DXR is a code search and navigation tool aimed at making sense of large projects. It supports full-text and regex searches as well as structural queries.

Mercurial (d38398e5144e)

VCS Links

Line Code
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 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
#!/usr/bin/env python
# -*- coding: utf-8 -*-
#
# This Source Code Form is subject to the terms of the Mozilla Public
# License, v. 2.0. If a copy of the MPL was not distributed with this
# file, You can obtain one at http://mozilla.org/MPL/2.0/.

""" Usage:
    make_intl_data.py langtags [language-subtag-registry.txt]
    make_intl_data.py tzdata
    make_intl_data.py currency

    Target "langtags":
    This script extracts information about mappings between deprecated and
    current BCP 47 language tags from the IANA Language Subtag Registry and
    converts it to JavaScript object definitions in IntlData.js. The definitions
    are used in Intl.js.

    The IANA Language Subtag Registry is imported from
    https://www.iana.org/assignments/language-subtag-registry
    and uses the syntax specified in
    https://tools.ietf.org/html/rfc5646#section-3


    Target "tzdata":
    This script computes which time zone informations are not up-to-date in ICU
    and provides the necessary mappings to workaround this problem.
    https://ssl.icu-project.org/trac/ticket/12044


    Target "currency":
    Generates the mapping from currency codes to decimal digits used for them.
"""

from __future__ import print_function
import os
import re
import io
import codecs
import sys
import tarfile
import tempfile
import urllib2
import urlparse
from contextlib import closing
from functools import partial
from itertools import chain, groupby, ifilter, ifilterfalse, imap, tee
from operator import attrgetter, itemgetter

def readRegistryRecord(registry):
    """ Yields the records of the IANA Language Subtag Registry as dictionaries. """
    record = {}
    for line in registry:
        line = line.strip()
        if line == "":
            continue
        if line == "%%":
            yield record
            record = {}
        else:
            if ":" in line:
                key, value = line.split(":", 1)
                key, value = key.strip(), value.strip()
                record[key] = value
            else:
                # continuation line
                record[key] += " " + line
    if record:
        yield record
    return


def readRegistry(registry):
    """ Reads IANA Language Subtag Registry and extracts information for Intl.js.

        Information extracted:
        - langTagMappings: mappings from complete language tags to preferred
          complete language tags
        - langSubtagMappings: mappings from subtags to preferred subtags
        - extlangMappings: mappings from extlang subtags to preferred subtags,
          with prefix to be removed
        Returns these three mappings as dictionaries, along with the registry's
        file date.

        We also check that mappings for language subtags don't affect extlang
        subtags and vice versa, so that CanonicalizeLanguageTag doesn't have
        to separate them for processing. Region codes are separated by case,
        and script codes by length, so they're unproblematic.
    """
    langTagMappings = {}
    langSubtagMappings = {}
    extlangMappings = {}
    languageSubtags = set()
    extlangSubtags = set()

    for record in readRegistryRecord(registry):
        if "File-Date" in record:
            fileDate = record["File-Date"]
            continue

        if record["Type"] == "grandfathered":
            # Grandfathered tags don't use standard syntax, so
            # CanonicalizeLanguageTag expects the mapping table to provide
            # the final form for all.
            # For langTagMappings, keys must be in lower case; values in
            # the case used in the registry.
            tag = record["Tag"]
            if "Preferred-Value" in record:
                langTagMappings[tag.lower()] = record["Preferred-Value"]
            else:
                langTagMappings[tag.lower()] = tag
        elif record["Type"] == "redundant":
            # For langTagMappings, keys must be in lower case; values in
            # the case used in the registry.
            if "Preferred-Value" in record:
                langTagMappings[record["Tag"].lower()] = record["Preferred-Value"]
        elif record["Type"] in ("language", "script", "region", "variant"):
            # For langSubtagMappings, keys and values must be in the case used
            # in the registry.
            subtag = record["Subtag"]
            if record["Type"] == "language":
                languageSubtags.add(subtag)
            if "Preferred-Value" in record:
                if subtag == "heploc":
                    # The entry for heploc is unique in its complexity; handle
                    # it as special case below.
                    continue
                if "Prefix" in record:
                    # This might indicate another heploc-like complex case.
                    raise Exception("Please evaluate: subtag mapping with prefix value.")
                langSubtagMappings[subtag] = record["Preferred-Value"]
        elif record["Type"] == "extlang":
            # For extlangMappings, keys must be in the case used in the
            # registry; values are records with the preferred value and the
            # prefix to be removed.
            subtag = record["Subtag"]
            extlangSubtags.add(subtag)
            if "Preferred-Value" in record:
                preferred = record["Preferred-Value"]
                prefix = record["Prefix"]
                extlangMappings[subtag] = {"preferred": preferred, "prefix": prefix}
        else:
            # No other types are allowed by
            # https://tools.ietf.org/html/rfc5646#section-3.1.3
            assert False, "Unrecognized Type: {0}".format(record["Type"])

    # Check that mappings for language subtags and extlang subtags don't affect
    # each other.
    for lang in languageSubtags:
        if lang in extlangMappings and extlangMappings[lang]["preferred"] != lang:
            raise Exception("Conflict: lang with extlang mapping: " + lang)
    for extlang in extlangSubtags:
        if extlang in langSubtagMappings:
            raise Exception("Conflict: extlang with lang mapping: " + extlang)

    # Special case for heploc.
    langTagMappings["ja-latn-hepburn-heploc"] = "ja-Latn-alalc97"

    return {"fileDate": fileDate,
            "langTagMappings": langTagMappings,
            "langSubtagMappings": langSubtagMappings,
            "extlangMappings": extlangMappings}


def writeMappingsVar(intlData, dict, name, description, fileDate, url):
    """ Writes a variable definition with a mapping table to file intlData.

