Modernize Python 2 compatibility code to Python 3.10+

README.md

@@ -38,13 +38,13 @@ It is now possible to use the Epitran class for English, Mandarin Chinese (Simpl
 
 The most useful public method of the Epitran class is `transliterate`:
 
-Epitran.**transliterate**(text, normpunc=False, ligatures=False). Convert `text` (in Unicode-encoded orthography of the language specified in the constructor) to IPA, which is returned. `normpunc` enables punctuation normalization and `ligatures` enables non-standard IPA ligatures like "ʤ" and "ʨ". Usage is illustrated below (Python 2):
+Epitran.**transliterate**(text, normpunc=False, ligatures=False). Convert `text` (in Unicode-encoded orthography of the language specified in the constructor) to IPA, which is returned. `normpunc` enables punctuation normalization and `ligatures` enables non-standard IPA ligatures like "ʤ" and "ʨ". Usage is illustrated below:
 
 
 ```python
->>> epi.transliterate(u'Düğün')
-u'dy\u0270yn'
->>> print(epi.transliterate(u'Düğün'))
+>>> epi.transliterate('Düğün')
+'dy\u0270yn'
+>>> print(epi.transliterate('Düğün'))
 dyɰyn
 ```
 
@@ -71,14 +71,14 @@ The codes for `character_category` are from the initial characters of the two ch
 )
 ```
 
-Here is an example of an interaction with ```word_to_tuples``` (Python 2):
+Here is an example of an interaction with ```word_to_tuples```:
 
 
 ```python
 >>> import epitran
 >>> epi = epitran.Epitran('tur-Latn')
->>> epi.word_to_tuples(u'Düğün')
-[(u'L', 1, u'D', u'd', [(u'd', [-1, -1, 1, -1, -1, -1, -1, -1, 1, -1, -1, 1, 1, -1, -1, -1, -1, -1, -1, 0, -1])]), (u'L', 0, u'u\u0308', u'y', [(u'y', [1, 1, -1, 1, -1, -1, -1, 0, 1, -1, -1, -1, -1, -1, 1, 1, -1, -1, 1, 1, -1])]), (u'L', 0, u'g\u0306', u'\u0270', [(u'\u0270', [-1, 1, -1, 1, 0, -1, -1, 0, 1, -1, -1, 0, -1, 0, -1, 1, -1, 0, -1, 1, -1])]), (u'L', 0, u'u\u0308', u'y', [(u'y', [1, 1, -1, 1, -1, -1, -1, 0, 1, -1, -1, -1, -1, -1, 1, 1, -1, -1, 1, 1, -1])]), (u'L', 0, u'n', u'n', [(u'n', [-1, 1, 1, -1, -1, -1, 1, -1, 1, -1, -1, 1, 1, -1, -1, -1, -1, -1, -1, 0, -1])])]
+>>> epi.word_to_tuples('Düğün')
+[('L', 1, 'D', 'd', [('d', [-1, -1, 1, -1, -1, -1, -1, -1, 1, -1, -1, 1, 1, -1, -1, -1, -1, -1, -1, 0, -1])]), ('L', 0, 'ü', 'y', [('y', [1, 1, -1, 1, -1, -1, -1, 0, 1, -1, -1, -1, -1, -1, 1, 1, -1, -1, 1, 1, -1])]), ('L', 0, 'ğ', 'ɰ', [('ɰ', [-1, 1, -1, 1, 0, -1, -1, 0, 1, -1, -1, 0, -1, 0, -1, 1, -1, 0, -1, 1, -1])]), ('L', 0, 'ü', 'y', [('y', [1, 1, -1, 1, -1, -1, -1, 0, 1, -1, -1, -1, -1, -1, 1, 1, -1, -1, 1, 1, -1])]), ('L', 0, 'n', 'n', [('n', [-1, 1, 1, -1, -1, -1, 1, -1, 1, -1, -1, 1, 1, -1, -1, -1, -1, -1, -1, 0, -1])])]
 ```
 
 ### The Backoff class
@@ -87,7 +87,7 @@ Sometimes, when parsing text in more than one script, it is useful to employ a g
 
 Backoff(lang_script_codes, cedict_file=None)
 
-Note that the Backoff class does not currently support parameterized preprocessor and postprocessor application and does not support non-standard ligatures. It also does not support punctuation normalization. `lang_script_codes` is a list of codes like `eng-Latn` or `hin-Deva`. For example, if one was transcribing a Hindi text with many English loanwords and some stray characters of Simplified Chinese, one might use the following code (Python 3):
+Note that the Backoff class does not currently support parameterized preprocessor and postprocessor application and does not support non-standard ligatures. It also does not support punctuation normalization. `lang_script_codes` is a list of codes like `eng-Latn` or `hin-Deva`. For example, if one was transcribing a Hindi text with many English loanwords and some stray characters of Simplified Chinese, one might use the following code:
 
