def _apply_meta_attach(meta, ctx):
action = ctx.new_action()
begin = meta.startswith(META_ATTACH_FLAG)
end = meta.endswith(META_ATTACH_FLAG)
if begin:
meta = meta[len(META_ATTACH_FLAG):]
action.prev_attach = True
if end:
meta = meta[:-len(META_ATTACH_FLAG)]
action.next_attach = True
last_word = ctx.last_action.word or ''
if not meta:
# We use an empty connection to indicate a "break" in the
# application of orthography rules. This allows the
# stenographer to tell Plover not to auto-correct a word.
action.orthography = False
elif (last_word and not meta.isspace() and ctx.last_action.orthography
and begin and (not end or _has_word_boundary(meta))):
new_word = add_suffix(last_word, meta)
common_len = len(commonprefix([last_word, new_word]))
replaced = last_word[common_len:]
action.prev_replace = ctx.last_text(len(replaced))
assert replaced.lower() == action.prev_replace.lower()
last_word = last_word[:common_len]
meta = new_word[common_len:]
action.text = meta
if action.prev_attach:
action.word = _rightmost_word(last_word + meta)
return action
def _apply_meta_attach(meta, ctx):
action = ctx.new_action()
begin = meta.startswith(META_ATTACH_FLAG)
end = meta.endswith(META_ATTACH_FLAG)
if begin:
meta = meta[len(META_ATTACH_FLAG):]
action.prev_attach = True
if end:
meta = meta[:-len(META_ATTACH_FLAG)]
action.next_attach = True
last_word = ctx.last_action.word or ''
if not meta:
# We use an empty connection to indicate a "break" in the
# application of orthography rules. This allows the
# stenographer to tell Plover not to auto-correct a word.
action.orthography = False
elif (
last_word and
not meta.isspace() and
ctx.last_action.orthography and
begin and (not end or _has_word_boundary(meta))
):
new_word = add_suffix(last_word, meta)
common_len = len(commonprefix([last_word, new_word]))
replaced = last_word[common_len:]
action.prev_replace = ctx.last_text(len(replaced))
assert replaced.lower() == action.prev_replace.lower()
last_word = last_word[:common_len]
meta = new_word[common_len:]
action.text = meta
if action.prev_attach:
action.word = _rightmost_word(last_word + meta)
return action
def test_runner(test, *args, **kwargs):
fn(test, *args, **kwargs)
result = add_suffix(test.word, test.suffix)
report = orthographic_rules_report(test.word, test.suffix)
msg = '%s\n%s' % (
test.format(location=False, result=result),
'\n'.join('• %s: %s' % (k, pformat(v)) for k, v in report.items()),
)
assert result == test.expected, msg
def _atom_to_action_spaces_after(atom, last_action):
"""Convert an atom into an action.
Arguments:
atom -- A string holding an atom. An atom is an irreducible string that is
either entirely a single meta command or entirely text containing no meta
commands.
last_action -- The context in which the new action takes place.
Returns: An action for the atom.
"""
action = _Action(space_char=last_action.space_char, case=last_action.case)
last_word = last_action.word
last_glue = last_action.glue
last_attach = last_action.attach
last_capitalize = last_action.capitalize
last_lower = last_action.lower
last_upper = last_action.upper
last_upper_carry = last_action.upper_carry
last_orthography = last_action.orthography
last_space = SPACE if last_action.text.endswith(SPACE) else NO_SPACE
was_space = len(last_space) is not 0
begin = False # for meta attach
meta = _get_meta(atom)
if meta is not None:
meta = _unescape_atom(meta)
if meta in META_COMMAS:
action.text = meta + SPACE
if last_action.text != '':
if was_space:
action.replace = SPACE
else:
action.replace = NO_SPACE
if last_attach:
action.replace = NO_SPACE
elif meta in META_STOPS:
action.text = meta + SPACE
action.capitalize = True
action.lower = False
if last_action.text != '':
if was_space:
action.replace = SPACE
else:
action.replace = NO_SPACE
if last_attach:
action.replace = NO_SPACE
elif meta == META_CAPITALIZE:
action = last_action.copy_state()
action.capitalize = True
action.lower = False
elif meta == META_LOWER:
action = last_action.copy_state()
if was_space:
# Persist space state
action.replace = SPACE
action.text = SPACE
action.lower = True
action.capitalize = False
elif meta == META_UPPER:
action = last_action.copy_state()
action.lower = False
action.upper = True
action.capitalize = False
elif (meta.startswith(META_CARRY_CAPITALIZATION) or
meta.startswith(META_ATTACH_FLAG + META_CARRY_CAPITALIZATION)):
action = _apply_carry_capitalize(meta, last_action, spaces_after=True)
elif meta == META_RETRO_CAPITALIZE:
action = last_action.copy_state()
action.word = _capitalize(action.word)
if len(last_action.text) < len(last_action.word):
action.replace = last_action.word + SPACE
action.text = _capitalize(last_action.word + SPACE)
else:
action.replace = last_action.text
action.text = _capitalize_nowhitespace(last_action.text)
elif meta == META_RETRO_LOWER:
action = last_action.copy_state()
action.word = _lower(action.word)
if len(last_action.text) < len(last_action.word):
action.replace = last_action.word + SPACE
action.text = _lower(last_action.word + SPACE)
else:
action.replace = last_action.text
action.text = _lower_nowhitespace(last_action.text)
elif meta == META_RETRO_UPPER:
action = last_action.copy_state()
action.word = _upper(action.word)
action.upper_carry = True
if len(last_action.text) < len(last_action.word):
action.replace = last_action.word + SPACE
action.text = _upper(last_action.word + SPACE)
else:
action.replace = last_action.text
action.text = _upper(last_action.text)
elif meta.startswith(META_RETRO_FORMAT):
if meta.startswith(META_RETRO_FORMAT) and meta.endswith(')'):
action = _apply_currency(meta, last_action, spaces_after=True)
elif meta.startswith(META_COMMAND):
action = last_action.copy_state()
action.command = meta[len(META_COMMAND):]
elif meta.startswith(META_MODE):
action = last_action.copy_state()
action = _change_mode(meta[len(META_MODE):], action)
elif meta.startswith(META_GLUE_FLAG):
action.glue = True
text = meta[len(META_GLUE_FLAG):]
if last_capitalize:
text = _capitalize(text)
if last_lower:
text = _lower(text)
action.text = text + SPACE
action.word = _rightmost_word(text)
if last_glue:
if was_space:
action.replace = SPACE
else:
action.replace = NO_SPACE
action.word = _rightmost_word(last_word + text)
if last_attach:
action.replace = NO_SPACE
action.word = _rightmost_word(last_word + text)
elif (meta.startswith(META_ATTACH_FLAG) or
meta.endswith(META_ATTACH_FLAG)):
begin = meta.startswith(META_ATTACH_FLAG)
end = meta.endswith(META_ATTACH_FLAG)
if begin:
meta = meta[len(META_ATTACH_FLAG):]
if end and len(meta) >= len(META_ATTACH_FLAG):
meta = meta[:-len(META_ATTACH_FLAG)]
space = NO_SPACE if end else SPACE
replace_space = NO_SPACE if last_attach else SPACE
if end:
action.attach = True
if begin and end and meta == '':
# We use an empty connection to indicate a "break" in the
# application of orthography rules. This allows the
# stenographer to tell plover not to auto-correct a word.
action.orthography = False
if last_action.text != '':
action.replace = replace_space
if (((begin and not end) or (begin and end and ' ' in meta)) and
last_orthography):
new = orthography.add_suffix(last_word.lower(), meta)
common = commonprefix([last_word.lower(), new])
if last_action.text == '':
replace_space = NO_SPACE
action.replace = last_word[len(common):] + replace_space
meta = new[len(common):]
if begin and end:
if last_action.text != '':
action.replace = replace_space
if last_capitalize:
meta = _capitalize(meta)
if last_lower:
meta = _lower(meta)
if last_upper_carry:
meta = _upper(meta)
action.upper_carry = True
action.text = meta + space
action.word = _rightmost_word(
last_word[:len(last_word + last_space)-len(action.replace)]
+ meta)
if end and not begin and last_space == SPACE:
action.word = _rightmost_word(meta)
elif meta.startswith(META_KEY_COMBINATION):
action = last_action.copy_state()
action.combo = meta[len(META_KEY_COMBINATION):]
else:
text = _unescape_atom(atom)
if last_capitalize:
text = _capitalize(text)
if last_lower:
text = _lower(text)
if last_upper:
text = _upper(text)
action.upper_carry = True
action.text = text + SPACE
action.word = _rightmost_word(text)
action.text = _apply_mode(action.text, action.case, action.space_char,
begin, last_attach, last_glue,
last_capitalize, last_upper, last_lower)
return action
def test_add_suffix(self):
cases = (
('artistic', 'ly', 'artistically'),
('cosmetic', 'ly', 'cosmetically'),
('establish', 's', 'establishes'),
('speech', 's', 'speeches'),
('approach', 's', 'approaches'),
('beach', 's', 'beaches'),
('arch', 's', 'arches'),
('larch', 's', 'larches'),
('march', 's', 'marches'),
('search', 's', 'searches'),
('starch', 's', 'starches'),
('stomach', 's', 'stomachs'),
('monarch', 's', 'monarchs'),
('patriarch', 's', 'patriarchs'),
('oligarch', 's', 'oligarchs'),
('cherry', 's', 'cherries'),
('day', 's', 'days'),
('penny', 's', 'pennies'),
('pharmacy', 'ist', 'pharmacist'),
('melody', 'ist', 'melodist'),
('pacify', 'ist', 'pacifist'),
('geology', 'ist', 'geologist'),
('metallurgy', 'ist', 'metallurgist'),
('anarchy', 'ist', 'anarchist'),
('monopoly', 'ist', 'monopolist'),
('alchemy', 'ist', 'alchemist'),
('botany', 'ist', 'botanist'),
('therapy', 'ist', 'therapist'),
('theory', 'ist', 'theorist'),
('psychiatry', 'ist', 'psychiatrist'),
('lobby', 'ist', 'lobbyist'),
('hobby', 'ist', 'hobbyist'),
('copy', 'ist', 'copyist'),
('beauty', 'ful', 'beautiful'),
('weary', 'ness', 'weariness'),
('weary', 'some', 'wearisome'),
('lonely', 'ness', 'loneliness'),
('narrate', 'ing', 'narrating'),
('narrate', 'or', 'narrator'),
('generalize', 'ability', 'generalizability'),
('reproduce', 'able', 'reproducible'),
('grade', 'ations', 'gradations'),
('urine', 'ary', 'urinary'),
('achieve', 'able', 'achievable'),
('polarize', 'ation', 'polarization'),
('done', 'or', 'donor'),
('analyze', 'ed', 'analyzed'),
('narrate', 'ing', 'narrating'),
('believe', 'able', 'believable'),
('animate', 'ors', 'animators'),
('discontinue', 'ation', 'discontinuation'),
('innovate', 'ive', 'innovative'),
('future', 'ists', 'futurists'),
('illustrate', 'or', 'illustrator'),
('emerge', 'ent', 'emergent'),
('equip', 'ed', 'equipped'),
('defer', 'ed', 'deferred'),
('defer', 'er', 'deferrer'),
('defer', 'ing', 'deferring'),
('pigment', 'ed', 'pigmented'),
('refer', 'ed', 'referred'),
('fix', 'ed', 'fixed'),
('alter', 'ed', 'altered'),
('interpret', 'ing', 'interpreting'),
('wonder', 'ing', 'wondering'),
('target', 'ing', 'targeting'),
('limit', 'er', 'limiter'),
('maneuver', 'ing', 'maneuvering'),
('monitor', 'ing', 'monitoring'),
('color', 'ing', 'coloring'),
('inhibit', 'ing', 'inhibiting'),
('master', 'ed', 'mastered'),
('target', 'ing', 'targeting'),
('fix', 'ed', 'fixed'),
('scrap', 'y', 'scrappy'),
('trip', 's', 'trips'),
('equip', 's', 'equips'),
('bat', 'en', 'batten'),
('smite', 'en', 'smitten'),
('got', 'en', 'gotten'),
('bite', 'en', 'bitten'),
('write', 'en', 'written'),
('flax', 'en', 'flaxen'),
('wax', 'en', 'waxen'),
('fast', 'est', 'fastest'),
('white', 'er', 'whiter'),
('crap', 'y', 'crappy'),
('lad', 'er', 'ladder'),
('translucent', 'cy', 'translucency'),
('bankrupt', 'cy', 'bankruptcy'),
('inadequate', 'cy', 'inadequacy'),
('secret', 'cy', 'secrecy'),
('impolite', 'cy', 'impolicy'),
('idiot', 'cy', 'idiocy'),
('free', 'ed', 'freed'),
('free', 'er', 'freer'),
('regulate', 'ry', 'regulatory'),
)
for word, suffix, expected in cases:
result = add_suffix(word, suffix)
msg = 'add_suffix(%r, %r) returned %r instead of %r' % (
word, suffix, result, expected,
)
self.assertEqual(result, expected, msg=msg)
def _atom_to_action_spaces_before(atom, last_action):
"""Convert an atom into an action.
Arguments:
atom -- A string holding an atom. An atom is an irreducible string that is
either entirely a single meta command or entirely text containing no meta
commands.
last_action -- The context in which the new action takes place.
Returns: An action for the atom.
"""
action = _Action()
last_word = last_action.word
last_glue = last_action.glue
last_attach = last_action.attach
last_capitalize = last_action.capitalize
last_lower = last_action.lower
last_upper = last_action.upper
last_upper_carry = last_action.upper_carry
last_orthography = last_action.orthography
meta = _get_meta(atom)
if meta is not None:
meta = _unescape_atom(meta)
if meta in META_COMMAS:
action.text = meta
elif meta in META_STOPS:
action.text = meta
action.capitalize = True
action.lower = False
action.upper = False
elif meta == META_CAPITALIZE:
action = last_action.copy_state()
action.capitalize = True
action.lower = False
action.upper = False
elif meta == META_LOWER:
action = last_action.copy_state()
action.lower = True
action.upper = False
action.capitalize = False
elif meta == META_UPPER:
action = last_action.copy_state()
action.lower = False
action.upper = True
action.capitalize = False
elif meta == META_RETRO_CAPITALIZE:
action = last_action.copy_state()
action.word = _capitalize(action.word)
if len(last_action.text) < len(last_action.word):
action.replace = last_action.word
action.text = _capitalize(last_action.word)
else:
action.replace = last_action.text
action.text = _capitalize_nowhitespace(last_action.text)
elif meta == META_RETRO_LOWER:
action = last_action.copy_state()
action.word = _lower(action.word)
if len(last_action.text) < len(last_action.word):
action.replace = last_action.word
action.text = _lower(last_action.word)
else:
action.replace = last_action.text
action.text = _lower_nowhitespace(last_action.text)
elif meta == META_RETRO_UPPER:
action = last_action.copy_state()
action.word = _upper(action.word)
action.upper_carry = True
if len(last_action.text) < len(last_action.word):
action.replace = last_action.word
action.text = _upper(last_action.word)
else:
action.replace = last_action.text
action.text = _upper(last_action.text)
elif meta.startswith(META_RETRO_FORMAT):
if (meta.startswith(META_RETRO_FORMAT) and meta.endswith(')')):
dict_format = meta[len(META_RETRO_FORMAT):-len(')')]
action = last_action.copy_state()
action.replace = last_action.word
try:
float(last_action.word)
except ValueError:
pass
else:
format = dict_format.replace('c', '{:,.2f}')
cast_input = float(last_action.word)
try:
int(last_action.word)
except ValueError:
pass
else:
format = dict_format.replace('c', '{:,}')
cast_input = int(last_action.word)
action.text = format.format(cast_input)
action.word = action.text
elif meta.startswith(META_COMMAND):
action = last_action.copy_state()
action.command = meta[len(META_COMMAND):]
elif meta.startswith(META_GLUE_FLAG):
action.glue = True
glue = last_glue or last_attach
space = NO_SPACE if glue else SPACE
text = meta[len(META_GLUE_FLAG):]
if last_capitalize:
text = _capitalize(text)
if last_lower:
text = _lower(text)
action.text = space + text
action.word = _rightmost_word(last_word + action.text)
elif (meta.startswith(META_ATTACH_FLAG) or
meta.endswith(META_ATTACH_FLAG)):
begin = meta.startswith(META_ATTACH_FLAG)
end = meta.endswith(META_ATTACH_FLAG)
if begin:
meta = meta[len(META_ATTACH_FLAG):]
if end and len(meta) >= len(META_ATTACH_FLAG):
meta = meta[:-len(META_ATTACH_FLAG)]
space = NO_SPACE if begin or last_attach else SPACE
if end:
action.attach = True
if begin and end and meta == '':
# We use an empty connection to indicate a "break" in the
# application of orthography rules. This allows the stenographer
# to tell plover not to auto-correct a word.
action.orthography = False
if (((begin and not end) or (begin and end and ' ' in meta)) and
last_orthography):
new = orthography.add_suffix(last_word.lower(), meta)
common = commonprefix([last_word.lower(), new])
action.replace = last_word[len(common):]
meta = new[len(common):]
if last_capitalize:
meta = _capitalize(meta)
if last_lower:
meta = _lower(meta)
if last_upper_carry:
meta = _upper(meta)
action.upper_carry = True
action.text = space + meta
action.word = _rightmost_word(
last_word[:len(last_word)-len(action.replace)] + action.text)
elif meta.startswith(META_KEY_COMBINATION):
action = last_action.copy_state()
action.combo = meta[len(META_KEY_COMBINATION):]
else:
text = _unescape_atom(atom)
if last_capitalize:
text = _capitalize(text)
if last_lower:
text = _lower(text)
if last_upper:
text = _upper(text)
action.upper_carry = True
space = NO_SPACE if last_attach else SPACE
action.text = space + text
action.word = _rightmost_word(text)
return action
def test_add_suffix(self):
cases = (
('artistic', 'ly', 'artistically'),
('cosmetic', 'ly', 'cosmetically'),
('establish', 's', 'establishes'),
('speech', 's', 'speeches'),
('approach', 's', 'approaches'),
('beach', 's', 'beaches'),
('arch', 's', 'arches'),
('larch', 's', 'larches'),
('march', 's', 'marches'),
('search', 's', 'searches'),
('starch', 's', 'starches'),
('stomach', 's', 'stomachs'),
('monarch', 's', 'monarchs'),
('patriarch', 's', 'patriarchs'),
('oligarch', 's', 'oligarchs'),
('cherry', 's', 'cherries'),
('day', 's', 'days'),
('penny', 's', 'pennies'),
('pharmacy', 'ist', 'pharmacist'),
('melody', 'ist', 'melodist'),
('pacify', 'ist', 'pacifist'),
('geology', 'ist', 'geologist'),
('metallurgy', 'ist', 'metallurgist'),
('anarchy', 'ist', 'anarchist'),
('monopoly', 'ist', 'monopolist'),
('alchemy', 'ist', 'alchemist'),
('botany', 'ist', 'botanist'),
('therapy', 'ist', 'therapist'),
('theory', 'ist', 'theorist'),
('psychiatry', 'ist', 'psychiatrist'),
('lobby', 'ist', 'lobbyist'),
('hobby', 'ist', 'hobbyist'),
('copy', 'ist', 'copyist'),
('beauty', 'ful', 'beautiful'),
('weary', 'ness', 'weariness'),
('weary', 'some', 'wearisome'),
('lonely', 'ness', 'loneliness'),
('narrate', 'ing', 'narrating'),
('narrate', 'or', 'narrator'),
('generalize', 'ability', 'generalizability'),
('reproduce', 'able', 'reproducible'),
('grade', 'ations', 'gradations'),
('urine', 'ary', 'urinary'),
('achieve', 'able', 'achievable'),
('polarize', 'ation', 'polarization'),
('done', 'or', 'donor'),
('analyze', 'ed', 'analyzed'),
('narrate', 'ing', 'narrating'),
('believe', 'able', 'believable'),
('animate', 'ors', 'animators'),
('discontinue', 'ation', 'discontinuation'),
('innovate', 'ive', 'innovative'),
('future', 'ists', 'futurists'),
('illustrate', 'or', 'illustrator'),
('emerge', 'ent', 'emergent'),
('equip', 'ed', 'equipped'),
('defer', 'ed', 'deferred'),
('defer', 'er', 'deferrer'),
('defer', 'ing', 'deferring'),
('pigment', 'ed', 'pigmented'),
('refer', 'ed', 'referred'),
('fix', 'ed', 'fixed'),
('alter', 'ed', 'altered'),
('interpret', 'ing', 'interpreting'),
('wonder', 'ing', 'wondering'),
('target', 'ing', 'targeting'),
('limit', 'er', 'limiter'),
('maneuver', 'ing', 'maneuvering'),
('monitor', 'ing', 'monitoring'),
('color', 'ing', 'coloring'),
('inhibit', 'ing', 'inhibiting'),
('master', 'ed', 'mastered'),
('target', 'ing', 'targeting'),
('fix', 'ed', 'fixed'),
('scrap', 'y', 'scrappy'),
('trip', 's', 'trips'),
('equip', 's', 'equips'),
('bat', 'en', 'batten'),
('smite', 'en', 'smitten'),
('got', 'en', 'gotten'),
('bite', 'en', 'bitten'),
('write', 'en', 'written'),
('flax', 'en', 'flaxen'),
('wax', 'en', 'waxen'),
('fast', 'est', 'fastest'),
('white', 'er', 'whiter'),
('crap', 'y', 'crappy'),
('lad', 'er', 'ladder'),
('translucent', 'cy', 'translucency'),
('bankrupt', 'cy', 'bankruptcy'),
('inadequate', 'cy', 'inadequacy'),
('secret', 'cy', 'secrecy'),
('impolite', 'cy', 'impolicy'),
('idiot', 'cy', 'idiocy'),
('free', 'ed', 'freed'),
('free', 'er', 'freer'),
('regulate', 'ry', 'regulatory'),
)
failed = []
for word, suffix, expected in cases:
if add_suffix(word, suffix) != expected:
failed.append((word, suffix, expected))
for word, suffix, expected in failed:
print 'add_suffix(%s, %s) is %s not %s' % (
word, suffix, add_suffix(word, suffix), expected)
self.assertEqual(len(failed), 0)
def test_add_suffix(word, suffix, expected):
assert add_suffix(word, suffix) == expected
def test_add_suffix(word, suffix, expected):
assert add_suffix(word, suffix) == expected