CMake.git - CMake is a cross-platform, open-source build system generator.

path: root/Help/guide/tutorial/Step11/tutorial.cxx

diff options

author	Craig Scott <craig.scott@crascit.com>	2022-06-04 08:30:17 (GMT)
committer	Craig Scott <craig.scott@crascit.com>	2022-06-04 08:35:37 (GMT)
commit	8cf8f4d13b18ad28cc33dbc765a095dcf679391b (patch)
tree	0e64a06849b5ed50b6285b3e8109947980c650df /Help/guide/tutorial/Step11/tutorial.cxx
parent	cc2c5d4eb536b220421835ed2eb40729337fd53a (diff)
download	CMake-8cf8f4d13b18ad28cc33dbc765a095dcf679391b.zip CMake-8cf8f4d13b18ad28cc33dbc765a095dcf679391b.tar.gz CMake-8cf8f4d13b18ad28cc33dbc765a095dcf679391b.tar.bz2

Help: State behavior for add_custom_command(TARGET) without event type

Projects should always have specified one of PRE_BUILD, PRE_LINK or POST_BUILD, and the documentation has always shown that one must be given. But the argument parsing logic was such that if none was given, POST_BUILD would be used and no error or warning would be raised. Projects may be relying on this behavior, so document it as formally supported, but not recommended. Fixes: #23488

Diffstat (limited to 'Help/guide/tutorial/Step11/tutorial.cxx')

0 files changed, 0 insertions, 0 deletions

='n123' href='#n123'>123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144

#
# (re)generate unicode property and type databases
#
# this script converts a unicode 3.2 database file to
# Modules/unicodedata_db.h, Modules/unicodename_db.h,
# and Objects/unicodetype_db.h
#
# history:
# 2000-09-24 fl   created (based on bits and pieces from unidb)
# 2000-09-25 fl   merged tim's splitbin fixes, separate decomposition table
# 2000-09-25 fl   added character type table
# 2000-09-26 fl   added LINEBREAK, DECIMAL, and DIGIT flags/fields (2.0)
# 2000-11-03 fl   expand first/last ranges
# 2001-01-19 fl   added character name tables (2.1)
# 2001-01-21 fl   added decomp compression; dynamic phrasebook threshold
# 2002-09-11 wd   use string methods
# 2002-10-18 mvl  update to Unicode 3.2
# 2002-10-22 mvl  generate NFC tables
# 2002-11-24 mvl  expand all ranges, sort names version-independently
# 2002-11-25 mvl  add UNIDATA_VERSION
# 2004-05-29 perky add east asian width information
# 2006-03-10 mvl  update to Unicode 4.1; add UCD 3.2 delta
# 2008-06-11 gb   add PRINTABLE_MASK for Atsuo Ishimoto's ascii() patch
#
# written by Fredrik Lundh (fredrik@pythonware.com)
#

import sys

SCRIPT = sys.argv[0]
VERSION = "2.6"

# The Unicode Database
UNIDATA_VERSION = "5.1.0"
UNICODE_DATA = "UnicodeData%s.txt"
COMPOSITION_EXCLUSIONS = "CompositionExclusions%s.txt"
EASTASIAN_WIDTH = "EastAsianWidth%s.txt"
UNIHAN = "Unihan%s.txt"
DERIVED_CORE_PROPERTIES = "DerivedCoreProperties%s.txt"
DERIVEDNORMALIZATION_PROPS = "DerivedNormalizationProps%s.txt"

old_versions = ["3.2.0"]

CATEGORY_NAMES = [ "Cn", "Lu", "Ll", "Lt", "Mn", "Mc", "Me", "Nd",
    "Nl", "No", "Zs", "Zl", "Zp", "Cc", "Cf", "Cs", "Co", "Cn", "Lm",
    "Lo", "Pc", "Pd", "Ps", "Pe", "Pi", "Pf", "Po", "Sm", "Sc", "Sk",
    "So" ]

BIDIRECTIONAL_NAMES = [ "", "L", "LRE", "LRO", "R", "AL", "RLE", "RLO",
    "PDF", "EN", "ES", "ET", "AN", "CS", "NSM", "BN", "B", "S", "WS",
    "ON" ]

EASTASIANWIDTH_NAMES = [ "F", "H", "W", "Na", "A", "N" ]

# note: should match definitions in Objects/unicodectype.c
ALPHA_MASK = 0x01
DECIMAL_MASK = 0x02
DIGIT_MASK = 0x04
LOWER_MASK = 0x08
LINEBREAK_MASK = 0x10
SPACE_MASK = 0x20
TITLE_MASK = 0x40
UPPER_MASK = 0x80
XID_START_MASK = 0x100
XID_CONTINUE_MASK = 0x200
PRINTABLE_MASK = 0x400
NODELTA_MASK = 0x800
NUMERIC_MASK = 0x1000

def maketables(trace=0):

    print("--- Reading", UNICODE_DATA % "", "...")