 ```python
 from epitran.backoff import Backoff
@@ -109,7 +109,7 @@ The Backoff class has the following public methods:
 * **xsampa_list**: returns a list of X-SAMPA (ASCII) strings, each of which is
   phoneme
 
-Consider the following example (Python 3):
+Consider the following example:
 
 ```python
 >>> backoff.transliterate('हिन्दी')
@@ -157,13 +157,13 @@ Its principle method is ```word_to_segs```:
 VectorWithIPASpace.**word_to_segs**(word, normpunc=False). `word` is a Unicode string. If the keyword argument *normpunc* is set to True, punctuation discovered in `word` is normalized to ASCII equivalents.
 
 
-A typical interaction with the ```VectorsWithIPASpace``` object via the ```word_to_segs``` method is illustrated here (Python 2):
+A typical interaction with the ```VectorsWithIPASpace``` object via the ```word_to_segs``` method is illustrated here:
 
 ```python
 >>> import epitran.vector
 >>> vwis = epitran.vector.VectorsWithIPASpace('uzb-Latn', ['uzb-Latn'])
->>> vwis.word_to_segs(u'darë')
-[(u'L', 0, u'd', u'd\u032a', u'40', [-1, -1, 1, -1, -1, -1, -1, -1, 1, -1, -1, 1, 1, 1, -1, -1, -1, -1, -1, 0, -1]), (u'L', 0, u'a', u'a', u'37', [1, 1, -1, 1, -1, -1, -1, 0, 1, -1, -1, -1, -1, -1, -1, -1, 1, 1, -1, 1, -1]), (u'L', 0, u'r', u'r', u'54', [-1, 1, 1, 1, 0, -1, -1, -1, 1, -1, -1, 1, 1, -1, -1, 0, 0, 0, -1, 0, -1]), (u'L', 0, u'e\u0308', u'ja', u'46', [-1, 1, -1, 1, -1, -1, -1, 0, 1, -1, -1, -1, -1, 0, -1, 1, -1, -1, -1, 0, -1]), (u'L', 0, u'e\u0308', u'ja', u'37', [1, 1, -1, 1, -1, -1, -1, 0, 1, -1, -1, -1, -1, -1, -1, -1, 1, 1, -1, 1, -1])]
+>>> vwis.word_to_segs('darë')
+[('L', 0, 'd', 'd̪', '40', [-1, -1, 1, -1, -1, -1, -1, -1, 1, -1, -1, 1, 1, 1, -1, -1, -1, -1, -1, 0, -1]), ('L', 0, 'a', 'a', '37', [1, 1, -1, 1, -1, -1, -1, 0, 1, -1, -1, -1, -1, -1, -1, -1, 1, 1, -1, 1, -1]), ('L', 0, 'r', 'r', '54', [-1, 1, 1, 1, 0, -1, -1, -1, 1, -1, -1, 1, 1, -1, -1, 0, 0, 0, -1, 0, -1]), ('L', 0, 'ë', 'ja', '46', [-1, 1, -1, 1, -1, -1, -1, 0, 1, -1, -1, -1, -1, 0, -1, 1, -1, -1, -1, 0, -1]), ('L', 0, 'ë', 'ja', '37', [1, 1, -1, 1, -1, -1, -1, 0, 1, -1, -1, -1, -1, -1, -1, -1, 1, 1, -1, 1, -1])]
 ```
 
 
@@ -379,7 +379,7 @@ Note that major languages, including **French**, are missing from this table due
 
 ## Installation of Flite (for English G2P)
 
-For use with most languages, Epitran requires no special installation steps. It can be installed as an ordinarary python package, either with `pip` or by running `python setup.py install` in the root of the source directory. However, English G2P in Epitran relies on CMU Flite, a speech synthesis package by Alan Black and other speech researchers at Carnegie Mellon University. For the current version of Epitran, you should follow the installation instructions for `lex_lookup`, which is used as the default G2P interface for Epitran.
+For use with most languages, Epitran requires no special installation steps. **Epitran requires Python 3.10 or higher.** It can be installed as an ordinary python package, either with `pip` or by running `python setup.py install` in the root of the source directory. However, English G2P in Epitran relies on CMU Flite, a speech synthesis package by Alan Black and other speech researchers at Carnegie Mellon University. For the current version of Epitran, you should follow the installation instructions for `lex_lookup`, which is used as the default G2P interface for Epitran.
 
 ### `t2p`
 
@@ -523,7 +523,7 @@ c -> s / _ (i|e)
 ```
 
 
-For a complete guide to `regex` regular expressions, see the documentation for [`re`](https://docs.python.org/2/library/re.html) and for [`regex`](https://pypi.python.org/pypi/regex), specifically.
+For a complete guide to `regex` regular expressions, see the documentation for [`re`](https://docs.python.org/3/library/re.html) and for [`regex`](https://pypi.org/project/regex/), specifically.
 
 Fragments of regular expressions can be assigned to symbols and reused throughout a file. For example, symbol for the disjunction of vowels in a language can be used in a rule that changes /u/ into /w/ before vowels:
 

epitran/backoff.py

@@ -1,6 +1,4 @@
 # -*- coding: utf-8 -*-
-from __future__ import (print_function, absolute_import,
-                        unicode_literals)
 
 import regex as re
 from . import _epitran
@@ -35,7 +33,7 @@ def transliterate(self, token):
         Args:
             token (unicode): orthographic text
         Returns:
-            unicode: transliteration as Unicode IPA string
+            str: transliteration as Unicode IPA string
         """
         tr_list = []
         while token:

epitran/bin/connl2engipaspace.py

@@ -5,7 +5,7 @@
 import logging
 from collections import Counter
 
-import unicodecsv as csv
+import csv
 
 import epitran
 import epitran.flite
@@ -56,7 +56,7 @@ def add_file(flite, ft, fn):
 def print_space(output, space):
     pairs = enumerate(sorted(filter(lambda x: x, space.keys())))
     with open(output, 'wb') as f:
-        writer = csv.writer(f, encoding='utf-8')
+        writer = csv.writer(f)
         for i, char in pairs:
             writer.writerow((i, char))
 

epitran/bin/connl2ipaspace.py

@@ -7,7 +7,7 @@
 
 import epitran
 import panphon
-import unicodecsv as csv
+import csv
 
 logger = logging.getLogger('epitran')
 
@@ -74,7 +74,7 @@ def add_file_op(epi, ft, fn):
 def print_space(output, space):
     pairs = enumerate(sorted(filter(lambda x: x, space.keys())))
     with open(output, 'wb') as f:
-        writer = csv.writer(f, encoding='utf-8')
+        writer = csv.writer(f)
         for i, char in pairs:
             writer.writerow((i, char))
 

epitran/bin/detectcaps.py

@@ -1,6 +1,5 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
-from __future__ import print_function
 
 import unicodedata
 import fileinput

epitran/bin/isbijective.py

@@ -1,14 +1,13 @@
 #!/usr/bin/env pythoh
-from __future__ import print_function
 
 import glob
 
-import unicodecsv as csv
+import csv
 
 
 def read_map(fn):
-    with open(fn, 'rb') as f:
-        reader = csv.reader(f, encoding='utf-8')
+    with open(fn, 'r', encoding='utf-8') as f:
+        reader = csv.reader(f)
         next(reader)
         return [(a, b) for [a, b] in reader]
 

epitran/bin/ltf2ipaspace.py

@@ -1,13 +1,12 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
-from __future__ import print_function
 