        Writes the contents of dictionary dict to file intlData with the given
        variable name and a comment with description, fileDate, and URL.
    """
    intlData.write("\n")
    intlData.write("// {0}.\n".format(description))
    intlData.write("// Derived from IANA Language Subtag Registry, file date {0}.\n".format(fileDate))
    intlData.write("// {0}\n".format(url))
    intlData.write("var {0} = {{\n".format(name))
    keys = sorted(dict)
    for key in keys:
        if isinstance(dict[key], basestring):
            value = '"{0}"'.format(dict[key])
        else:
            preferred = dict[key]["preferred"]
            prefix = dict[key]["prefix"]
            value = '{{preferred: "{0}", prefix: "{1}"}}'.format(preferred, prefix)
        intlData.write('    "{0}": {1},\n'.format(key, value))
    intlData.write("};\n")


def writeLanguageTagData(intlData, fileDate, url, langTagMappings, langSubtagMappings, extlangMappings):
    """ Writes the language tag data to the Intl data file. """
    writeMappingsVar(intlData, langTagMappings, "langTagMappings",
                     "Mappings from complete tags to preferred values", fileDate, url)
    writeMappingsVar(intlData, langSubtagMappings, "langSubtagMappings",
                     "Mappings from non-extlang subtags to preferred values", fileDate, url)
    writeMappingsVar(intlData, extlangMappings, "extlangMappings",
                     "Mappings from extlang subtags to preferred values", fileDate, url)

def updateLangTags(args):
    """ Update the IntlData.js file. """
    url = args.url
    out = args.out
    filename = args.file

    print("Arguments:")
    print("\tDownload url: %s" % url)
    print("\tLocal registry: %s" % filename)
    print("\tOutput file: %s" % out)
    print("")

    if filename is not None:
        print("Always make sure you have the newest language-subtag-registry.txt!")
        registry = codecs.open(filename, "r", encoding="utf-8")
    else:
        print("Downloading IANA Language Subtag Registry...")
        with closing(urllib2.urlopen(url)) as reader:
            text = reader.read().decode("utf-8")
        registry = codecs.open("language-subtag-registry.txt", "w+", encoding="utf-8")
        registry.write(text)
        registry.seek(0)

    print("Processing IANA Language Subtag Registry...")
    with closing(registry) as reg:
        data = readRegistry(reg)
    fileDate = data["fileDate"]
    langTagMappings = data["langTagMappings"]
    langSubtagMappings = data["langSubtagMappings"]
    extlangMappings = data["extlangMappings"]

    print("Writing Intl data...")
    with codecs.open(out, "w", encoding="utf-8") as intlData:
        intlData.write("// Generated by make_intl_data.py. DO NOT EDIT.\n")
        writeLanguageTagData(intlData, fileDate, url, langTagMappings, langSubtagMappings, extlangMappings)

def flines(filepath, encoding="utf-8"):
    """ Open filepath and iterate over its content. """
    with io.open(filepath, mode="r", encoding=encoding) as f:
        for line in f:
            yield line

class Zone:
    """ Time zone with optional file name. """

    def __init__(self, name, filename=""):
        self.name = name
        self.filename = filename
    def __eq__(self, other):
        return hasattr(other, "name") and self.name == other.name
    def __cmp__(self, other):
        if self.name == other.name:
            return 0
        if self.name < other.name:
            return -1
        return 1
    def __hash__(self):
        return hash(self.name)
    def __str__(self):
        return self.name
    def __repr__(self):
        return self.name

class TzDataDir:
    """ tzdata source from a directory. """

    def __init__(self, obj):
        self.name = partial(os.path.basename, obj)
        self.resolve = partial(os.path.join, obj)
        self.basename = os.path.basename
        self.isfile = os.path.isfile
        self.listdir = partial(os.listdir, obj)
        self.readlines = flines

class TzDataFile:
    """ tzdata source from a file (tar or gzipped). """

    def __init__(self, obj):
        self.name = lambda: os.path.splitext(os.path.splitext(os.path.basename(obj))[0])[0]
        self.resolve = obj.getmember
        self.basename = attrgetter("name")
        self.isfile = tarfile.TarInfo.isfile
        self.listdir = obj.getnames
        self.readlines = partial(self._tarlines, obj)

    def _tarlines(self, tar, m):
        with closing(tar.extractfile(m)) as f:
            for line in codecs.EncodedFile(f, "utf-8"):
                yield line

def validateTimeZones(zones, links):
    """ Validate the zone and link entries. """
    linkZones = set(links.viewkeys())
    intersect = linkZones.intersection(zones)
    if intersect:
        raise RuntimeError("Links also present in zones: %s" % intersect)