    version = ""
    unicode = UnicodeData(UNICODE_DATA % version,
                          COMPOSITION_EXCLUSIONS % version,
                          EASTASIAN_WIDTH % version,
                          UNIHAN % version,
                          DERIVED_CORE_PROPERTIES % version,
                          DERIVEDNORMALIZATION_PROPS % version)

    print(len(list(filter(None, unicode.table))), "characters")

    for version in old_versions:
        print("--- Reading", UNICODE_DATA % ("-"+version), "...")
        old_unicode = UnicodeData(UNICODE_DATA % ("-"+version),
                                  COMPOSITION_EXCLUSIONS % ("-"+version),
                                  EASTASIAN_WIDTH % ("-"+version),
                                  UNIHAN % ("-"+version),
                                  DERIVED_CORE_PROPERTIES % ("-"+version))
        print(len(list(filter(None, old_unicode.table))), "characters")
        merge_old_version(version, unicode, old_unicode)

    makeunicodename(unicode, trace)
    makeunicodedata(unicode, trace)
    makeunicodetype(unicode, trace)

# --------------------------------------------------------------------
# unicode character properties

def makeunicodedata(unicode, trace):

    dummy = (0, 0, 0, 0, 0, 0)
    table = [dummy]
    cache = {0: dummy}
    index = [0] * len(unicode.chars)

    FILE = "Modules/unicodedata_db.h"

    print("--- Preparing", FILE, "...")

    # 1) database properties

    for char in unicode.chars:
        record = unicode.table[char]
        if record:
            # extract database properties
            category = CATEGORY_NAMES.index(record[2])
            combining = int(record[3])
            bidirectional = BIDIRECTIONAL_NAMES.index(record[4])
            mirrored = record[9] == "Y"
            eastasianwidth = EASTASIANWIDTH_NAMES.index(record[15])
            normalizationquickcheck = record[17]
            item = (
                category, combining, bidirectional, mirrored, eastasianwidth,
                normalizationquickcheck
                )
            # add entry to index and item tables
            i = cache.get(item)
            if i is None:
                cache[item] = i = len(table)
                table.append(item)
            index[char] = i

    # 2) decomposition data

    decomp_data = [0]
    decomp_prefix = [""]
    decomp_index = [0] * len(unicode.chars)
    decomp_size = 0

    comp_pairs = []
    comp_first = [None] * len(unicode.chars)
    comp_last = [None] * len(unicode.chars)

    for char in unicode.chars:
        record = unicode.table[char]
        if record:
            if record[5]:
                decomp = record[5].split()
                if len(decomp) > 19:
                    raise Exception("character %x has a decomposition too large for nfd_nfkd" % char)
                # prefix
                if decomp[0][0] == "<":
                    prefix = decomp.pop(0)
                else:
                    prefix = ""
                try:
                    i = decomp_prefix.index(prefix)
                except ValueError:
                    i = len(decomp_prefix)
                    decomp_prefix.append(prefix)
                prefix = i
                assert prefix < 256
                # content
                decomp = [prefix + (len(decomp)<<8)] + [int(s, 16) for s in decomp]
                # Collect NFC pairs
                if not prefix and len(decomp) == 3 and \
                   char not in unicode.exclusions and \
                   unicode.table[decomp[1]][3] == "0":
                    p, l, r = decomp
                    comp_first[l] = 1
                    comp_last[r] = 1
                    comp_pairs.append((l,r,char))
                try:
                    i = decomp_data.index(decomp)
                except ValueError:
                    i = len(decomp_data)
                    decomp_data.extend(decomp)
                    decomp_size = decomp_size + len(decomp) * 2
            else:
                i = 0
            decomp_index[char] = i

    f = l = 0
    comp_first_ranges = []
    comp_last_ranges = []
    prev_f = prev_l = None
    for i in unicode.chars:
        if comp_first[i] is not None:
            comp_first[i] = f
            f += 1
            if prev_f is None:
                prev_f = (i,i)
            elif prev_f[1]+1 == i:
                prev_f = prev_f[0],i
            else:
                comp_first_ranges.append(prev_f)
                prev_f = (i,i)
        if comp_last[i] is not None:
            comp_last[i] = l
            l += 1
            if prev_l is None:
                prev_l = (i,i)
            elif prev_l[1]+1 == i:
                prev_l = prev_l[0],i
            else:
                comp_last_ranges.append(prev_l)
                prev_l = (i,i)
    comp_first_ranges.append(prev_f)
    comp_last_ranges.append(prev_l)
    total_first = f
    total_last = l

    comp_data = [0]*(total_first*total_last)
    for f,l,char in comp_pairs:
        f = comp_first[f]
        l = comp_last[l]
        comp_data[f*total_last+l] = char

    print(len(table), "unique properties")
    print(len(decomp_prefix), "unique decomposition prefixes")
    print(len(decomp_data), "unique decomposition entries:", end=' ')
    print(decomp_size, "bytes")
    print(total_first, "first characters in NFC")
    print(total_last, "last characters in NFC")
    print(len(comp_pairs), "NFC pairs")

    print("--- Writing", FILE, "...")

    fp = open(FILE, "w")
    print("/* this file was generated by %s %s */" % (SCRIPT, VERSION), file=fp)
    print(file=fp)
    print('#define UNIDATA_VERSION "%s"' % UNIDATA_VERSION, file=fp)
    print("/* a list of unique database records */", file=fp)
    print("const _PyUnicode_DatabaseRecord _PyUnicode_Database_Records[] = {", file=fp)
    for item in table:
        print("    {%d, %d, %d, %d, %d, %d}," % item, file=fp)
    print("};", file=fp)
    print(file=fp)

    print("/* Reindexing of NFC first characters. */", file=fp)
    print("#define TOTAL_FIRST",total_first, file=fp)
    print("#define TOTAL_LAST",total_last, file=fp)
    print("struct reindex{int start;short count,index;};", file=fp)
    print("static struct reindex nfc_first[] = {", file=fp)
    for start,end in comp_first_ranges:
        print("  { %d, %d, %d}," % (start,end-start,comp_first[start]), file=fp)
    print("  {0,0,0}", file=fp)
    print("};\n", file=fp)
    print("static struct reindex nfc_last[] = {", file=fp)
    for start,end in comp_last_ranges:
        print("  { %d, %d, %d}," % (start,end-start,comp_last[start]), file=fp)
    print("  {0,0,0}", file=fp)
    print("};\n", file=fp)

    # FIXME: <fl> the following tables could be made static, and
    # the support code moved into unicodedatabase.c

    print("/* string literals */", file=fp)
    print("const char *_PyUnicode_CategoryNames[] = {", file=fp)
    for name in CATEGORY_NAMES:
        print("    \"%s\"," % name, file=fp)
    print("    NULL", file=fp)
    print("};", file=fp)

    print("const char *_PyUnicode_BidirectionalNames[] = {", file=fp)
    for name in BIDIRECTIONAL_NAMES:
        print("    \"%s\"," % name, file=fp)
    print("    NULL", file=fp)
    print("};", file=fp)

    print("const char *_PyUnicode_EastAsianWidthNames[] = {", file=fp)
    for name in EASTASIANWIDTH_NAMES:
        print("    \"%s\"," % name, file=fp)
    print("    NULL", file=fp)
    print("};", file=fp)

    print("static const char *decomp_prefix[] = {", file=fp)
    for name in decomp_prefix:
        print("    \"%s\"," % name, file=fp)
    print("    NULL", file=fp)
    print("};", file=fp)

    # split record index table
    index1, index2, shift = splitbins(index, trace)

    print("/* index tables for the database records */", file=fp)
    print("#define SHIFT", shift, file=fp)
    Array("index1", index1).dump(fp, trace)
    Array("index2", index2).dump(fp, trace)

    # split decomposition index table
    index1, index2, shift = splitbins(decomp_index, trace)

    print("/* decomposition data */", file=fp)
    Array("decomp_data", decomp_data).dump(fp, trace)

    print("/* index tables for the decomposition data */", file=fp)
    print("#define DECOMP_SHIFT", shift, file=fp)
    Array("decomp_index1", index1).dump(fp, trace)
    Array("decomp_index2", index2).dump(fp, trace)

    index, index2, shift = splitbins(comp_data, trace)
    print("/* NFC pairs */", file=fp)
    print("#define COMP_SHIFT", shift, file=fp)
    Array("comp_index", index).dump(fp, trace)
    Array("comp_data", index2).dump(fp, trace)

    # Generate delta tables for old versions
    for version, table, normalization in unicode.changed:
        cversion = version.replace(".","_")
        records = [table[0]]
        cache = {table[0]:0}
        index = [0] * len(table)
        for i, record in enumerate(table):
            try:
                index[i] = cache[record]
            except KeyError:
                index[i] = cache[record] = len(records)
                records.append(record)
        index1, index2, shift = splitbins(index, trace)
        print("static const change_record change_records_%s[] = {" % cversion, file=fp)
        for record in records:
            print("\t{ %s }," % ", ".join(map(str,record)), file=fp)
        print("};", file=fp)
        Array("changes_%s_index" % cversion, index1).dump(fp, trace)
        Array("changes_%s_data" % cversion, index2).dump(fp, trace)
        print("static const change_record* get_change_%s(Py_UCS4 n)" % cversion, file=fp)
        print("{", file=fp)
        print("\tint index;", file=fp)
        print("\tif (n >= 0x110000) index = 0;", file=fp)
        print("\telse {", file=fp)
        print("\t\tindex = changes_%s_index[n>>%d];" % (cversion, shift), file=fp)
        print("\t\tindex = changes_%s_data[(index<<%d)+(n & %d)];" % \
              (cversion, shift, ((1<<shift)-1)), file=fp)
        print("\t}", file=fp)
        print("\treturn change_records_%s+index;" % cversion, file=fp)
        print("}\n", file=fp)
        print("static Py_UCS4 normalization_%s(Py_UCS4 n)" % cversion, file=fp)
        print("{", file=fp)
        print("\tswitch(n) {", file=fp)
        for k, v in normalization:
            print("\tcase %s: return 0x%s;" % (hex(k), v), file=fp)
        print("\tdefault: return 0;", file=fp)
        print("\t}\n}\n", file=fp)

    fp.close()

# --------------------------------------------------------------------
# unicode character type tables

def makeunicodetype(unicode, trace):

    FILE = "Objects/unicodetype_db.h"

    print("--- Preparing", FILE, "...")

