6 print >>sys.stderr, "usage: ./gen-indic-table.py IndicSyllabicCategory.txt IndicMatraCategory.txt Blocks.txt"
9 files = [file (sys.argv[i+1]) for i in range (3)]
11 headers = [[f.readline () for i in range (2)] for f in files]
14 data = [{} for f in files]
15 values = [{} for f in files]
16 for i, f in enumerate (files):
23 fields = [x.strip () for x in line.split (';')]
27 uu = fields[0].split ('..')
28 start = int (uu[0], 16)
36 for u in range (start, end + 1):
38 values[i][t] = values[i].get (t, 0) + 1
41 blocks[t] = (start, end)
43 # Merge data into one dict:
44 defaults = ('Other', 'Not_Applicable', 'No_Block')
45 for i,v in enumerate (defaults):
46 values[i][v] = values[i].get (v, 0) + 1
48 for i,d in enumerate (data):
49 for u,v in d.items ():
50 if i == 2 and not u in combined:
53 combined[u] = list (defaults)
59 # Move the outliers NO-BREAK SPACE and DOTTED CIRCLE out
61 for u in [0x00A0, 0x25CC]:
65 print "/* == Start of generated table == */"
67 print " * The following table is generated by running:"
69 print " * ./gen-indic-table.py IndicSyllabicCategory.txt IndicMatraCategory.txt Blocks.txt"
71 print " * on files with these headers:"
75 print " * %s" % (l.strip())
86 "Not_Applicable": 'X',
90 # Add some of the values, to make them more readable, and to avoid duplicates
94 for v,s in short[i].items ():
95 all_shorts[i].append (s)
97 what = ["INDIC_SYLLABIC_CATEGORY", "INDIC_MATRA_CATEGORY"]
98 what_short = ["ISC", "IMC"]
101 vv = values[i].keys ()
104 v_no_and = v.replace ('_And_', '_')
108 s = ''.join ([c for c in v_no_and if ord ('A') <= ord (c) <= ord ('Z')])
109 if s in all_shorts[i]:
110 raise Exception ("Duplicate short value alias", v, s)
111 all_shorts[i].append (s)
113 print "#define %s_%s %s_%s %s/* %3d chars; %s */" % \
114 (what_short[i], s, what[i], v.upper (), \
115 ' '* ((48-1 - len (what[i]) - 1 - len (v)) / 8), \
118 print "#define _(S,M) INDIC_COMBINE_CATEGORIES (ISC_##S, IMC_##M)"
122 def print_block (block, start, end, data):
125 print " /* %s (%04X..%04X) */" % (block, start, end)
127 for u in range (start, end+1):
130 print " /* %04X */" % u,
133 d = data.get (u, defaults)
134 print "%10s" % ("_(%s,%s)," % (short[0][d[0]], short[1][d[1]])),
137 # Filler block, don't check occupancy
139 total = end - start + 1
140 occupancy = num * 100. / total
141 # Maintain at least 30% occupancy in the table */
143 raise Exception ("Table too sparse, please investigate: ", occupancy, block)
156 (start, end) = blocks[block]
158 if start != last + 1:
159 if start - last <= 33:
160 print_block ("FILLER", last+1, start-1, data)
167 print "static const INDIC_TABLE_ELEMENT_TYPE indic_table_0x%04x[] =" % start
169 tables.append (start)
171 print_block (block, start, end, data)
178 print "static INDIC_TABLE_ELEMENT_TYPE"
179 print "get_indic_categories (hb_codepoint_t u)"
182 t = "indic_table_0x%04x" % u
183 print " if (0x%04X <= u && u <= 0x%04X + ARRAY_LENGTH (%s)) return %s[u - 0x%04X];" % (u, u, t, t, u)
184 for u,d in singles.items ():
185 print " if (unlikely (u == 0x%04X)) return _(%s,%s);" % (u, short[0][d[0]], short[1][d[1]])
186 print " return _(X,X);"
193 vv = values[i].keys ()
196 print "#undef %s_%s" % \
197 (what_short[i], short[i][v])
201 print "/* == End of generated table == */"