    zoneNames = set(z.name for z in zones)
    linkTargets = set(links.viewvalues())
    if not linkTargets.issubset(zoneNames):
        raise RuntimeError("Link targets not found: %s" % linkTargets.difference(zoneNames))

def partition(iterable, *predicates):
    def innerPartition(pred, it):
        it1, it2 = tee(it)
        return (ifilter(pred, it1), ifilterfalse(pred, it2))
    if len(predicates) == 0:
        return iterable
    (left, right) = innerPartition(predicates[0], iterable)
    if len(predicates) == 1:
        return (left, right)
    return tuple([left] + list(partition(right, *predicates[1:])))

def listIANAFiles(tzdataDir):
    def isTzFile(d, m, f):
        return m(f) and d.isfile(d.resolve(f))
    return ifilter(partial(isTzFile, tzdataDir, re.compile("^[a-z0-9]+$").match), tzdataDir.listdir())

def readIANAFiles(tzdataDir, files):
    """ Read all IANA time zone files from the given iterable. """
    nameSyntax = "[\w/+\-]+"
    pZone = re.compile(r"Zone\s+(?P<name>%s)\s+.*" % nameSyntax)
    pLink = re.compile(r"Link\s+(?P<target>%s)\s+(?P<name>%s)(?:\s+#.*)?" % (nameSyntax, nameSyntax))

    def createZone(line, fname):
        match = pZone.match(line)
        name = match.group("name")
        return Zone(name, fname)

    def createLink(line, fname):
        match = pLink.match(line)
        (name, target) = match.group("name", "target")
        return (Zone(name, fname), target)

    zones = set()
    links = dict()
    for filename in files:
        filepath = tzdataDir.resolve(filename)
        for line in tzdataDir.readlines(filepath):
            if line.startswith("Zone"):
                zones.add(createZone(line, filename))
            if line.startswith("Link"):
                (link, target) = createLink(line, filename)
                links[link] = target

    return (zones, links)

def readIANATimeZones(tzdataDir, ignoreBackzone, ignoreFactory):
    """ Read the IANA time zone information from `tzdataDir`. """

    backzoneFiles = {"backzone"}
    (bkfiles, tzfiles) = partition(listIANAFiles(tzdataDir), backzoneFiles.__contains__)

    # Read zone and link infos.
    (zones, links) = readIANAFiles(tzdataDir, tzfiles)
    (backzones, backlinks) = readIANAFiles(tzdataDir, bkfiles)

    # Remove the placeholder time zone "Factory".
    if ignoreFactory:
        zones.remove(Zone("Factory"))

    # Merge with backzone data.
    if not ignoreBackzone:
        zones |= backzones
        links = {name: target for name, target in links.iteritems() if name not in backzones}
        links.update(backlinks)

    validateTimeZones(zones, links)

    return (zones, links)

def readICUResourceFile(filename):
    """ Read an ICU resource file.

        Yields (<table-name>, <startOrEnd>, <value>) for each table.
    """

    numberValue = r"-?\d+"
    stringValue = r'".+?"'
    asVector = lambda val: r"%s(?:\s*,\s*%s)*" % (val, val)
    numberVector = asVector(numberValue)
    stringVector = asVector(stringValue)

    reNumberVector = re.compile(numberVector)
    reStringVector = re.compile(stringVector)
    reNumberValue = re.compile(numberValue)
    reStringValue = re.compile(stringValue)
    def parseValue(value):
        m = reNumberVector.match(value)
        if m:
            return [int(v) for v in reNumberValue.findall(value)]
        m = reStringVector.match(value)
        if m:
            return [v[1:-1] for v in reStringValue.findall(value)]
        raise RuntimeError("unknown value type: %s" % value)

    def extractValue(values):
        if len(values) == 0:
            return None
        if len(values) == 1:
            return values[0]
        return values

    def line(*args):
        maybeMultiComments = r"(?:/\*[^*]*\*/)*"
        maybeSingleComment = r"(?://.*)?"
        lineStart = "^%s" % maybeMultiComments
        lineEnd = "%s\s*%s$" % (maybeMultiComments, maybeSingleComment)
        return re.compile(r"\s*".join(chain([lineStart], args, [lineEnd])))

    tableName = r'(?P<quote>"?)(?P<name>.+?)(?P=quote)'
    tableValue = r"(?P<value>%s|%s)" % (numberVector, stringVector)

    reStartTable = line(tableName, r"\{")
    reEndTable = line(r"\}")
    reSingleValue = line(r",?", tableValue, r",?")
    reCompactTable = line(tableName, r"\{", tableValue, r"\}")
    reEmptyLine = line()

    tables = []
    currentTable = lambda: "|".join(tables)
    values = []
    for line in flines(filename, "utf-8-sig"):
        line = line.strip()
        if line == "":
            continue

        m = reEmptyLine.match(line)
        if m:
            continue

        m = reStartTable.match(line)
        if m:
            assert len(values) == 0
            tables.append(m.group("name"))
            continue

        m = reEndTable.match(line)
        if m:
            yield (currentTable(), extractValue(values))
            tables.pop()
            values = []
            continue