    # extract unicode types
    dummy = (0, 0, 0, 0, 0, 0)
    table = [dummy]
    cache = {0: dummy}
    index = [0] * len(unicode.chars)
    numeric = {}
    spaces = []
    linebreaks = []

    for char in unicode.chars:
        record = unicode.table[char]
        if record:
            # extract database properties
            category = record[2]
            bidirectional = record[4]
            properties = record[16]
            flags = 0
            delta = True
            if category in ["Lm", "Lt", "Lu", "Ll", "Lo"]:
                flags |= ALPHA_MASK
            if category == "Ll":
                flags |= LOWER_MASK
            if category == "Zl" or bidirectional == "B":
                flags |= LINEBREAK_MASK
                linebreaks.append(char)
            if category == "Zs" or bidirectional in ("WS", "B", "S"):
                flags |= SPACE_MASK
                spaces.append(char)
            if category == "Lt":
                flags |= TITLE_MASK
            if category == "Lu":
                flags |= UPPER_MASK
            if char == ord(" ") or category[0] not in ("C", "Z"):
                flags |= PRINTABLE_MASK
            if "XID_Start" in properties:
                flags |= XID_START_MASK
            if "XID_Continue" in properties:
                flags |= XID_CONTINUE_MASK
            # use delta predictor for upper/lower/title if it fits
            if record[12]:
                upper = int(record[12], 16)
            else:
                upper = char
            if record[13]:
                lower = int(record[13], 16)
            else:
                lower = char
            if record[14]:
                title = int(record[14], 16)
            else:
                # UCD.html says that a missing title char means that
                # it defaults to the uppercase character, not to the
                # character itself. Apparently, in the current UCD (5.x)
                # this feature is never used
                title = upper
            upper_d = upper - char
            lower_d = lower - char
            title_d = title - char
            if -32768 <= upper_d <= 32767 and \
               -32768 <= lower_d <= 32767 and \
               -32768 <= title_d <= 32767:
                # use deltas
                upper = upper_d & 0xffff
                lower = lower_d & 0xffff
                title = title_d & 0xffff
            else:
                flags |= NODELTA_MASK
            # decimal digit, integer digit
            decimal = 0
            if record[6]:
                flags |= DECIMAL_MASK
                decimal = int(record[6])
            digit = 0
            if record[7]:
                flags |= DIGIT_MASK
                digit = int(record[7])
            if record[8]:
                flags |= NUMERIC_MASK
                numeric.setdefault(record[8], []).append(char)
            item = (
                upper, lower, title, decimal, digit, flags
                )
            # add entry to index and item tables
            i = cache.get(item)
            if i is None:
                cache[item] = i = len(table)
                table.append(item)
            index[char] = i

    print(len(table), "unique character type entries")
    print(sum(map(len, numeric.values())), "numeric code points")
    print(len(spaces), "whitespace code points")
    print(len(linebreaks), "linebreak code points")

    print("--- Writing", FILE, "...")

    fp = open(FILE, "w")
    print("/* this file was generated by %s %s */" % (SCRIPT, VERSION), file=fp)
    print(file=fp)
    print("/* a list of unique character type descriptors */", file=fp)
    print("const _PyUnicode_TypeRecord _PyUnicode_TypeRecords[] = {", file=fp)
    for item in table:
        print("    {%d, %d, %d, %d, %d, %d}," % item, file=fp)
    print("};", file=fp)
    print(file=fp)

    # split decomposition index table
    index1, index2, shift = splitbins(index, trace)

    print("/* type indexes */", file=fp)
    print("#define SHIFT", shift, file=fp)
    Array("index1", index1).dump(fp, trace)
    Array("index2", index2).dump(fp, trace)

    # Generate code for _PyUnicode_ToNumeric()
    numeric_items = sorted(numeric.items())
    print('/* Returns the numeric value as double for Unicode characters', file=fp)
    print(' * having this property, -1.0 otherwise.', file=fp)
    print(' */', file=fp)
    print('double _PyUnicode_ToNumeric(Py_UNICODE ch)', file=fp)
    print('{', file=fp)
    print('    switch (ch) {', file=fp)
    for value, codepoints in numeric_items:
        haswide = False
        hasnonewide = False
        codepoints.sort()
        for codepoint in codepoints:
            if codepoint < 0x10000:
                hasnonewide = True
            if codepoint >= 0x10000 and not haswide:
                print('#ifdef Py_UNICODE_WIDE', file=fp)
                haswide = True
            print('    case 0x%04X:' % (codepoint,), file=fp)
        if haswide and hasnonewide:
            print('#endif', file=fp)
        print('        return (double) %s;' % (value,), file=fp)
        if haswide and not hasnonewide:
            print('#endif', file=fp)
    print('    }', file=fp)
    print('    return -1.0;', file=fp)
    print('}', file=fp)
    print(file=fp)

