def _check_skintone(sorted_seq_to_filepath):
"""Ensure skin tone modifiers are not applied to emoji that are not defined
to take them. May appear standalone, though. Also check that emoji that take
skin tone modifiers have a complete set."""
base_to_modifiers = collections.defaultdict(set)
for seq, fp in sorted_seq_to_filepath.iteritems():
for i, cp in enumerate(seq):
if _is_skintone_modifier(cp):
if i == 0:
if len(seq) > 1:
print >> sys.stderr, 'skin color selector first in sequence %s' % fp
# standalone are ok
continue
pcp = seq[i - 1]
if not unicode_data.is_emoji_modifier_base(pcp):
print >> sys.stderr, (
'emoji skintone modifier applied to non-base at %d: %s'
% (i, fp))
elif unicode_data.is_emoji_modifier_base(cp):
if i < len(seq) - 1 and _is_skintone_modifier(seq[i + 1]):
base_to_modifiers[cp].add(seq[i + 1])
elif cp not in base_to_modifiers:
base_to_modifiers[cp] = set()
for cp, modifiers in sorted(base_to_modifiers.iteritems()):
if len(modifiers) != 5:
print >> sys.stderr, 'emoji base %04x has %d modifiers defined (%s) in %s' % (
cp, len(modifiers), ', '.join('%04x' % cp
for cp in sorted(modifiers)), fp)
def _check_skintone(sorted_seqs):
"""Ensure skin tone modifiers are not applied to emoji that are not defined
to take them. May appear standalone, though. Also check that emoji that take
skin tone modifiers have a complete set."""
base_to_modifiers = collections.defaultdict(set)
for seq in sorted_seqs:
for i, cp in enumerate(seq):
if _is_skintone_modifier(cp):
if i == 0:
if len(seq) > 1:
print >> sys.stderr, 'skin color selector first in sequence %s'
# standalone are ok
continue
pcp = seq[i-1]
if not unicode_data.is_emoji_modifier_base(pcp):
print >> sys.stderr, (
'emoji skintone modifier applied to non-base at %d: %s' % (
i, _seq_string(seq)))
elif unicode_data.is_emoji_modifier_base(cp):
if i < len(seq) - 1 and _is_skintone_modifier(seq[i+1]):
base_to_modifiers[cp].add(seq[i+1])
elif cp not in base_to_modifiers:
base_to_modifiers[cp] = set()
for cp, modifiers in sorted(base_to_modifiers.iteritems()):
if len(modifiers) != 5:
print 'emoji base %04X has %d modifiers defined (%s)' % (
cp, len(modifiers),
', '.join('%04x' % cp for cp in sorted(modifiers)))
def _check_skintone(sorted_seq_to_filepath):
"""Ensure skin tone modifiers are not applied to emoji that are not defined
to take them. May appear standalone, though. Also check that emoji that take
skin tone modifiers have a complete set."""
base_to_modifiers = collections.defaultdict(set)
for seq, fp in sorted_seq_to_filepath.iteritems():
for i, cp in enumerate(seq):
if unicode_data.is_skintone_modifier(cp):
if i == 0:
if len(seq) > 1:
print(
'check skintone: skin color selector first in sequence %s' % fp,
file=sys.stderr)
# standalone are ok
continue
pcp = seq[i-1]
if not unicode_data.is_emoji_modifier_base(pcp):
print(
'check skintone: emoji skintone modifier applied to non-base ' +
'at %d: %s' % (i, fp), file=sys.stderr)
else:
if pcp not in base_to_modifiers:
base_to_modifiers[pcp] = set()
base_to_modifiers[pcp].add(cp)
for cp, modifiers in sorted(base_to_modifiers.iteritems()):
if len(modifiers) != 5:
print(
'check skintone: base %04x has %d modifiers defined (%s) in %s' % (
cp, len(modifiers),
', '.join('%04x' % cp for cp in sorted(modifiers)), fp),
file=sys.stderr)
def _check_skintone(sorted_seq_to_filepath):
"""Ensure skin tone modifiers are not applied to emoji that are not defined
to take them. May appear standalone, though. Also check that emoji that take
skin tone modifiers have a complete set."""
base_to_modifiers = collections.defaultdict(set)
for seq, fp in sorted_seq_to_filepath.items():
for i, cp in enumerate(seq):
if unicode_data.is_skintone_modifier(cp):
if i == 0:
if len(seq) > 1:
print(
f'check skintone: skin color selector first in sequence {fp}',
file=sys.stderr)
# standalone are ok
continue
pcp = seq[i - 1]
if not unicode_data.is_emoji_modifier_base(pcp):
print(
f'check skintone: emoji skintone modifier applied to non-base at {i}: {fp}',
file=sys.stderr)
else:
if pcp not in base_to_modifiers:
base_to_modifiers[pcp] = set()
base_to_modifiers[pcp].add(cp)
for cp, modifiers in sorted(base_to_modifiers.items()):
if len(modifiers) != 5:
print(
'check skintone: base %04x has %d modifiers defined (%s) in %s'
% (cp, len(modifiers), ', '.join(
'%04x' % cp for cp in sorted(modifiers)), fp),
file=sys.stderr)
def _fix_estrs(estrs):
"""Add fitzpatrick colors, and remove emoji variation selectors.
1) The emoji ordering html page omits the skin color variants. Since
we want to display them, we have to add them. However, we don't add
them for sequences with multiple people, since we don't support
multiracial groups at the moment.
2) Currently we don't include emoji variation selectors in our sequences
except after heavy black heart (and we probably don't need them there
either). The font will work fine without them, but some systems doing
fallback might break the sequences apart if the variation selectors are
not present in text. So they should be there in the text, and not there
in the font. Anyway, while we clean this up, we currently strip them
except for the four cases where we retain them for legacy reasons.
"""
new_estrs = []
for estr in estrs:
if estr in _keep_varsel_estrs:
nestr = estr
else:
if u'\u2764' in estr:
print '# oops', u''.join('\\u%04x' % ord(cp) for cp in estr)
nestr = u''.join(cp for cp in estr if cp != u'\ufe0f')
new_estrs.append(nestr)
num_bases = sum(
unicode_data.is_emoji_modifier_base(ord(cp)) for cp in estr)
if num_bases != 1:
continue
split_before = len(nestr)
for i, cp in enumerate(nestr):
if unicode_data.is_emoji_modifier_base(ord(cp)):
split_before = i + 1
break
prefix = nestr[:split_before]
suffix = nestr[split_before:]
for cp in range(0x1f3fb, 0x1f3ff + 1):
new_estrs.append(u''.join([prefix, unichr(cp), suffix]))
return new_estrs
def _fix_estrs(estrs):
"""Add fitzpatrick colors, and remove emoji variation selectors.
1) The emoji ordering html page omits the skin color variants. Since
we want to display them, we have to add them. However, we don't add
them for sequences with multiple people, since we don't support
multiracial groups at the moment.
2) Currently we don't include emoji variation selectors in our sequences
except after heavy black heart (and we probably don't need them there
either). The font will work fine without them, but some systems doing
fallback might break the sequences apart if the variation selectors are
not present in text. So they should be there in the text, and not there
in the font. Anyway, while we clean this up, we currently strip them
except for the four cases where we retain them for legacy reasons.
"""
new_estrs = []
for estr in estrs:
if estr in _keep_varsel_estrs:
nestr = estr
else:
if u'\u2764' in estr:
print '# oops', u''.join('\\u%04x' % ord(cp) for cp in estr)
nestr = u''.join(cp for cp in estr if cp != u'\ufe0f')
new_estrs.append(nestr)
num_bases = sum(unicode_data.is_emoji_modifier_base(ord(cp)) for cp in estr)
if num_bases != 1:
continue
split_before = len(nestr)
for i, cp in enumerate(nestr):
if unicode_data.is_emoji_modifier_base(ord(cp)):
split_before = i + 1
break
prefix = nestr[:split_before]
suffix = nestr[split_before:]
for cp in range(0x1f3fb, 0x1f3ff + 1):
new_estrs.append(u''.join([prefix, unichr(cp), suffix]))
return new_estrs