        m = reCompactTable.match(line)
        if m:
            assert len(values) == 0
            tables.append(m.group("name"))
            yield (currentTable(), extractValue(parseValue(m.group("value"))))
            tables.pop()
            continue

        m = reSingleValue.match(line)
        if m and tables:
            values.extend(parseValue(m.group("value")))
            continue

        raise RuntimeError("unknown entry: %s" % line)

def readICUTimeZonesFromTimezoneTypes(icuTzDir):
    """ Read the ICU time zone information from `icuTzDir`/timezoneTypes.txt
        and returns the tuple (zones, links).
    """
    typeMapTimeZoneKey = "timezoneTypes:table(nofallback)|typeMap|timezone|"
    typeAliasTimeZoneKey = "timezoneTypes:table(nofallback)|typeAlias|timezone|"
    toTimeZone = lambda name: Zone(name.replace(":", "/"))

    zones = set()
    links = dict()

    for name, value in readICUResourceFile(os.path.join(icuTzDir, "timezoneTypes.txt")):
        if name.startswith(typeMapTimeZoneKey):
            zones.add(toTimeZone(name[len(typeMapTimeZoneKey):]))
        if name.startswith(typeAliasTimeZoneKey):
            links[toTimeZone(name[len(typeAliasTimeZoneKey):])] = value

    # Remove the ICU placeholder time zone "Etc/Unknown".
    zones.remove(Zone("Etc/Unknown"))

    validateTimeZones(zones, links)

    return (zones, links)

def readICUTimeZonesFromZoneInfo(icuTzDir, ignoreFactory):
    """ Read the ICU time zone information from `icuTzDir`/zoneinfo64.txt
        and returns the tuple (zones, links).
    """
    zoneKey = "zoneinfo64:table(nofallback)|Zones:array|:table"
    linkKey = "zoneinfo64:table(nofallback)|Zones:array|:int"
    namesKey = "zoneinfo64:table(nofallback)|Names"

    tzId = 0
    tzLinks = dict()
    tzNames = []

    for name, value in readICUResourceFile(os.path.join(icuTzDir, "zoneinfo64.txt")):
        if name == zoneKey:
            tzId += 1
        elif name == linkKey:
            tzLinks[tzId] = int(value)
            tzId += 1
        elif name == namesKey:
            tzNames.extend(value)

    links = dict((Zone(tzNames[zone]), tzNames[target]) for (zone, target) in tzLinks.iteritems())
    zones = set([Zone(v) for v in tzNames if Zone(v) not in links])

    # Remove the ICU placeholder time zone "Etc/Unknown".
    zones.remove(Zone("Etc/Unknown"))

    # Remove the placeholder time zone "Factory".
    if ignoreFactory:
        zones.remove(Zone("Factory"))

    validateTimeZones(zones, links)

    return (zones, links)

def readICUTimeZones(icuDir, icuTzDir, ignoreFactory):
    # zoneinfo64.txt contains the supported time zones by ICU. This data is
    # generated from tzdata files, it doesn't include "backzone" in stock ICU.
    (zoneinfoZones, zoneinfoLinks) = readICUTimeZonesFromZoneInfo(icuTzDir, ignoreFactory)

    # timezoneTypes.txt contains the canonicalization information for ICU. This
    # data is generated from CLDR files. It includes data about time zones from
    # tzdata's "backzone" file.
    (typesZones, typesLinks) = readICUTimeZonesFromTimezoneTypes(icuTzDir)

    # Information in zoneinfo64 should be a superset of timezoneTypes.
    inZoneInfo64 = lambda zone: zone in zoneinfoZones or zone in zoneinfoLinks

    # Remove legacy ICU time zones from zoneinfo64 data.
    (legacyZones, legacyLinks) = readICULegacyZones(icuDir)
    zoneinfoZones = set(zone for zone in zoneinfoZones if zone not in legacyZones)
    zoneinfoLinks = dict((zone, target) for (zone, target) in zoneinfoLinks.iteritems() if zone not in legacyLinks)

    notFoundInZoneInfo64 = [zone for zone in typesZones if not inZoneInfo64(zone)]
    if notFoundInZoneInfo64:
        raise RuntimeError("Missing time zones in zoneinfo64.txt: %s" % notFoundInZoneInfo64)

    notFoundInZoneInfo64 = [zone for zone in typesLinks.iterkeys() if not inZoneInfo64(zone)]
    if notFoundInZoneInfo64:
        raise RuntimeError("Missing time zones in zoneinfo64.txt: %s" % notFoundInZoneInfo64)

    # zoneinfo64.txt only defines the supported time zones by ICU, the canonicalization
    # rules are defined through timezoneTypes.txt. Merge both to get the actual zones
    # and links used by ICU.
    icuZones = set(chain(
                 (zone for zone in zoneinfoZones if zone not in typesLinks),
                 (zone for zone in typesZones)
               ))
    icuLinks = dict(chain(
                 ((zone, target) for (zone, target) in zoneinfoLinks.iteritems() if zone not in typesZones),
                 ((zone, target) for (zone, target) in typesLinks.iteritems())
               ))

    return (icuZones, icuLinks)


def readICULegacyZones(icuDir):
    """ Read the ICU legacy time zones from `icuTzDir`/tools/tzcode/icuzones
        and returns the tuple (zones, links).
    """
    tzdir = TzDataDir(os.path.join(icuDir, "tools/tzcode"))
    (zones, links) = readIANAFiles(tzdir, ["icuzones"])

    # Remove the ICU placeholder time zone "Etc/Unknown".
    zones.remove(Zone("Etc/Unknown"))

    return (zones, links)

def icuTzDataVersion(icuTzDir):
    """ Read the ICU time zone version from `icuTzDir`/zoneinfo64.txt. """
    def searchInFile(pattern, f):
        p = re.compile(pattern)
        for line in flines(f, "utf-8-sig"):
            m = p.search(line)
            if m:
                return m.group(1)
        return None

    zoneinfo = os.path.join(icuTzDir, "zoneinfo64.txt")
    if not os.path.isfile(zoneinfo):
        raise RuntimeError("file not found: %s" % zoneinfo)
    version = searchInFile("^//\s+tz version:\s+([0-9]{4}[a-z])$", zoneinfo)
    if version is None:
        raise RuntimeError("%s does not contain a valid tzdata version string" % zoneinfo)
    return version

def findIncorrectICUZones(ianaZones, ianaLinks, icuZones, icuLinks, ignoreBackzone):
    """ Find incorrect ICU zone entries. """
    isIANATimeZone = lambda zone: zone in ianaZones or zone in ianaLinks
    isICUTimeZone = lambda zone: zone in icuZones or zone in icuLinks
    isICULink = lambda zone: zone in icuLinks

    # All IANA zones should be present in ICU.
    missingTimeZones = [zone for zone in ianaZones if not isICUTimeZone(zone)]
    # Normally zones in backzone are also present as links in one of the other
    # time zone files. The only exception to this rule is the Asia/Hanoi time
    # zone, this zone is only present in the backzone file.
    expectedMissing = [] if ignoreBackzone else [Zone("Asia/Hanoi")]
    if missingTimeZones != expectedMissing:
        raise RuntimeError("Not all zones are present in ICU, did you forget "
                           "to run intl/update-tzdata.sh? %s" % missingTimeZones)

    # Zones which are only present in ICU?
    additionalTimeZones = [zone for zone in icuZones if not isIANATimeZone(zone)]
    if additionalTimeZones:
        raise RuntimeError("Additional zones present in ICU, did you forget "
                           "to run intl/update-tzdata.sh? %s" % additionalTimeZones)

    # Zones which are marked as links in ICU.
    result = ((zone, icuLinks[zone]) for zone in ianaZones if isICULink(zone))

    # Remove unnecessary UTC mappings.
    utcnames = ["Etc/UTC", "Etc/UCT", "Etc/GMT"]
    result = ifilterfalse(lambda (zone, target): zone.name in utcnames, result)

    return sorted(result, key=itemgetter(0))

def findIncorrectICULinks(ianaZones, ianaLinks, icuZones, icuLinks):
    """ Find incorrect ICU link entries. """
    isIANATimeZone = lambda zone: zone in ianaZones or zone in ianaLinks
    isICUTimeZone = lambda zone: zone in icuZones or zone in icuLinks
    isICULink = lambda zone: zone in icuLinks
    isICUZone = lambda zone: zone in icuZones

    # All links should be present in ICU.
    missingTimeZones = [zone for zone in ianaLinks.iterkeys() if not isICUTimeZone(zone)]
    if missingTimeZones:
        raise RuntimeError("Not all zones are present in ICU, did you forget "
                           "to run intl/update-tzdata.sh? %s" % missingTimeZones)

    # Links which are only present in ICU?
    additionalTimeZones = [zone for zone in icuLinks.iterkeys() if not isIANATimeZone(zone)]
    if additionalTimeZones:
        raise RuntimeError("Additional links present in ICU, did you forget "
                           "to run intl/update-tzdata.sh? %s" % additionalTimeZones)

    result = chain(
        # IANA links which have a different target in ICU.
        ((zone, target, icuLinks[zone]) for (zone, target) in ianaLinks.iteritems() if isICULink(zone) and target != icuLinks[zone]),

        # IANA links which are zones in ICU.
        ((zone, target, zone.name) for (zone, target) in ianaLinks.iteritems() if isICUZone(zone))
    )

    # Remove unnecessary UTC mappings.
    utcnames = ["Etc/UTC", "Etc/UCT", "Etc/GMT"]
    result = ifilterfalse(lambda (zone, target, icuTarget): target in utcnames and icuTarget in utcnames, result)

    return sorted(result, key=itemgetter(0))

generatedFileWarning = u"// Generated by make_intl_data.py. DO NOT EDIT."
tzdataVersionComment = u"// tzdata version = {0}"

def processTimeZones(tzdataDir, icuDir, icuTzDir, version, ignoreBackzone, ignoreFactory, out):
    """ Read the time zone info and create a new time zone cpp file. """
    print("Processing tzdata mapping...")
    (ianaZones, ianaLinks) = readIANATimeZones(tzdataDir, ignoreBackzone, ignoreFactory)
    (icuZones, icuLinks) = readICUTimeZones(icuDir, icuTzDir, ignoreFactory)
    (legacyZones, legacyLinks) = readICULegacyZones(icuDir)

    incorrectZones = findIncorrectICUZones(ianaZones, ianaLinks, icuZones, icuLinks, ignoreBackzone)
    if not incorrectZones:
        print("<<< No incorrect ICU time zones found, please update Intl.js! >>>")
        print("<<< Maybe https://ssl.icu-project.org/trac/ticket/12044 was fixed? >>>")

    incorrectLinks = findIncorrectICULinks(ianaZones, ianaLinks, icuZones, icuLinks)
    if not incorrectLinks:
        print("<<< No incorrect ICU time zone links found, please update Intl.js! >>>")
        print("<<< Maybe https://ssl.icu-project.org/trac/ticket/12044 was fixed? >>>")

    print("Writing Intl tzdata file...")
    with io.open(out, mode="w", encoding="utf-8", newline="") as f:
        println = partial(print, file=f)

        println(generatedFileWarning)
        println(tzdataVersionComment.format(version))
        println(u"")

        println(u"#ifndef builtin_IntlTimeZoneData_h")
        println(u"#define builtin_IntlTimeZoneData_h")
        println(u"")

        println(u"namespace js {")
        println(u"namespace timezone {")
        println(u"")

        println(u"// Format:")
        println(u'// "ZoneName" // ICU-Name [time zone file]')
        println(u"const char* const ianaZonesTreatedAsLinksByICU[] = {")
        for (zone, icuZone) in incorrectZones:
            println(u'    "%s", // %s [%s]' % (zone, icuZone, zone.filename))
        println(u"};")
        println(u"")

        println(u"// Format:")
        println(u'// "LinkName", "Target" // ICU-Target [time zone file]')
        println(u"struct LinkAndTarget");
        println(u"{");
        println(u"    const char* const link;");
        println(u"    const char* const target;");
        println(u"};");
        println(u"")
        println(u"const LinkAndTarget ianaLinksCanonicalizedDifferentlyByICU[] = {")
        for (zone, target, icuTarget) in incorrectLinks:
            println(u'    { "%s", "%s" }, // %s [%s]' % (zone, target, icuTarget, zone.filename))
        println(u"};")
        println(u"")

        println(u"// Legacy ICU time zones, these are not valid IANA time zone names. We also")
        println(u"// disallow the old and deprecated System V time zones.")
        println(u"// https://ssl.icu-project.org/repos/icu/trunk/icu4c/source/tools/tzcode/icuzones")
        println(u"const char* const legacyICUTimeZones[] = {")
        for zone in chain(sorted(legacyLinks.keys()), sorted(legacyZones)):
            println(u'    "%s",' % zone)
        println(u"};")
        println(u"")

        println(u"} // namespace timezone")
        println(u"} // namespace js")
        println(u"")
        println(u"#endif /* builtin_IntlTimeZoneData_h */")

def updateBackzoneLinks(tzdataDir, links):
    (backzoneZones, backzoneLinks) = readIANAFiles(tzdataDir, ["backzone"])
    (stableZones, updatedLinks, updatedZones) = partition(
        links.iteritems(),
        # Link not changed in backzone.
        lambda (zone, target): zone not in backzoneLinks and zone not in backzoneZones,
        # Link has a new target.
        lambda (zone, target): zone in backzoneLinks,
    )
    # Keep stable zones and links with updated target.
    return dict(chain(
                stableZones,
                imap(lambda (zone, target): (zone, backzoneLinks[zone]), updatedLinks)
           ))

def generateTzDataLinkTestContent(testDir, version, fileName, description, links):
    with io.open(os.path.join(testDir, fileName), mode="w", encoding="utf-8", newline="") as f:
        println = partial(print, file=f)

        println(u'// |reftest| skip-if(!this.hasOwnProperty("Intl"))')
        println(u"")
        println(generatedFileWarning)
        println(tzdataVersionComment.format(version))
        println(u"""
const tzMapper = [
    x => x,
    x => x.toUpperCase(),
    x => x.toLowerCase(),
];
""")

        println(description)
        println(u"const links = {")
        for (zone, target) in sorted(links, key=itemgetter(0)):
            println(u'    "%s": "%s",' % (zone, target))
        println(u"};")

        println(u"""
for (let [linkName, target] of Object.entries(links)) {
    if (target === "Etc/UTC" || target === "Etc/GMT")
        target = "UTC";

    for (let map of tzMapper) {
        let dtf = new Intl.DateTimeFormat(undefined, {timeZone: map(linkName)});
        let resolvedTimeZone = dtf.resolvedOptions().timeZone;
        assertEq(resolvedTimeZone, target, `${linkName} -> ${target}`);
    }
}
""")
        println(u"""
if (typeof reportCompare === "function")
    reportCompare(0, 0, "ok");
""")

def generateTzDataTestBackwardLinks(tzdataDir, version, ignoreBackzone, testDir):
    (zones, links) = readIANAFiles(tzdataDir, ["backward"])
    assert len(zones) == 0

    if not ignoreBackzone:
        links = updateBackzoneLinks(tzdataDir, links)

    generateTzDataLinkTestContent(
        testDir, version,
        "timeZone_backward_links.js",
        u"// Link names derived from IANA Time Zone Database, backward file.",
        links.iteritems()
    )

def generateTzDataTestNotBackwardLinks(tzdataDir, version, ignoreBackzone, testDir):
    tzfiles = ifilterfalse({"backward", "backzone"}.__contains__, listIANAFiles(tzdataDir))
    (zones, links) = readIANAFiles(tzdataDir, tzfiles)

    if not ignoreBackzone:
        links = updateBackzoneLinks(tzdataDir, links)

    generateTzDataLinkTestContent(
        testDir, version,
        "timeZone_notbackward_links.js",
        u"// Link names derived from IANA Time Zone Database, excluding backward file.",
        links.iteritems()
    )

def generateTzDataTestBackzone(tzdataDir, version, ignoreBackzone, testDir):
    backzoneFiles = {"backzone"}
    (bkfiles, tzfiles) = partition(listIANAFiles(tzdataDir), backzoneFiles.__contains__)

    # Read zone and link infos.
    (zones, links) = readIANAFiles(tzdataDir, tzfiles)
    (backzones, backlinks) = readIANAFiles(tzdataDir, bkfiles)

    if not ignoreBackzone:
        comment=u"""\
// This file was generated with historical, pre-1970 backzone information
// respected. Therefore, every zone key listed below is its own Zone, not
// a Link to a modern-day target as IANA ignoring backzones would say.

"""
    else:
        comment=u"""\
// This file was generated while ignoring historical, pre-1970 backzone
// information. Therefore, every zone key listed below is part of a Link
// whose target is the corresponding value.

"""

    generateTzDataLinkTestContent(
        testDir, version,
        "timeZone_backzone.js",
        comment + u"// Backzone zones derived from IANA Time Zone Database.",
        ((zone, zone if not ignoreBackzone else links[zone]) for zone in backzones if zone in links)
    )

def generateTzDataTestBackzoneLinks(tzdataDir, version, ignoreBackzone, testDir):
    backzoneFiles = {"backzone"}
    (bkfiles, tzfiles) = partition(listIANAFiles(tzdataDir), backzoneFiles.__contains__)

    # Read zone and link infos.
    (zones, links) = readIANAFiles(tzdataDir, tzfiles)
    (backzones, backlinks) = readIANAFiles(tzdataDir, bkfiles)

    if not ignoreBackzone:
        comment=u"""\
// This file was generated with historical, pre-1970 backzone information
// respected. Therefore, every zone key listed below points to a target
// in the backzone file and not to its modern-day target as IANA ignoring
// backzones would say.

"""
    else:
        comment=u"""\
// This file was generated while ignoring historical, pre-1970 backzone
// information. Therefore, every zone key listed below is part of a Link
// whose target is the corresponding value ignoring any backzone entries.

"""

    generateTzDataLinkTestContent(
        testDir, version,
        "timeZone_backzone_links.js",
        comment +  u"// Backzone links derived from IANA Time Zone Database.",
        ((zone, target if not ignoreBackzone else links[zone]) for (zone, target) in backlinks.iteritems())
    )

def generateTzDataTests(tzdataDir, version, ignoreBackzone, testDir):
    generateTzDataTestBackwardLinks(tzdataDir, version, ignoreBackzone, testDir)
    generateTzDataTestNotBackwardLinks(tzdataDir, version, ignoreBackzone, testDir)
    generateTzDataTestBackzone(tzdataDir, version, ignoreBackzone, testDir)
    generateTzDataTestBackzoneLinks(tzdataDir, version, ignoreBackzone, testDir)

def updateTzdata(topsrcdir, args):
    """ Update the time zone cpp file. """

    icuDir = os.path.join(topsrcdir, "intl/icu/source")
    if not os.path.isdir(icuDir):
        raise RuntimeError("not a directory: %s" % icuDir)

    icuTzDir = os.path.join(topsrcdir, "intl/tzdata/source")
    if not os.path.isdir(icuTzDir):
        raise RuntimeError("not a directory: %s" % icuTzDir)

    dateTimeFormatTestDir = os.path.join(topsrcdir, "js/src/tests/Intl/DateTimeFormat")
    if not os.path.isdir(dateTimeFormatTestDir):
        raise RuntimeError("not a directory: %s" % dateTimeFormatTestDir)

    tzDir = args.tz
    if tzDir is not None and not (os.path.isdir(tzDir) or os.path.isfile(tzDir)):
        raise RuntimeError("not a directory or file: %s" % tzDir)
    ignoreBackzone = args.ignore_backzone
    # TODO: Accept or ignore the placeholder time zone "Factory"?
    ignoreFactory = False
    out = args.out

    version = icuTzDataVersion(icuTzDir)
    url = "https://www.iana.org/time-zones/repository/releases/tzdata%s.tar.gz" % version

    print("Arguments:")
    print("\ttzdata version: %s" % version)
    print("\ttzdata URL: %s" % url)
    print("\ttzdata directory|file: %s" % tzDir)
    print("\tICU directory: %s" % icuDir)
    print("\tICU timezone directory: %s" % icuTzDir)
    print("\tIgnore backzone file: %s" % ignoreBackzone)
    print("\tOutput file: %s" % out)
    print("")