    # Generate code for _PyUnicode_IsWhitespace()
    print("/* Returns 1 for Unicode characters having the bidirectional", file=fp)
    print(" * type 'WS', 'B' or 'S' or the category 'Zs', 0 otherwise.", file=fp)
    print(" */", file=fp)
    print('int _PyUnicode_IsWhitespace(register const Py_UNICODE ch)', file=fp)
    print('{', file=fp)
    print('#ifdef WANT_WCTYPE_FUNCTIONS', file=fp)
    print('    return iswspace(ch);', file=fp)
    print('#else', file=fp)
    print('    switch (ch) {', file=fp)

    haswide = False
    hasnonewide = False
    spaces.sort()
    for codepoint in spaces:
        if codepoint < 0x10000:
            hasnonewide = True
        if codepoint >= 0x10000 and not haswide:
            print('#ifdef Py_UNICODE_WIDE', file=fp)
            haswide = True
        print('    case 0x%04X:' % (codepoint,), file=fp)
    if haswide and hasnonewide:
        print('#endif', file=fp)
    print('        return 1;', file=fp)
    if haswide and not hasnonewide:
        print('#endif', file=fp)

    print('    }', file=fp)
    print('    return 0;', file=fp)
    print('#endif', file=fp)
    print('}', file=fp)
    print(file=fp)

    # Generate code for _PyUnicode_IsLinebreak()
    print("/* Returns 1 for Unicode characters having the category 'Zl',", file=fp)
    print(" * 'Zp' or type 'B', 0 otherwise.", file=fp)
    print(" */", file=fp)
    print('int _PyUnicode_IsLinebreak(register const Py_UNICODE ch)', file=fp)
    print('{', file=fp)
    print('    switch (ch) {', file=fp)
    haswide = False
    hasnonewide = False
    linebreaks.sort()
    for codepoint in linebreaks:
        if codepoint < 0x10000:
            hasnonewide = True
        if codepoint >= 0x10000 and not haswide:
            print('#ifdef Py_UNICODE_WIDE', file=fp)
            haswide = True
        print('    case 0x%04X:' % (codepoint,), file=fp)
    if haswide and hasnonewide:
        print('#endif', file=fp)
    print('        return 1;', file=fp)
    if haswide and not hasnonewide:
        print('#endif', file=fp)

    print('    }', file=fp)
    print('    return 0;', file=fp)
    print('}', file=fp)
    print(file=fp)

    fp.close()

# --------------------------------------------------------------------
# unicode name database

def makeunicodename(unicode, trace):

    FILE = "Modules/unicodename_db.h"

    print("--- Preparing", FILE, "...")

    # collect names
    names = [None] * len(unicode.chars)

    for char in unicode.chars:
        record = unicode.table[char]
        if record:
            name = record[1].strip()
            if name and name[0] != "<":
                names[char] = name + chr(0)

    print(len(list(n for n in names if n is not None)), "distinct names")

    # collect unique words from names (note that we differ between
    # words inside a sentence, and words ending a sentence.  the
    # latter includes the trailing null byte.

    words = {}
    n = b = 0
    for char in unicode.chars:
        name = names[char]
        if name:
            w = name.split()
            b = b + len(name)
            n = n + len(w)
            for w in w:
                l = words.get(w)
                if l:
                    l.append(None)
                else:
                    words[w] = [len(words)]

    print(n, "words in text;", b, "bytes")

    wordlist = list(words.items())

    # sort on falling frequency, then by name
    def word_key(a):
        aword, alist = a
        return -len(alist), aword
    wordlist.sort(key=word_key)

    # figure out how many phrasebook escapes we need
    escapes = 0
    while escapes * 256 < len(wordlist):
        escapes = escapes + 1
    print(escapes, "escapes")

    short = 256 - escapes

    assert short > 0

    print(short, "short indexes in lexicon")

    # statistics
    n = 0
    for i in range(short):
        n = n + len(wordlist[i][1])
    print(n, "short indexes in phrasebook")

    # pick the most commonly used words, and sort the rest on falling
    # length (to maximize overlap)

    wordlist, wordtail = wordlist[:short], wordlist[short:]
    wordtail.sort(key=lambda a: a[0], reverse=True)
    wordlist.extend(wordtail)

    # generate lexicon from words

    lexicon_offset = [0]
    lexicon = ""
    words = {}

    # build a lexicon string
    offset = 0
    for w, x in wordlist:
        # encoding: bit 7 indicates last character in word (chr(128)
        # indicates the last character in an entire string)
        ww = w[:-1] + chr(ord(w[-1])+128)
        # reuse string tails, when possible
        o = lexicon.find(ww)
        if o < 0:
            o = offset
            lexicon = lexicon + ww
            offset = offset + len(w)
        words[w] = len(lexicon_offset)
        lexicon_offset.append(o)

    lexicon = list(map(ord, lexicon))

    # generate phrasebook from names and lexicon
    phrasebook = [0]
    phrasebook_offset = [0] * len(unicode.chars)
    for char in unicode.chars:
        name = names[char]
        if name:
            w = name.split()
            phrasebook_offset[char] = len(phrasebook)
            for w in w:
                i = words[w]
                if i < short:
                    phrasebook.append(i)
                else:
                    # store as two bytes
                    phrasebook.append((i>>8) + short)
                    phrasebook.append(i&255)

    assert getsize(phrasebook) == 1

    #
    # unicode name hash table

    # extract names
    data = []
    for char in unicode.chars:
        record = unicode.table[char]
        if record:
            name = record[1].strip()
            if name and name[0] != "<":
                data.append((name, char))

    # the magic number 47 was chosen to minimize the number of
    # collisions on the current data set.  if you like, change it
    # and see what happens...

    codehash = Hash("code", data, 47)

    print("--- Writing", FILE, "...")

    fp = open(FILE, "w")
    print("/* this file was generated by %s %s */" % (SCRIPT, VERSION), file=fp)
    print(file=fp)
    print("#define NAME_MAXLEN", 256, file=fp)
    print(file=fp)
    print("/* lexicon */", file=fp)
    Array("lexicon", lexicon).dump(fp, trace)
    Array("lexicon_offset", lexicon_offset).dump(fp, trace)

    # split decomposition index table
    offset1, offset2, shift = splitbins(phrasebook_offset, trace)

    print("/* code->name phrasebook */", file=fp)
    print("#define phrasebook_shift", shift, file=fp)
    print("#define phrasebook_short", short, file=fp)

    Array("phrasebook", phrasebook).dump(fp, trace)
    Array("phrasebook_offset1", offset1).dump(fp, trace)
    Array("phrasebook_offset2", offset2).dump(fp, trace)

    print("/* name->code dictionary */", file=fp)
    codehash.dump(fp, trace)

    fp.close()


def merge_old_version(version, new, old):
    # Changes to exclusion file not implemented yet
    if old.exclusions != new.exclusions:
        raise NotImplementedError("exclusions differ")

    # In these change records, 0xFF means "no change"
    bidir_changes = [0xFF]*0x110000
    category_changes = [0xFF]*0x110000
    decimal_changes = [0xFF]*0x110000
    mirrored_changes = [0xFF]*0x110000
    # In numeric data, 0 means "no change",
    # -1 means "did not have a numeric value
    numeric_changes = [0] * 0x110000
    # normalization_changes is a list of key-value pairs
    normalization_changes = []
    for i in range(0x110000):
        if new.table[i] is None:
            # Characters unassigned in the new version ought to
            # be unassigned in the old one
            assert old.table[i] is None
            continue
        # check characters unassigned in the old version
        if old.table[i] is None:
            # category 0 is "unassigned"
            category_changes[i] = 0
            continue
        # check characters that differ
        if old.table[i] != new.table[i]:
            for k in range(len(old.table[i])):
                if old.table[i][k] != new.table[i][k]:
                    value = old.table[i][k]
                    if k == 2:
                        #print "CATEGORY",hex(i), old.table[i][k], new.table[i][k]
                        category_changes[i] = CATEGORY_NAMES.index(value)
                    elif k == 4:
                        #print "BIDIR",hex(i), old.table[i][k], new.table[i][k]
                        bidir_changes[i] = BIDIRECTIONAL_NAMES.index(value)
                    elif k == 5:
                        #print "DECOMP",hex(i), old.table[i][k], new.table[i][k]
                        # We assume that all normalization changes are in 1:1 mappings
                        assert " " not in value
                        normalization_changes.append((i, value))
                    elif k == 6:
                        #print "DECIMAL",hex(i), old.table[i][k], new.table[i][k]
                        # we only support changes where the old value is a single digit
                        assert value in "0123456789"
                        decimal_changes[i] = int(value)
                    elif k == 8:
                        # print "NUMERIC",hex(i), `old.table[i][k]`, new.table[i][k]
                        # Since 0 encodes "no change", the old value is better not 0
                        if not value:
                            numeric_changes[i] = -1
                        else:
                            numeric_changes[i] = float(value)
                            assert numeric_changes[i] not in (0, -1)
                    elif k == 9:
                        if value == 'Y':
                            mirrored_changes[i] = '1'
                        else:
                            mirrored_changes[i] = '0'
                    elif k == 11:
                        # change to ISO comment, ignore
                        pass
                    elif k == 12:
                        # change to simple uppercase mapping; ignore
                        pass
                    elif k == 13:
                        # change to simple lowercase mapping; ignore
                        pass
                    elif k == 14:
                        # change to simple titlecase mapping; ignore
                        pass
                    elif k == 16:
                        # derived property changes; not yet
                        pass
                    else:
                        class Difference(Exception):pass
                        raise Difference(hex(i), k, old.table[i], new.table[i])
    new.changed.append((version, list(zip(bidir_changes, category_changes,
                                     decimal_changes, mirrored_changes,
                                     numeric_changes)),
                        normalization_changes))


# --------------------------------------------------------------------
# the following support code is taken from the unidb utilities
# Copyright (c) 1999-2000 by Secret Labs AB

# load a unicode-data file from disk

class UnicodeData:
    # Record structure:
    # [ID, name, category, combining, bidi, decomp,  (6)
    #  decimal, digit, numeric, bidi-mirrored, Unicode-1-name, (11)
    #  ISO-comment, uppercase, lowercase, titlecase, ea-width, (16)
    #  derived-props] (17)

    def __init__(self, filename, exclusions, eastasianwidth, unihan,
                 derivedprops, derivednormalizationprops=None, expand=1):
        self.changed = []
        file = open(filename)
        table = [None] * 0x110000
        while 1:
            s = file.readline()
            if not s:
                break
            s = s.strip().split(";")
            char = int(s[0], 16)
            table[char] = s

        # expand first-last ranges
        if expand:
            field = None
            for i in range(0, 0x110000):
                s = table[i]
                if s:
                    if s[1][-6:] == "First>":
                        s[1] = ""
                        field = s
                    elif s[1][-5:] == "Last>":
                        s[1] = ""
                        field = None
                elif field:
                    f2 = field[:]
                    f2[0] = "%X" % i
                    table[i] = f2

        # public attributes
        self.filename = filename
        self.table = table
        self.chars = list(range(0x110000)) # unicode 3.2

        file = open(exclusions)
        self.exclusions = {}
        for s in file:
            s = s.strip()
            if not s:
                continue
            if s[0] == '#':
                continue
            char = int(s.split()[0],16)
            self.exclusions[char] = 1

        widths = [None] * 0x110000
        for s in open(eastasianwidth):
            s = s.strip()
            if not s:
                continue
            if s[0] == '#':
                continue
            s = s.split()[0].split(';')
            if '..' in s[0]:
                first, last = [int(c, 16) for c in s[0].split('..')]
                chars = list(range(first, last+1))
            else:
                chars = [int(s[0], 16)]
            for char in chars:
                widths[char] = s[1]
        for i in range(0, 0x110000):
            if table[i] is not None:
                table[i].append(widths[i])

        for i in range(0, 0x110000):
            if table[i] is not None:
                table[i].append(set())
        for s in open(derivedprops):
            s = s.split('#', 1)[0].strip()
            if not s:
                continue

            r, p = s.split(";")
            r = r.strip()
            p = p.strip()
            if ".." in r:
                first, last = [int(c, 16) for c in r.split('..')]
                chars = list(range(first, last+1))
            else:
                chars = [int(r, 16)]
            for char in chars:
                if table[char]:
                    # Some properties (e.g. Default_Ignorable_Code_Point)
                    # apply to unassigned code points; ignore them
                    table[char][-1].add(p)

        if derivednormalizationprops:
            quickchecks = [0] * 0x110000 # default is Yes
            qc_order = 'NFD_QC NFKD_QC NFC_QC NFKC_QC'.split()
            for s in open(derivednormalizationprops):
                if '#' in s:
                    s = s[:s.index('#')]
                s = [i.strip() for i in s.split(';')]
                if len(s) < 2 or s[1] not in qc_order:
                    continue
                quickcheck = 'MN'.index(s[2]) + 1 # Maybe or No
                quickcheck_shift = qc_order.index(s[1])*2
                quickcheck <<= quickcheck_shift
                if '..' not in s[0]:
                    first = last = int(s[0], 16)
                else:
                    first, last = [int(c, 16) for c in s[0].split('..')]
                for char in range(first, last+1):
                    assert not (quickchecks[char]>>quickcheck_shift)&3
                    quickchecks[char] |= quickcheck
            for i in range(0, 0x110000):
                if table[i] is not None:
                    table[i].append(quickchecks[i])

        for line in open(unihan, encoding='utf-8'):
            if not line.startswith('U+'):
                continue
            code, tag, value = line.split(None, 3)[:3]
            if tag not in ('kAccountingNumeric', 'kPrimaryNumeric',
                           'kOtherNumeric'):
                continue
            value = value.strip().replace(',', '')
            i = int(code[2:], 16)
            # Patch the numeric field
            if table[i] is not None:
                table[i][8] = value

    def uselatin1(self):
        # restrict character range to ISO Latin 1
        self.chars = list(range(256))

# hash table tools

# this is a straight-forward reimplementation of Python's built-in
# dictionary type, using a static data structure, and a custom string
# hash algorithm.

def myhash(s, magic):
    h = 0
    for c in map(ord, s.upper()):
        h = (h * magic) + c
        ix = h & 0xff000000
        if ix:
            h = (h ^ ((ix>>24) & 0xff)) & 0x00ffffff
    return h