 import argparse
 import glob
 import os.path
 
 from lxml import etree
-import unicodecsv as csv
+import csv
 
 import epitran
 import panphon.featuretable
@@ -34,7 +33,7 @@ def read_input(input_, langscript):
 
 def write_output(output, space):
     with open(output, 'wb') as f:
-        writer = csv.writer(f, encoding='utf-8')
+        writer = csv.writer(f)
         for n, ch in enumerate(sorted(list(space))):
             writer.writerow((n, ch))
 

epitran/bin/migraterules.py

@@ -1,13 +1,12 @@
 #!/usr/bin/env Python
 # -*- coding: utf-8 -*-
 
-from __future__ import (print_function, unicode_literals, absolute_import)
 
 import glob
 import re
 import io
 
-import unicodecsv
+import csv
 
 
 def build_rule(fields):
@@ -21,10 +20,10 @@
 
 
 def main():
     for csv in glob.glob('*.csv'):
         txt = re.match('[A-Za-z-]+', csv).group(0) + '.txt'
-        with open(csv, 'rb') as f, io.open(txt, 'w', encoding='utf-8') as g:
-            reader = unicodecsv.reader(f, encoding='utf-8')
+        with open(csv, 'r', encoding='utf-8') as f, open(txt, 'w', encoding='utf-8') as g:
+            reader = csv.reader(f)
             next(reader)
             for fields in reader:
                 if re.match('\s*%', fields[0]):

epitran/bin/reromanize.py

@@ -1,6 +1,5 @@
-#!/usr/bin/env python2
+#!/usr/bin/env python3
 
-from __future__ import print_function
 
 import epitran.reromanize
 import argparse

epitran/bin/space2punc.py

@@ -2,19 +2,19 @@
 
 import sys
 import unicodedata
-import unicodecsv as csv
+import csv
 
 
 def main(fns, fnn):
     punc = set()
     for fn in fns:
-        with open(fn, 'rb') as f:
-            reader = csv.reader(f, encoding='utf-8')
+        with open(fn, 'r', encoding='utf-8') as f:
+            reader = csv.reader(f)
             for _, s in reader:
                 if len(s) == 1 and unicodedata.category(s)[0] == u'P':
                     punc.add(s)
     with open(fnn, 'wb') as f:
-        writer = csv.writer(f, encoding='utf-8')
+        writer = csv.writer(f)
         for mark in sorted(list(punc)):
             writer.writerow([mark])
 

epitran/bin/testvectorgen.py

@@ -1,6 +1,5 @@
 #!/usr/bin/env python
 
-from __future__ import print_function
 
 import argparse
 import codecs

epitran/bin/uigtransliterate.py

@@ -1,5 +1,4 @@
 #!/usr/bin/env python
-from __future__ import print_function
 
 import fileinput
 import epitran

epitran/cedict.py

@@ -1,6 +1,4 @@
 # -*- coding: utf-8 -*-
-from __future__ import (absolute_import, division, print_function,
-                        unicode_literals)
 
 import codecs
 

epitran/data/map/count_phones.py

@@ -2,14 +2,14 @@
 
 import epitran.xsampa
 import panphon
-import unicodecsv as csv
+import csv
 
 
 def main(fn):
     ft = panphon.FeatureTable()
     xs = epitran.xsampa.XSampa()
-    with open(fn, 'rb') as f:
-        reader = csv.reader(f, encoding='utf-8')
+    with open(fn, 'r', encoding='utf-8') as f:
+        reader = csv.reader(f)
         next(reader)
         phones = set()
         for orth, phon in reader:

epitran/epihan.py

@@ -1,5 +1,4 @@
 # -*- utf-8 -*-
-from __future__ import print_function, unicode_literals, division, absolute_import
 
 import os.path
 
@@ -56,10 +55,10 @@ def normalize_punc(self, text):
         """Normalize punctutation in a string
 
         Args:
-            text (unicode): an orthographic string
+            text (str): an orthographic string
 
         Return:
-            unicode: an orthographic string with punctation normalized to
+            str: an orthographic string with punctation normalized to
                      Western equivalents
         """
         for a, b in self.punc: