summaryrefslogtreecommitdiff
path: root/Build/source/libs/harfbuzz/harfbuzz-src/src/gen-indic-table.py
blob: 912b1d7ea2c35d5b6356e953bc339da3e867c846 (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
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
#!/usr/bin/env python

from __future__ import print_function, division, absolute_import

import io, sys

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

ALLOWED_SINGLES = [0x00A0, 0x25CC]
ALLOWED_BLOCKS = [
	'Basic Latin',
	'Latin-1 Supplement',
	'Devanagari',
	'Bengali',
	'Gurmukhi',
	'Gujarati',
	'Oriya',
	'Tamil',
	'Telugu',
	'Kannada',
	'Malayalam',
	'Sinhala',
	'Myanmar',
	'Khmer',
	'Vedic Extensions',
	'General Punctuation',
	'Superscripts and Subscripts',
	'Devanagari Extended',
	'Myanmar Extended-B',
	'Myanmar Extended-A',
]

files = [io.open (x, encoding='utf-8') for x in sys.argv[1:]]

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

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) + end - start + 1

# 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
combined = {k:v for k,v in combined.items() if k in ALLOWED_SINGLES or v[2] in ALLOWED_BLOCKS}
data = combined
del combined
num = len (data)

# Move the outliers NO-BREAK SPACE and DOTTED CIRCLE out
singles = {}
for u in ALLOWED_SINGLES:
	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 IndicPositionalCategory.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 ('#include "hb.hh"')
print ()
print ('#ifndef HB_NO_OT_SHAPE')
print ()
print ('#include "hb-ot-shape-complex-indic.hh"')
print ()

# Shorten values
short = [{
	"Bindu":		'Bi',
	"Cantillation_Mark":	'Ca',
	"Joiner":		'ZWJ',
	"Non_Joiner":		'ZWNJ',
	"Number":		'Nd',
	"Visarga":		'Vs',
	"Vowel":		'Vo',
	"Vowel_Dependent":	'M',
	"Consonant_Prefixed":	'CPrf',
	"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][s] = v

what = ["INDIC_SYLLABIC_CATEGORY", "INDIC_MATRA_CATEGORY"]
what_short = ["ISC", "IMC"]
print ('#pragma GCC diagnostic push')
print ('#pragma GCC diagnostic ignored "-Wunused-macros"')
cat_defs = []
for i in range (2):
	vv = sorted (values[i].keys ())
	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, all_shorts[i][s])
			all_shorts[i][s] = v
			short[i][v] = s
		cat_defs.append ((what_short[i] + '_' + s, what[i] + '_' + v.upper (), str (values[i][v]), v))

maxlen_s = max ([len (c[0]) for c in cat_defs])
maxlen_l = max ([len (c[1]) for c in cat_defs])
maxlen_n = max ([len (c[2]) for c in cat_defs])
for s in what_short:
	print ()
	for c in [c for c in cat_defs if s in c[0]]:
		print ("#define %s %s /* %s chars; %s */" %
			(c[0].ljust (maxlen_s), c[1].ljust (maxlen_l), c[2].rjust (maxlen_n), c[3]))
print ()
print ('#pragma GCC diagnostic pop')
print ()
print ("#define _(S,M) INDIC_COMBINE_CATEGORIES (ISC_##S, IMC_##M)")
print ()
print ()

total = 0
used = 0
last_block = None
def print_block (block, start, end, data):
	global total, used, last_block
	if block and block != last_block:
		print ()
		print ()
		print ("  /* %s */" % block)
	num = 0
	assert start % 8 == 0
	assert (end+1) % 8 == 0
	for u in range (start, end+1):
		if u % 8 == 0:
			print ()
			print ("  /* %04X */" % u, end="")
		if u in data:
			num += 1
		d = data.get (u, defaults)
		print ("%9s" % ("_(%s,%s)," % (short[0][d[0]], short[1][d[1]])), end="")

	total += end - start + 1
	used += num
	if block:
		last_block = block

uu = sorted (data.keys ())

last = -100000
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 = u//8*8
	end = start+1
	while end in uu and block == data[end][2]:
		end += 1
	end = (end-1)//8*8 + 7

	if start != last + 1:
		if start - last <= 1+16*3:
			print_block (None, 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%04xu %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 ()
occupancy = used * 100. / total
page_bits = 12
print ("}; /* Table items: %d; occupancy: %d%% */" % (offset, occupancy))
print ()
print ("INDIC_TABLE_ELEMENT_TYPE")
print ("hb_indic_get_categories (hb_codepoint_t u)")
print ("{")
print ("  switch (u >> %d)" % page_bits)
print ("  {")
pages = set ([u>>page_bits for u in starts+ends+list (singles.keys ())])
for p in sorted(pages):
	print ("    case 0x%0Xu:" % p)
	for u,d in singles.items ():
		if p != u>>page_bits: continue
		print ("      if (unlikely (u == 0x%04Xu)) return _(%s,%s);" % (u, short[0][d[0]], short[1][d[1]]))
	for (start,end) in zip (starts, ends):
		if p not in [start>>page_bits, end>>page_bits]: continue
		offset = "indic_offset_0x%04xu" % start
		print ("      if (hb_in_range<hb_codepoint_t> (u, 0x%04Xu, 0x%04Xu)) return indic_table[u - 0x%04Xu + %s];" % (start, end-1, start, offset))
	print ("      break;")
	print ("")
print ("    default:")
print ("      break;")
print ("  }")
print ("  return _(x,x);")
print ("}")
print ()
print ("#undef _")
for i in range (2):
	print ()
	vv = sorted (values[i].keys ())
	for v in vv:
		print ("#undef %s_%s" %
			(what_short[i], short[i][v]))
print ()
print ('#endif')
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)