print (" * on file with this header:")
print (" *")
for l in header:
print (" * %s" % (l.strip()))
print (" */")
print ()
print ("#ifndef HB_UNICODE_EMOJI_TABLE_HH")
print ("#define HB_UNICODE_EMOJI_TABLE_HH")
print ()
print ('#include "hb-unicode.hh"')
print ()
for typ, s in ranges.items():
if typ != "Extended_Pictographic": continue
arr = dict()
for start,end in s:
for i in range(start,end):
arr[i] = 1
sol = packTab.pack_table(arr, 0, compression=3)
code = packTab.Code('_hb_emoji')
sol.genCode(code, 'is_'+typ)
code.print_c(linkage='static inline')
print()
print ()
print ("#endif /* HB_UNICODE_EMOJI_TABLE_HH */")
print ()
print ("/* == End of generated table == */")
print("/* == Start of generated table == */")
print("/*")
print(" * The following table is generated by running:")
print(" *")
print(" * ./gen-ucd-table.py ucd.nounihan.grouped.xml")
print(" *")
print(" * on file with this description:", ucdxml.description)
print(" */")
print()
print("#ifndef HB_UCD_TABLE_HH")
print("#define HB_UCD_TABLE_HH")
print()
print('#include "hb.hh"')
print()
code = packTab.Code('_hb_ucd')
sc_array, _ = code.addArray('hb_script_t', 'sc_map', sc_array)
dm1_p0_array, _ = code.addArray('uint16_t', 'dm1_p0_map', dm1_p0_array)
dm1_p2_array, _ = code.addArray('uint16_t', 'dm1_p2_map', dm1_p2_array)
dm2_u32_array, _ = code.addArray('uint32_t', 'dm2_u32_map', dm2_u32_array)
dm2_u64_array, _ = code.addArray('uint64_t', 'dm2_u64_map', dm2_u64_array)
code.print_c(linkage='static inline')
datasets = [
('gc', gc, 'Cn', gc_order),
('ccc', ccc, 0, None),
('bmg', bmg, 0, None),
('sc', sc, 'Zzzz', sc_order),
('dm', dm, None, dm_order),
]
print()
print("extern const ucdb_entry UCDB_TABLE[] = {")
for script, bclass, lbreak, zspace, cbreak, paired in records:
print(
f" {{ UCDB_SCRIPT_{script.upper()}, UCDB_BIDI_{bclass}, UCDB_LBREAK_{lbreak}, {'true' if zspace else 'false'}, UCDB_CBREAK_{cbreak.upper()}, {'true' if paired else 'false'} }},"
)
print("};")
print()
# Use packTab to build an index mapping a character to a record.
solutions = packTab.pack_table(data, 0, None)
solution = packTab.pick_solution(solutions, compression=1)
code = packTab.Code('ucdb')
expr = solution.genCode(code, 'index')
code.print_c()
print()
else:
# Build enums
cbreak_enum = {record[4] for record in records}
print()
bclass_enum = {record[1].upper() for record in records}
bclass_enum.remove("L")
bclass_enum.remove("R")
import packTab
print("/* == Start of generated table == */")
print("/*")
print(" * The following table is generated by running:")
print(" *")
print(" * ./gen-arabic-pua.py")
print(" *")
print(" */")
print()
print("#ifndef HB_OT_SHAPER_ARABIC_PUA_HH")
print("#define HB_OT_SHAPER_ARABIC_PUA_HH")
print()
code = packTab.Code('_hb_arabic')
for p in ("ArabicPUASimplified.txt", "ArabicPUATraditional.txt"):
with open(p, encoding='utf-8') as f:
fields = [l.split('\t') for l in f if l[:1] != '#']
data = {int(fs[1], 16): int(fs[0], 16) for fs in fields}
sol = packTab.pack_table(data, compression=9)
sol.genCode(code, f'pua_{p[9:13].lower()}_map')
code.print_c(linkage='static inline')
print()
print("#endif /* HB_OT_SHAPER_ARABIC_PUA_HH */")
print()
print("/* == End of generated table == */")
print('#pragma GCC diagnostic push')
print('#pragma GCC diagnostic ignored "-Wunused-macros"')
for k, v in sorted(use_mapping.items()):
if k in use_positions and use_positions[k]: continue
print("#define %s USE(%s) /* %s */" % (k, k, v.__name__[3:]))
for k, v in sorted(use_positions.items()):
if not v: continue
for suf in v.keys():
tag = k + suf
print("#define %s USE(%s)" % (tag, tag))
print('#pragma GCC diagnostic pop')
print("")
import packTab
data = {u: v[0] for u, v in use_data.items()}
code = packTab.Code('hb_use')
sol = packTab.pack_table(data, compression=5, default='O')
sol.genCode(code, f'get_category')
code.print_c(linkage='static inline')
print()
for k in sorted(use_mapping.keys()):
if k in use_positions and use_positions[k]: continue
print("#undef %s" % k)
for k, v in sorted(use_positions.items()):
if not v: continue
for suf in v.keys():
tag = k + suf
print("#undef %s" % tag)
print()
print(r"""