SIZES = [
    (4,3), (8,3), (16,3), (32,5), (64,3), (128,3), (256,29), (512,17),
    (1024,9), (2048,5), (4096,83), (8192,27), (16384,43), (32768,3),
    (65536,45), (131072,9), (262144,39), (524288,39), (1048576,9),
    (2097152,5), (4194304,3), (8388608,33), (16777216,27)
]

class Hash:
    def __init__(self, name, data, magic):
        # turn a (key, value) list into a static hash table structure

        # determine table size
        for size, poly in SIZES:
            if size > len(data):
                poly = size + poly
                break
        else:
            raise AssertionError("ran out of polynominals")

        print(size, "slots in hash table")

        table = [None] * size

        mask = size-1

        n = 0

        hash = myhash

        # initialize hash table
        for key, value in data:
            h = hash(key, magic)
            i = (~h) & mask
            v = table[i]
            if v is None:
                table[i] = value
                continue
            incr = (h ^ (h >> 3)) & mask;
            if not incr:
                incr = mask
            while 1:
                n = n + 1
                i = (i + incr) & mask
                v = table[i]
                if v is None:
                    table[i] = value
                    break
                incr = incr << 1
                if incr > mask:
                    incr = incr ^ poly

        print(n, "collisions")
        self.collisions = n

        for i in range(len(table)):
            if table[i] is None:
                table[i] = 0

        self.data = Array(name + "_hash", table)
        self.magic = magic
        self.name = name
        self.size = size
        self.poly = poly

    def dump(self, file, trace):
        # write data to file, as a C array
        self.data.dump(file, trace)
        file.write("#define %s_magic %d\n" % (self.name, self.magic))
        file.write("#define %s_size %d\n" % (self.name, self.size))
        file.write("#define %s_poly %d\n" % (self.name, self.poly))

# stuff to deal with arrays of unsigned integers

class Array:

    def __init__(self, name, data):
        self.name = name
        self.data = data

    def dump(self, file, trace=0):
        # write data to file, as a C array
        size = getsize(self.data)
        if trace:
            print(self.name+":", size*len(self.data), "bytes", file=sys.stderr)
        file.write("static ")
        if size == 1:
            file.write("unsigned char")
        elif size == 2:
            file.write("unsigned short")
        else:
            file.write("unsigned int")
        file.write(" " + self.name + "[] = {\n")
        if self.data:
            s = "    "
            for item in self.data:
                i = str(item) + ", "
                if len(s) + len(i) > 78:
                    file.write(s + "\n")
                    s = "    " + i
                else:
                    s = s + i
            if s.strip():
                file.write(s + "\n")
        file.write("};\n\n")

def getsize(data):
    # return smallest possible integer size for the given array
    maxdata = max(data)
    if maxdata < 256:
        return 1
    elif maxdata < 65536:
        return 2
    else:
        return 4

def splitbins(t, trace=0):
    """t, trace=0 -> (t1, t2, shift).  Split a table to save space.

    t is a sequence of ints.  This function can be useful to save space if
    many of the ints are the same.  t1 and t2 are lists of ints, and shift
    is an int, chosen to minimize the combined size of t1 and t2 (in C
    code), and where for each i in range(len(t)),
        t[i] == t2[(t1[i >> shift] << shift) + (i & mask)]
    where mask is a bitmask isolating the last "shift" bits.

    If optional arg trace is non-zero (default zero), progress info
    is printed to sys.stderr.  The higher the value, the more info
    you'll get.
    """

    if trace:
        def dump(t1, t2, shift, bytes):
            print("%d+%d bins at shift %d; %d bytes" % (
                len(t1), len(t2), shift, bytes), file=sys.stderr)
        print("Size of original table:", len(t)*getsize(t), \
                            "bytes", file=sys.stderr)
    n = len(t)-1    # last valid index
    maxshift = 0    # the most we can shift n and still have something left
    if n > 0:
        while n >> 1:
            n >>= 1
            maxshift += 1
    del n
    bytes = sys.maxsize  # smallest total size so far
    t = tuple(t)    # so slices can be dict keys
    for shift in range(maxshift + 1):
        t1 = []
        t2 = []
        size = 2**shift
        bincache = {}
        for i in range(0, len(t), size):
            bin = t[i:i+size]
            index = bincache.get(bin)
            if index is None:
                index = len(t2)
                bincache[bin] = index
                t2.extend(bin)
            t1.append(index >> shift)
        # determine memory size
        b = len(t1)*getsize(t1) + len(t2)*getsize(t2)
        if trace > 1:
            dump(t1, t2, shift, b)
        if b < bytes:
            best = t1, t2, shift
            bytes = b
    t1, t2, shift = best
    if trace:
        print("Best:", end=' ', file=sys.stderr)
        dump(t1, t2, shift, bytes)
    if __debug__:
        # exhaustively verify that the decomposition is correct
        mask = ~((~0) << shift) # i.e., low-bit mask of shift bits
        for i in range(len(t)):
            assert t[i] == t2[(t1[i >> shift] << shift) + (i & mask)]
    return best

if __name__ == "__main__":
    maketables(1)


context:
space:
mode: