Add comprehensive type hints to all Python code

epitran/_epitran.py

@@ -85,7 +85,7 @@ def trans_list(self, word: str, normpunc: bool=False, ligatures: bool=False) ->
         """
         return self.ft.segs_safe(self.epi.transliterate(word, normpunc, ligatures))
 
-    def trans_delimiter(self, text: str, delimiter: str=str(' '), normpunc: bool=False, ligatures: bool=False):
+    def trans_delimiter(self, text: str, delimiter: str=str(' '), normpunc: bool=False, ligatures: bool=False) -> str:
         """Return IPA transliteration with a delimiter between segments
 
         :param text str: An orthographic text
@@ -98,7 +98,7 @@ def trans_delimiter(self, text: str, delimiter: str=str(' '), normpunc: bool=Fal
         return delimiter.join(self.trans_list(text, normpunc=normpunc,
                                               ligatures=ligatures))
 
-    def xsampa_list(self, word: str, normpunc: bool=False, ligaturize: bool=False):
+    def xsampa_list(self, word: str, normpunc: bool=False, ligaturize: bool=False) -> "list[str]":
         """Transliterates/transcribes a word as X-SAMPA
 
         :param word str: An orthographic word
@@ -111,7 +111,7 @@ def xsampa_list(self, word: str, normpunc: bool=False, ligaturize: bool=False):
                                                           ligaturize))
         return list(map(self.xsampa.ipa2xs, ipa_segs))
 
-    def word_to_tuples(self, word: str, normpunc: bool=False, _ligaturize: bool=False):
+    def word_to_tuples(self, word: str, normpunc: bool=False, _ligaturize: bool=False) -> "list[tuple[str, str, str, str, list[int]]]":
         """Given a word, returns a list of tuples corresponding to IPA segments. The "feature
         vectors" form a list consisting of (segment, vector) pairs.
         For IPA segments, segment is a substring of phonetic_form such that the

epitran/backoff.py

@@ -1,5 +1,6 @@
 # -*- coding: utf-8 -*-
 
+from typing import List, Optional, Union
 import regex as re
 from . import _epitran
 import panphon.featuretable
@@ -11,7 +12,7 @@
 class Backoff(object):
     """Implements rudimentary language ID and backoff."""
 
-    def __init__(self, lang_script_codes, cedict_file=None):
+    def __init__(self, lang_script_codes: List[str], cedict_file: Optional[str] = None) -> None:
         """Construct a Backoff object.
 
         Args:
@@ -28,7 +29,7 @@ def __init__(self, lang_script_codes, cedict_file=None):
         self.puncnorm = PuncNorm()
         self.dias = [StripDiacritics(c) for c in lang_script_codes]
 
-    def transliterate(self, token):
+    def transliterate(self, token: str) -> str:
         """Return IPA transliteration given by first acceptable mode.
         Args:
             token (unicode): orthographic text
@@ -60,7 +61,7 @@ def transliterate(self, token):
                     token = token[1:]
         return ''.join(tr_list)
 
-    def trans_list(self, token):
+    def trans_list(self, token: str) -> List[str]:
         """Transliterate/transcribe a word into list of IPA phonemes.
 
         Args:
@@ -71,7 +72,7 @@ def trans_list(self, token):
         """
         return self.ft.segs_safe(self.transliterate(token))
 
-    def xsampa_list(self, token):
+    def xsampa_list(self, token: str) -> Union[str, List[str]]:
         """Transcribe a word into a list of X-SAMPA phonemes.
 
         Args:

epitran/bin/connl2engipaspace.py

@@ -4,6 +4,7 @@
 import codecs
 import logging
 from collections import Counter
+from typing import List
 
 import csv
 
@@ -14,16 +15,16 @@
 logger = logging.getLogger('epitran')
 
 
-def normpunc(flite, s):
-    def norm(c):
+def normpunc(flite: epitran.flite.Flite, s: str) -> str:
+    def norm(c: str) -> str:
         if c in flite.puncnorm:
             return flite.puncnorm[c]
         else:
             return c
     return ''.join(map(norm, s))
 
