summaryrefslogtreecommitdiff
path: root/Build/source/libs/harfbuzz/harfbuzz-0.9.7/src/gen-indic-table.py
blob: 94aa2ab0728e3e03201697cfe1b4e645209f96a5 (plain)
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
200
201
202
203
204
205
206
207
208
209
210
211
#!/usr/bin/python

import sys

if len (sys.argv) != 4:
	print >>sys.stderr, "usage: ./gen-indic-table.py IndicSyllabicCategory.txt IndicMatraCategory.txt Blocks.txt"
	sys.exit (1)

files = [file (x) for x in sys.argv[1:]]

headers = [[f.readline () for i in range (2)] for f in files]

blocks = {}
data = [{} for f in files]
values = [{} for f in files]
for i, f in enumerate (files):
	for line in f:

		j = line.find ('#')
		if j >= 0:
			line = line[:j]

		fields = [x.strip () for x in line.split (';')]
		if len (fields) == 1:
			continue

		uu = fields[0].split ('..')
		start = int (uu[0], 16)
		if len (uu) == 1:
			end = start
		else:
			end = int (uu[1], 16)

		t = fields[1]

		for u in range (start, end + 1):
			data[i][u] = t
		values[i][t] = values[i].get (t, 0) + 1

		if i == 2:
			blocks[t] = (start, end)

# Merge data into one dict:
defaults = ('Other', 'Not_Applicable', 'No_Block')
for i,v in enumerate (defaults):
	values[i][v] = values[i].get (v, 0) + 1
combined = {}
for i,d in enumerate (data):
	for u,v in d.items ():
		if i == 2 and not u in combined:
			continue
		if not u in combined:
			combined[u] = list (defaults)
		combined[u][i] = v
data = combined
del combined
num = len (data)

# Move the outliers NO-BREAK SPACE and DOTTED CIRCLE out
singles = {}
for u in [0x00A0, 0x25CC]:
	singles[u] = data[u]
	del data[u]

print "/* == Start of generated table == */"
print "/*"
print " * The following table is generated by running:"
print " *"
print " *   ./gen-indic-table.py IndicSyllabicCategory.txt IndicMatraCategory.txt Blocks.txt"
print " *"
print " * on files with these headers:"
print " *"
for h in headers:
	for l in h:
		print " * %s" % (l.strip())
print " */"
print
print "#ifndef HB_OT_SHAPE_COMPLEX_INDIC_TABLE_HH"
print "#define HB_OT_SHAPE_COMPLEX_INDIC_TABLE_HH"
print

# Shorten values
short = [{
	"Bindu":		'Bi',
	"Visarga":		'Vs',
	"Vowel":		'Vo',
	"Vowel_Dependent":	'M',
	"Other":		'x',
},{
	"Not_Applicable":	'x',
}]
all_shorts = [[],[]]

# Add some of the values, to make them more readable, and to avoid duplicates


for i in range (2):
	for v,s in short[i].items ():
		all_shorts[i].append (s)

what = ["INDIC_SYLLABIC_CATEGORY", "INDIC_MATRA_CATEGORY"]
what_short = ["ISC", "IMC"]
for i in range (2):
	print
	vv = values[i].keys ()
	vv.sort ()
	for v in vv:
		v_no_and = v.replace ('_And_', '_')
		if v in short[i]:
			s = short[i][v]
		else:
			s = ''.join ([c for c in v_no_and if ord ('A') <= ord (c) <= ord ('Z')])
			if s in all_shorts[i]:
				raise Exception ("Duplicate short value alias", v, s)
			all_shorts[i].append (s)
			short[i][v] = s
		print "#define %s_%s	%s_%s	%s/* %3d chars; %s */" % \
			(what_short[i], s, what[i], v.upper (), \
			'	'* ((48-1 - len (what[i]) - 1 - len (v)) / 8), \
			values[i][v], v)
print
print "#define _(S,M) INDIC_COMBINE_CATEGORIES (ISC_##S, IMC_##M)"
print
print

total = 0
used = 0
def print_block (block, start, end, data):
	print
	print
	print "  /* %s  (%04X..%04X) */" % (block, start, end)
	num = 0
	for u in range (start, end+1):
		if u % 8 == 0:
			print
			print "  /* %04X */" % u,
		if u in data:
			num += 1
		d = data.get (u, defaults)
		sys.stdout.write ("%9s" % ("_(%s,%s)," % (short[0][d[0]], short[1][d[1]])))

	global total, used
	total += end - start + 1
	used += num

uu = data.keys ()
uu.sort ()

last = -1
num = 0
offset = 0
starts = []
ends = []
print "static const INDIC_TABLE_ELEMENT_TYPE indic_table[] = {"
for u in uu:
	if u <= last:
		continue
	block = data[u][2]
	(start, end) = blocks[block]

	if start != last + 1:
		if start - last <= 33:
			print_block ("FILLER", last+1, start-1, data)
			last = start-1
		else:
			if last >= 0:
				ends.append (last + 1)
				offset += ends[-1] - starts[-1]
			print
			print
			print "#define indic_offset_0x%04x %d" % (start, offset)
			starts.append (start)

	print_block (block, start, end, data)
	last = end
ends.append (last + 1)
offset += ends[-1] - starts[-1]
print
print
print "#define indic_offset_total %d" % offset
print
occupancy = used * 100. / total
print "}; /* Table occupancy: %d%% */" % occupancy
print
print "static INDIC_TABLE_ELEMENT_TYPE"
print "get_indic_categories (hb_codepoint_t u)"
print "{"
for (start,end) in zip (starts, ends):
	offset = "indic_offset_0x%04x" % start
	print "  if (0x%04X <= u && u <= 0x%04X) return indic_table[u - 0x%04X + %s];" % (start, end, start, offset)
for u,d in singles.items ():
	print "  if (unlikely (u == 0x%04X)) return _(%s,%s);" % (u, short[0][d[0]], short[1][d[1]])
print "  return _(x,x);"
print "}"
print
print "#undef _"
for i in range (2):
	print
	vv = values[i].keys ()
	vv.sort ()
	for v in vv:
		print "#undef %s_%s" % \
			(what_short[i], short[i][v])
print
print "#endif /* HB_OT_SHAPE_COMPLEX_INDIC_TABLE_HH */"
print
print "/* == End of generated table == */"

# Maintain at least 30% occupancy in the table */
if occupancy < 30:
	raise Exception ("Table too sparse, please investigate: ", occupancy)