1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
|
from .IndicShaperData import IndicPosition, make_syllable_machine
from .SyllabicShaper import SyllabicShaper
from collections import OrderedDict
myanmar_category_reassignments = {
0x002D: "GB",
0x00A0: "GB",
0x00D7: "GB",
0x1004: "Ra",
0x101B: "Ra",
0x1032: "A",
0x1036: "A",
0x1038: "SM",
0x1039: "H",
0x103A: "As",
0x103B: "MY",
0x103C: "MR",
0x103D: "MW",
0x103E: "MH",
0x1040: "D", # XXX The spec says D0, but Uniscribe doesn't seem to do.
0x104A: "P",
0x104B: "P",
0x104E: "C",
0x105A: "Ra",
0x105E: "MY",
0x105F: "MY",
0x1060: "MH",
0x1082: "MW",
0x109A: "SM",
0x109B: "SM",
0x109C: "SM",
0x2012: "GB",
0x2013: "GB",
0x2014: "GB",
0x2015: "GB",
0x2022: "GB",
0x25CC: "GB",
0x25FB: "GB",
0x25FC: "GB",
0x25FD: "GB",
0x25FE: "GB",
0xAA74: "C",
0xAA75: "C",
0xAA76: "C", # https://github.com/harfbuzz/harfbuzz/issues/218
}
for cp in range(0x1041, 0x1049 + 1):
myanmar_category_reassignments[cp] = "D"
for cp in range(0x1090, 0x1099 + 1):
myanmar_category_reassignments[cp] = "D"
for cp in range(0xF300, 0xFE0F + 1):
myanmar_category_reassignments[cp] = "VS"
for cp in range(0x1087, 0x108F + 1):
myanmar_category_reassignments[cp] = "SM"
for cp in [0x1063, 0x1064, 0x1069, 0x106A, 0x106B, 0x106C, 0x106D, 0xAA7B]:
myanmar_category_reassignments[cp] = "PT"
states = OrderedDict(
j="ZWJ|ZWNJ", # Joiners
k="(Ra As H)", # Kinzi
c="C|Ra", # is_consonant
medial_group="MY? As? MR? ((MW MH? | MH) As?)?",
main_vowel_group="(VPre VS?)* VAbv* VBlw* A* (DB As?)?",
post_vowel_group="VPst MH? As* VAbv* A* (DB As?)?",
pwo_tone_group="PT A* DB? As?",
complex_syllable_tail="As* medial_group main_vowel_group post_vowel_group* pwo_tone_group* V* j?",
syllable_tail="(H (c|IV) VS?)* (H | complex_syllable_tail)",
consonant_syllable="(k|CS)? (c|IV|D|GB) VS? syllable_tail",
punctuation_cluster="P V",
broken_cluster="k? VS? syllable_tail",
)
class MyanmarShaper(SyllabicShaper):
basic_features = ["rphf", "pref", "blwf", "pstf"]
other_features = ["pres", "abvs", "blws", "psts"]
repha = None
syllable_machine = make_syllable_machine(
states,
additional_categories=[
"VS",
"MW",
"P",
"As",
"PT",
"MY",
"MH",
"D",
"GB",
"MR",
"VPre",
"VAbv",
"VBlw",
"VPst",
],
)
syllable_types = [
"consonant_syllable",
"punctuation_cluster",
"broken_cluster",
"other",
]
def reassign_category(self, item):
cp = item.codepoint
if cp in myanmar_category_reassignments:
item.syllabic_category = myanmar_category_reassignments[cp]
if item.syllabic_category == "M":
if item.syllabic_position == IndicPosition.PRE_C:
item.syllabic_category = "VPre"
item.syllabic_position = IndicPosition.PRE_M
elif item.syllabic_position == IndicPosition.ABOVE_C:
item.syllabic_category = "VAbv"
elif item.syllabic_position == IndicPosition.BELOW_C:
item.syllabic_category = "VBlw"
elif item.syllabic_position == IndicPosition.POST_C:
item.syllabic_category = "VPst"
def initial_reordering_consonant_syllable(self, start, end):
def cat(i):
return self.buffer.items[i].syllabic_category
def get_pos(i):
return self.buffer.items[i].syllabic_position
def set_pos(i, pos):
self.buffer.items[i].syllabic_position = pos
def is_consonant(n): # XXX Something else is Placeholder too
isc = cat(n)
is_medial = isc == "CM"
return (
isc in ["C", "CS", "Ra", "V", "PLACEHOLDER", "DOTTEDCIRCLE"]
or is_medial
)
base = end
has_reph = False
limit = start
if (
start + 3 <= end
and cat(start) == "Ra"
and cat(start + 1) == "As"
and cat(start + 2) == "H"
):
limit += 3
base = start
has_reph = True
if not has_reph:
base = limit
for i in range(limit, end):
if is_consonant(i):
base = i
break
i = start
while i < start + (3 if has_reph else 0):
set_pos(i, IndicPosition.AFTER_MAIN)
i += 1
while i < base:
set_pos(i, IndicPosition.PRE_C)
i += 1
if i < end:
set_pos(i, IndicPosition.BASE_C)
i += 1
pos = IndicPosition.AFTER_MAIN
while i < end:
if cat(i) == "MR":
set_pos(i, IndicPosition.PRE_C)
elif get_pos(i) < IndicPosition.BASE_C: # Left Matra
pass
elif cat(i) == "VS":
set_pos(i, get_pos(i - 1))
elif pos == IndicPosition.AFTER_MAIN and cat(i) == "VBlw":
pos = IndicPosition.BELOW_C
set_pos(i, pos)
elif pos == IndicPosition.BELOW_C and cat(i) == "A":
set_pos(i, IndicPosition.BEFORE_SUB)
elif pos == IndicPosition.BELOW_C and cat(i) == "VBlw":
set_pos(i, pos)
elif pos == IndicPosition.BELOW_C and cat(i) != "A":
pos = IndicPosition.AFTER_SUB
set_pos(i, pos)
else:
set_pos(i, pos)
i = i + 1
continue
self.buffer.items[start:end] = sorted(
self.buffer.items[start:end], key=lambda x: x.syllabic_position
)
initial_reordering_syllable = {
"broken_cluster": initial_reordering_consonant_syllable,
"consonant_syllable": initial_reordering_consonant_syllable,
}
|