 
-def add_record(flite, ft, orth):
+def add_record(flite: epitran.flite.Flite, ft: panphon.FeatureTable, orth: str) -> Counter[str]:
     space = Counter()
     orth = normpunc(flite, orth)
     trans = flite.transliterate(orth)
@@ -41,7 +42,7 @@ def add_record(flite, ft, orth):
     return space
 
 
-def add_file(flite, ft, fn):
+def add_file(flite: epitran.flite.Flite, ft: panphon.FeatureTable, fn: str) -> Counter[str]:
     space = Counter()
     with codecs.open(fn, 'r', 'utf-8') as f:
         for line in f:
@@ -53,15 +54,15 @@ def add_file(flite, ft, fn):
     return space
 
 
-def print_space(output, space):
+def print_space(output: str, space: Counter[str]) -> None:
     pairs = enumerate(sorted(filter(lambda x: x, space.keys())))
     with open(output, 'wb') as f:
         writer = csv.writer(f)
         for i, char in pairs:
             writer.writerow((i, char))
 
 
-def main(infiles, output):
+def main(infiles: List[str], output: str) -> None:
     flite = epitran.flite.Flite()
     ft = panphon.FeatureTable()
     space = Counter()

epitran/bin/connl2ipaspace.py

@@ -4,6 +4,7 @@
 import codecs
 import logging
 from collections import Counter
+from typing import List
 
 import epitran
 import panphon
@@ -12,16 +13,16 @@
 logger = logging.getLogger('epitran')
 
 
-def normpunc(epi, s):
-    def norm(c):
+def normpunc(epi: epitran.Epitran, s: str) -> str:
+    def norm(c: str) -> str:
         if c in epi.puncnorm:
             return epi.puncnorm[c]
         else:
             return c
     return ''.join(map(norm, s))
 
 
-def add_record_gen(epi, ft, orth):
+def add_record_gen(epi: epitran.Epitran, ft: panphon.FeatureTable, orth: str) -> Counter[str]:
     space = Counter()
     orth = normpunc(epi, orth)
     trans = epi.transliterate(orth)
@@ -36,7 +37,7 @@ def add_record_gen(epi, ft, orth):
     return space
 
 
-def add_file_gen(epi, ft, fn):
+def add_file_gen(epi: epitran.Epitran, ft: panphon.FeatureTable, fn: str) -> Counter[str]:
     space = Counter()
     with codecs.open(fn, 'r', 'utf-8') as f:
         for line in f:
@@ -48,7 +49,7 @@ def add_file_gen(epi, ft, fn):
     return space
 
 
-def add_file_op(epi, ft, fn):
+def add_file_op(epi: epitran.Epitran, ft: panphon.FeatureTable, fn: str) -> Counter[str]:
     space = Counter()
     with codecs.open(fn, 'r', 'utf-8') as f:
         for line in f:
@@ -71,15 +72,15 @@ def add_file_op(epi, ft, fn):
     return space
 
 
-def print_space(output, space):
+def print_space(output: str, space: Counter[str]) -> None:
     pairs = enumerate(sorted(filter(lambda x: x, space.keys())))
     with open(output, 'wb') as f:
         writer = csv.writer(f)
         for i, char in pairs:
             writer.writerow((i, char))
 
 
-def main(code, op, infiles, output):
+def main(code: str, op: bool, infiles: List[str], output: str) -> None:
     epi = epitran.Epitran(code)
     ft = panphon.FeatureTable()
     space = Counter()

epitran/bin/decompose.py

@@ -4,7 +4,7 @@
 import sys
 
 
-def main(fn):
+def main(fn: str) -> None:
     with open(fn, encoding='utf-8') as f:
         print(unicodedata.normalize('NFD', f.read()))
 

epitran/bin/detectcaps.py

@@ -5,7 +5,7 @@
 import fileinput
 
 
-def main():
+def main() -> None:
     for line in fileinput.input():
         line = line.decode('utf-8')
         token = line.strip()

epitran/bin/epitranscribe.py

@@ -7,7 +7,7 @@
 import argparse
 
 
-def main(code):
+def main(code: str) -> None:
     epi = epitran.Epitran(code)
     for line in sys.stdin:  # pointless
         line = line.decode('utf-8')

epitran/bin/isbijective.py

@@ -1,24 +1,25 @@
 #!/usr/bin/env pythoh
 
 import glob
+from typing import List, Tuple
 
 import csv
 
 
-def read_map(fn):
+def read_map(fn: str) -> List[Tuple[str, str]]:
     with open(fn, 'r', encoding='utf-8') as f:
         reader = csv.reader(f)
         next(reader)
         return [(a, b) for [a, b] in reader]
 
 
-def is_bijection(mapping):
+def is_bijection(mapping: List[Tuple[str, str]]) -> bool:
     a, b = zip(*mapping)
     distinct_a, distinct_b = set(a), set(b)
     return len(distinct_a) == len(mapping) and len(distinct_b) == len(mapping)
 
 
-def main(map_fns):
+def main(map_fns: List[str]) -> None:
     for fn in map_fns:
         mapping = read_map(fn)
         is_b = is_bijection(mapping)

epitran/bin/ltf2ipaspace.py

@@ -4,6 +4,7 @@
 import argparse
 import glob
 import os.path
+from typing import List, Set
 
 from lxml import etree
 import csv
@@ -12,13 +13,13 @@
 import panphon.featuretable
 
 
-def read_tokens(fn):
+def read_tokens(fn: str) -> List[str]:
     tree = etree.parse(fn)
     root = tree.getroot()
     return [tok.text for tok in root.findall('.//TOKEN')]
 
 
-def read_input(input_, langscript):
+def read_input(input_: List[List[str]], langscript: str) -> Set[str]:
     space = set()
     epi = epitran.Epitran(langscript)
     ft = panphon.featuretable.FeatureTable()
@@ -31,14 +32,14 @@ def read_input(input_, langscript):
     return space
 
 
-def write_output(output, space):
+def write_output(output: str, space: Set[str]) -> None:
     with open(output, 'wb') as f:
         writer = csv.writer(f)
         for n, ch in enumerate(sorted(list(space))):
             writer.writerow((n, ch))
 
 
-def main(langscript, input_, output):
+def main(langscript: str, input_: List[List[str]], output: str) -> None:
     space = read_input(input_, langscript)
     write_output(output, space)
 

epitran/bin/migraterules.py

@@ -4,23 +4,25 @@
 
 import glob
 import re
 import io
+from typing import List, Optional
 
 import csv
 
 
-def build_rule(fields):
+def build_rule(fields: List[str]) -> Optional[str]:
     try:
         a, b, X, Y = fields
         b = "0" if not b else b
         a = "0" if not a else a
         return '{} -> {} / {} _ {}'.format(a, b, X, Y)
     except ValueError:
         print('Malformed rule: {}'.format(','.join(fields)))
+        return None
 
 
-def main():
+def main() -> None:
     for csv in glob.glob('*.csv'):
         txt = re.match('[A-Za-z-]+', csv).group(0) + '.txt'
         with open(csv, 'r', encoding='utf-8') as f, open(txt, 'w', encoding='utf-8') as g:
             reader = csv.reader(f)

epitran/bin/space2punc.py

@@ -2,10 +2,11 @@
 
 import sys
 import unicodedata
+from typing import List
 import csv
 
 
-def main(fns, fnn):
+def main(fns: List[str], fnn: str) -> None:
     punc = set()
     for fn in fns:
         with open(fn, 'r', encoding='utf-8') as f:

epitran/bin/testvectorgen.py

@@ -3,11 +3,12 @@
 
 import argparse
 import codecs
+from typing import List
 
 import epitran.vector
 
 
-def main(code, space, infile):
+def main(code: str, space: List[str], infile: str) -> None:
     vec = epitran.vector.VectorsWithIPASpace(code, space)
     with codecs.open(infile, 'r', 'utf-8') as f:
         for line in f:

epitran/bin/vie-tones.py

@@ -16,7 +16,7 @@
 }
 
 
-def shuffle_tone(orth, phon):
+def shuffle_tone(orth: str, phon: str) -> str:
     orth = unicodedata.normalize('NFD', orth)
     if re.search('[aeiouơư]', orth):
         for tone in tones:
@@ -27,7 +27,7 @@ def shuffle_tone(orth, phon):
     return phon
 
 
-def main():
+def main() -> None:
     fnin = sys.argv[1]
     fnout = os.path.basename(fnin)
     with open(fnin) as fin, open(fnout, 'w') as fout: