.. _GitHub: https://github.com/holgern/pyedflib
+.. _GitHub: https://github.com/holgern/pyedflib

doc/source/dev/preparing_windows_build_environment.rst

@@ -7,16 +7,16 @@ Preparing Windows build environment
 To start developing pyedflib code on Windows you will have to install
 a C compiler and prepare the build environment.
 
-Installing Microsoft Visual C++ Compiler for Python 2.7
--------------------------------------------------------
+Installing Microsoft Visual C++ Compiler
+----------------------------------------
 
-Downloading  Microsoft Visual C++ Compiler for Python 2.7 from https://www.microsoft.com/en-us/download/details.aspx?id=44266.
+Downloading Microsoft Visual C++ Compiler from https://visualstudio.microsoft.com/fr/downloads/.
 
 
 After installing the Compiler and before compiling the extension you have
 to configure some environment variables.
 
-For  build execute the ``util/setenv_win.bat`` script in the cmd
+For build execute the ``util/setenv_win.bat`` script in the cmd
 window:
 
   .. sourcecode:: bat
@@ -39,4 +39,4 @@ After completing these steps continue with
 .. _numpy: https://numpy.org/
 .. _Cython: https://cython.org/
 .. _Sphinx: https://www.sphinx-doc.org/
-.. _Microsoft Visual C++ Compiler for Python 2.7: https://www.microsoft.com/en-us/download/details.aspx?id=44266
+.. _Microsoft Visual C++ Compiler: https://visualstudio.microsoft.com/fr/downloads/

doc/source/resources.rst

@@ -11,7 +11,7 @@ The `GitHub repository`_ is now the main
 code repository.
 
 If you are using the Mercurial repository at Bitbucket, please switch
-to Git/GitHub and follow  for development updates.
+to Git/GitHub and follow for development updates.
 
 
 Questions and bug reports

pyedflib/_extensions/_pyedflib.pyx

@@ -202,7 +202,7 @@ cdef class CyEdfReader:
         """
         open(file_name, annotations_mode, check_file_size)
         """
-        file_name_str = file_name.encode('utf_8','strict')
+        file_name_str = file_name.encode(errors='strict')
         result = c_edf.edfopen_file_readonly(file_name_str, &self.hdr, annotations_mode, check_file_size)
 
         self.file_name = file_name
@@ -511,7 +511,7 @@ def get_handle(file_number):
     return c_edf.edflib_get_handle(file_number)
 
 def is_file_used(path):
-    path_byte = path.encode('utf_8','strict')
+    path_byte = path.encode(errors='strict')
     return c_edf.edflib_is_file_used(path_byte)
 
 # so you can use the same name if defining a python only function
@@ -521,7 +521,7 @@ def set_physical_maximum(handle, edfsignal, phys_max):
 def open_file_writeonly(path, filetype, number_of_signals):
     """int edfopen_file_writeonly(char *path, int filetype, int number_of_signals)"""
 
-    py_byte_string  = path.encode('utf_8','strict')
+    py_byte_string  = path.encode(errors='strict')
     cdef char* path_str = py_byte_string
     return c_edf.edfopen_file_writeonly(path_str, filetype, number_of_signals)
 

pyedflib/edfreader.py

@@ -44,32 +44,32 @@ def _debug_parse_header(filename: str, printout=True) -> None:  # pragma: no cov
     header = OrderedDict()
     with open(filename, "rb") as f:
         f.seek(0)
-        header["version"] = f.read(8).decode()
-        header["patient_id"] = f.read(80).decode().strip()
-        header["recording_id"] = f.read(80).decode().strip()
-        header["startdate"] = f.read(8).decode()
-        header["starttime"] = f.read(8).decode()
-        header["header_n_bytes"] = f.read(8).decode()
-        header["reserved"] = f.read(44).decode().strip()
-        header["n_records"] = f.read(8).decode()
-        header["record_duration"] = f.read(8).decode()
-        header["n_signals"] = f.read(4).decode()
+        header["version"] = f.read(8).decode("ascii")
+        header["patient_id"] = f.read(80).decode("ascii").strip()
+        header["recording_id"] = f.read(80).decode("ascii").strip()
+        header["startdate"] = f.read(8).decode("ascii")
+        header["starttime"] = f.read(8).decode("ascii")
+        header["header_n_bytes"] = f.read(8).decode("ascii")
+        header["reserved"] = f.read(44).decode("ascii").strip()
+        header["n_records"] = f.read(8).decode("ascii")
+        header["record_duration"] = f.read(8).decode("ascii")
+        header["n_signals"] = f.read(4).decode("ascii")
 
         if printout:
             print("\n##### Header")
             print(json.dumps(header, indent=2))
 
         nsigs = int(header["n_signals"])
-        label = [f.read(16).decode() for i in range(nsigs)]
-        transducer = [f.read(80).decode().strip() for i in range(nsigs)]
-        dimension = [f.read(8).decode().strip() for i in range(nsigs)]
-        pmin = [f.read(8).decode() for i in range(nsigs)]
-        pmax = [f.read(8).decode() for i in range(nsigs)]
-        dmin = [f.read(8).decode() for i in range(nsigs)]
-        dmax = [f.read(8).decode() for i in range(nsigs)]
-        prefilter = [f.read(80).decode().strip() for i in range(nsigs)]
-        n_samples = [f.read(8).decode() for i in range(nsigs)]
-        reserved = [f.read(32).decode() for i in range(nsigs)]
+        label = [f.read(16).decode("ascii") for i in range(nsigs)]
+        transducer = [f.read(80).decode("ascii").strip() for i in range(nsigs)]
+        dimension = [f.read(8).decode("ascii").strip() for i in range(nsigs)]
+        pmin = [f.read(8).decode("ascii") for i in range(nsigs)]
+        pmax = [f.read(8).decode("ascii") for i in range(nsigs)]
+        dmin = [f.read(8).decode("ascii") for i in range(nsigs)]
+        dmax = [f.read(8).decode("ascii") for i in range(nsigs)]
+        prefilter = [f.read(80).decode("ascii").strip() for i in range(nsigs)]
+        n_samples = [f.read(8).decode("ascii") for i in range(nsigs)]
+        reserved = [f.read(32).decode("ascii") for i in range(nsigs)]
     _ = zip(
         label,
         transducer,
@@ -168,7 +168,7 @@ def _get_float(self, v: np.ndarray) -> np.ndarray:
 
     def _convert_string(self, s: Union[bytes, str]) -> str:
         if isinstance(s, bytes):
-            return s.decode("latin")
+            return s.decode("latin_1")
         elif isinstance(s, str):
             return s
         else:

pyedflib/edfwriter.py

@@ -146,17 +146,6 @@ def check_signal_header_correct(channels: List[Dict[str, Union[str, None, float]
                                                                       str(ch['physical_max'])[:8]))
 
 
-def u(x: bytes) -> str:
-    return x.decode("utf_8", "strict")
-
-
-def du(x: Union[str, bytes]) -> bytes:
-    if isinstance(x, bytes):
-        return x
-    else:
-        return x.encode("utf_8")
-
-
 def isstr(s: Any) -> bool:
     warnings.warn("Function 'isstr' is deprecated.", DeprecationWarning, stacklevel=2)
     return isinstance(s, str)
@@ -389,13 +378,13 @@ def update_header(self) -> None:
                 f'{sample_freqs=} contains non int/float'
             self.record_duration = _calculate_record_duration(sample_freqs)
 
-        set_technician(self.handle, du(self.technician))
-        set_recording_additional(self.handle, du(self.recording_additional))
-        set_patientname(self.handle, du(self.patient_name))
-        set_patientcode(self.handle, du(self.patient_code))
-        set_patient_additional(self.handle, du(self.patient_additional))
-        set_equipment(self.handle, du(self.equipment))
-        set_admincode(self.handle, du(self.admincode))
+        set_technician(self.handle, self.technician.encode('ascii'))
+        set_recording_additional(self.handle, self.recording_additional.encode('ascii'))
+        set_patientname(self.handle, self.patient_name.encode('ascii'))
+        set_patientcode(self.handle, self.patient_code.encode('ascii'))
+        set_patient_additional(self.handle, self.patient_additional.encode('ascii'))
+        set_equipment(self.handle, self.equipment.encode('ascii'))
+        set_admincode(self.handle, self.admincode.encode('ascii'))
         set_sex(self.handle, sex2int(self.sex))
 
         set_datarecord_duration(self.handle, self.record_duration)
@@ -420,10 +409,10 @@ def update_header(self) -> None:
             set_physical_minimum(self.handle, i, self.channels[i]['physical_min'])
             set_digital_maximum(self.handle, i, self.channels[i]['digital_max'])
             set_digital_minimum(self.handle, i, self.channels[i]['digital_min'])
-            set_label(self.handle, i, du(self.channels[i]['label']))
-            set_physical_dimension(self.handle, i, du(self.channels[i]['dimension']))
-            set_transducer(self.handle, i, du(self.channels[i]['transducer']))
-            set_prefilter(self.handle, i, du(self.channels[i]['prefilter']))
+            set_label(self.handle, i, self.channels[i]['label'].encode('ascii'))
+            set_physical_dimension(self.handle, i, self.channels[i]['dimension'].encode('ascii'))
+            set_transducer(self.handle, i, self.channels[i]['transducer'].encode('ascii'))
+            set_prefilter(self.handle, i, self.channels[i]['prefilter'].encode('ascii'))
 
 
 
@@ -1017,16 +1006,14 @@ def writeAnnotation(self, onset_in_seconds: Union[int, float], duration_in_secon
 
         if str_format == 'utf_8':
             if duration_in_seconds >= 0:
-                return write_annotation_utf8(self.handle, np.round(onset_in_seconds*10000).astype(np.int64), np.round(duration_in_seconds*10000).astype(int), du(description))
+                return write_annotation_utf8(self.handle, np.round(onset_in_seconds*10000).astype(np.int64), np.round(duration_in_seconds*10000).astype(int), description.encode('utf_8'))
             else:
-                return write_annotation_utf8(self.handle, np.round(onset_in_seconds*10000).astype(np.int64), -1, du(description))
+                return write_annotation_utf8(self.handle, np.round(onset_in_seconds*10000).astype(np.int64), -1, description.encode('utf_8'))
         else:
             if duration_in_seconds >= 0:
-                # FIX: description must be bytes. string will fail in u function
-                return write_annotation_latin1(self.handle, np.round(onset_in_seconds*10000).astype(np.int64), np.round(duration_in_seconds*10000).astype(int), u(description).encode('latin1'))  # type: ignore
+                return write_annotation_latin1(self.handle, np.round(onset_in_seconds*10000).astype(np.int64), np.round(duration_in_seconds*10000).astype(int), description.encode('latin1'))  # type: ignore
             else:
-                # FIX: description must be bytes. string will fail in u function
-                return write_annotation_latin1(self.handle, np.round(onset_in_seconds*10000).astype(np.int64), -1, u(description).encode('latin1'))  # type: ignore
+                return write_annotation_latin1(self.handle, np.round(onset_in_seconds*10000).astype(np.int64), -1, description.encode('latin1'))  # type: ignore
 
     def close(self) -> None:
         """

pyedflib/highlevel.py

@@ -753,7 +753,7 @@ def drop_channels(
             if isinstance(ch,str):
                 ch_idx = ch_names.index(ch.lower())
                 to_keep[i] = ch_idx
-        load_channels = list(to_keep) # copy list compatible with py2.7
+        load_channels = to_keep.copy()
     elif to_drop is not None:
         for i,ch in enumerate(to_drop):
             if isinstance(ch,str):

pyedflib/tests/test_edfreader.py

@@ -374,19 +374,19 @@ def test_read_incorrect_file(self):
     def test_EdfReader_Legacy_Header_Info(self):
         expected_header = {
             # Legacy header fields
-            'patient': b'Legacy patient description',
-            'recording': b'Legacy recording description',
+            'patient': 'Legacy patient description',
+            'recording': 'Legacy recording description',
             # All the rest must be empty
-            'technician': b'',
-            'recording_additional': b'',
-            'patientname': b'',
-            'patient_additional': b'',
-            'patientcode': b'',
-            'equipment': b'',
-            'admincode': b'',
-            'sex': b'',
-            'gender': b'',  # deprecated
-            'birthdate': b''
+            'technician': '',
+            'recording_additional': '',
+            'patientname': '',
+            'patient_additional': '',
+            'patientcode': '',
+            'equipment': '',
+            'admincode': '',
+            'sex': '',
+            'gender': '',  # deprecated
+            'birthdate': ''
         }
 
         with pyedflib.EdfReader(self.edf_legacy) as f:

pyedflib/tests/test_edfwriter.py

@@ -715,10 +715,10 @@ def test_AnnotationWritingUTF8(self):
         np.testing.assert_equal(ann_text[2], "abc")
 
     def test_BytesChars(self):
-        channel_info = {'label': b'test_label', 'dimension': b'mV', 'sample_frequency': 100,
+        channel_info = {'label': 'test_label', 'dimension': 'mV', 'sample_frequency': 100,
                         'physical_max': 1.0, 'physical_min': -1.0,
                         'digital_max': 8388607, 'digital_min': -8388608,
-                        'prefilter': b'      ', 'transducer': b'trans1'}
+                        'prefilter': '      ', 'transducer': 'trans1'}
         f = pyedflib.EdfWriter(self.bdfplus_data_file, 1,
                                 file_type=pyedflib.FILETYPE_BDFPLUS)
         f.setSignalHeader(0,channel_info)
@@ -727,9 +727,9 @@ def test_BytesChars(self):
         f.writePhysicalSamples(data)
         f.writePhysicalSamples(data)
         f.writePhysicalSamples(data)
-        f.writeAnnotation(1.23, 0.2, b'Zaehne')
-        f.writeAnnotation(0.25, -1, b'Fuss')
-        f.writeAnnotation(1.25, 0, b'abc')
+        f.writeAnnotation(1.23, 0.2, 'Zaehne')
+        f.writeAnnotation(0.25, -1, 'Fuss')
+        f.writeAnnotation(1.25, 0, 'abc')
 
         del f
 

setup.py

@@ -66,12 +66,8 @@ def get_numpy_include():
         # versions.
         # setuptools forgets to unset numpy's setup flag and we get a crippled
         # version of it unless we do it ourselves.
-        try:
-            import __builtin__  # py2
-            __builtin__.__NUMPY_SETUP__ = False
-        except:
-            import builtins  # py3
-            builtins.__NUMPY_SETUP__ = False
+        import builtins
+        builtins.__NUMPY_SETUP__ = False
         import numpy as np
     except ImportError as e:
         try:

util/authors.py

@@ -48,7 +48,7 @@ def main():
     authors = collections.Counter()
 
     def analyze_line(line, names, disp=False):
-        line = line.strip().decode('utf-8')
+        line = line.strip().decode()
 
         # Check the commit author name
         m = re.match(u'^@@@([^@]*)@@@', line)
@@ -58,7 +58,7 @@ def analyze_line(line, names, disp=False):
             name = NAME_MAP.get(name, name)
             if disp:
                 if name not in names:
-                    stdout_b.write(("    - Author: %s\n" % name).encode('utf-8'))
+                    stdout_b.write(("    - Author: %s\n" % name).encode())
             names.update((name,))
 
         # Look for "thanks to" messages in the commit log
@@ -67,13 +67,13 @@ def analyze_line(line, names, disp=False):
             name = m.group(2)
             if name not in (u'this',):
                 if disp:
-                    stdout_b.write("    - Log   : %s\n" % line.strip().encode('utf-8'))
+                    stdout_b.write("    - Log   : %s\n" % line.strip().encode())
                 name = NAME_MAP.get(name, name)
                 names.update((name,))
 
             line = line[m.end():].strip()
             line = re.sub(r'^(and|, and|, ) ', u'Thanks to ', line)
-            analyze_line(line.encode('utf-8'), names)
+            analyze_line(line.encode(), names)
 
     # Find all authors before the named range
     for line in git.pipe('log', '--pretty=@@@%an@@@%n@@@%cn@@@%n%b',
@@ -108,9 +108,9 @@ def name_key(fullname):
         n_authors = list(new_authors)
         n_authors.sort(key=name_key)
         # Print some empty lines to separate
-        stdout_b.write(("\n\n").encode('utf-8'))
+        stdout_b.write(("\n\n").encode())
         for author in n_authors:
-            stdout_b.write(("- %s\n" % author).encode('utf-8'))
+            stdout_b.write(("- %s\n" % author).encode())
         # return for early exit so we only print new authors
         return
 
@@ -133,19 +133,19 @@ def name_key(fullname):
         author_clean = author.strip('@')
 
         if author in all_authors:
-            stdout_b.write((f"* {author_clean} ({count})\n").encode('utf-8'))
+            stdout_b.write((f"* {author_clean} ({count})\n").encode())
         else:
-            stdout_b.write((f"* {author_clean} ({count}) +\n").encode('utf-8'))
+            stdout_b.write((f"* {author_clean} ({count}) +\n").encode())
 
     stdout_b.write(("""
 A total of %(count)d people contributed to this release.
 People with a "+" by their names contributed a patch for the first time.
 This list of names is automatically generated, and may not be fully complete.
 
-""" % dict(count=len(authors))).encode('utf-8'))
+""" % dict(count=len(authors))).encode())
 
     stdout_b.write(("\nNOTE: Check this list manually! It is automatically generated "
-                    "and some names\n      may be missing.\n").encode('utf-8'))
+                    "and some names\n      may be missing.\n").encode())
 
 
 def load_name_map(filename):