    def updateFrom(f):
        if os.path.isfile(f) and tarfile.is_tarfile(f):
            with tarfile.open(f, "r:*") as tar:
                processTimeZones(TzDataFile(tar), icuDir, icuTzDir, version, ignoreBackzone, ignoreFactory, out)
                generateTzDataTests(TzDataFile(tar), version, ignoreBackzone, dateTimeFormatTestDir)
        elif os.path.isdir(f):
            processTimeZones(TzDataDir(f), icuDir, icuTzDir, version, ignoreBackzone, ignoreFactory, out)
            generateTzDataTests(TzDataDir(f), version, ignoreBackzone, dateTimeFormatTestDir)
        else:
            raise RuntimeError("unknown format")

    if tzDir is None:
        print("Downloading tzdata file...")
        with closing(urllib2.urlopen(url)) as tzfile:
            fname = urlparse.urlsplit(tzfile.geturl()).path.split("/")[-1]
            with tempfile.NamedTemporaryFile(suffix=fname) as tztmpfile:
                print("File stored in %s" % tztmpfile.name)
                tztmpfile.write(tzfile.read())
                tztmpfile.flush()
                updateFrom(tztmpfile.name)
    else:
        updateFrom(tzDir)

def readCurrencyFile(tree):
    reCurrency = re.compile(r"^[A-Z]{3}$")
    reIntMinorUnits = re.compile(r"^\d+$")

    for country in tree.iterfind(".//CcyNtry"):
        # Skip entry if no currency information is available.
        currency = country.findtext("Ccy")
        if currency is None:
            continue
        assert reCurrency.match(currency)

        minorUnits = country.findtext("CcyMnrUnts")
        assert minorUnits is not None

        # Skip all entries without minorUnits or which use the default minorUnits.
        if reIntMinorUnits.match(minorUnits) and int(minorUnits) != 2:
            currencyName = country.findtext("CcyNm")
            countryName = country.findtext("CtryNm")
            yield (currency, int(minorUnits), currencyName, countryName)

def writeCurrencyFile(published, currencies, out):
    with io.open(out, mode="w", encoding="utf-8", newline="") as f:
        println = partial(print, file=f)

        println(generatedFileWarning)
        println(u"// Version: {}".format(published))

        println(u"""
/**
 * Mapping from currency codes to the number of decimal digits used for them.
 * Default is 2 digits.
 *
 * Spec: ISO 4217 Currency and Funds Code List.
 * http://www.currency-iso.org/en/home/tables/table-a1.html
 */""")
        println(u"var currencyDigits = {")
        for (currency, entries) in groupby(sorted(currencies, key=itemgetter(0)), itemgetter(0)):
            for (_, minorUnits, currencyName, countryName) in entries:
                println(u"    // {} ({})".format(currencyName, countryName))
            println(u"    {}: {},".format(currency, minorUnits))
        println(u"};")

def updateCurrency(topsrcdir, args):
    """ Update the IntlCurrency.js file. """
    import xml.etree.ElementTree as ET
    from random import randint

    url = args.url
    out = args.out
    filename = args.file

    print("Arguments:")
    print("\tDownload url: %s" % url)
    print("\tLocal currency file: %s" % filename)
    print("\tOutput file: %s" % out)
    print("")

    def updateFrom(currencyFile):
        print("Processing currency code list file...")
        tree = ET.parse(currencyFile)
        published = tree.getroot().attrib["Pblshd"]
        currencies = readCurrencyFile(tree)

        print("Writing IntlCurrency file...")
        writeCurrencyFile(published, currencies, out)

    if filename is not None:
        print("Always make sure you have the newest currency code list file!")
        updateFrom(filename)
    else:
        print("Downloading currency & funds code list...")
        request = urllib2.Request(url)
        # Fake a random user agent string to circumvent the bot detection from
        # currency-iso.org...
        request.add_header("User-agent", "Mozilla/5.0 (Mobile; rv:{0}.0) Gecko/{0}.0 Firefox/{0}.0".format(randint(1, 999)))
        with closing(urllib2.urlopen(request)) as currencyFile:
            fname = urlparse.urlsplit(currencyFile.geturl()).path.split("/")[-1]
            with tempfile.NamedTemporaryFile(suffix=fname) as currencyTmpFile:
                print("File stored in %s" % currencyTmpFile.name)
                currencyTmpFile.write(currencyFile.read())
                currencyTmpFile.flush()
                updateFrom(currencyTmpFile.name)

if __name__ == "__main__":
    import argparse

    # This script must reside in js/src/builtin to work correctly.
    (thisDir, thisFile) = os.path.split(os.path.abspath(sys.argv[0]))
    dirPaths = os.path.normpath(thisDir).split(os.sep)
    if "/".join(dirPaths[-3:]) != "js/src/builtin":
        raise RuntimeError("%s must reside in js/src/builtin" % sys.argv[0])
    topsrcdir = "/".join(dirPaths[:-3])

    def EnsureHttps(v):
        if not v.startswith("https:"):
            raise argparse.ArgumentTypeError("URL protocol must be https: " % v)
        return v

    parser = argparse.ArgumentParser(description="Update intl data.")
    subparsers = parser.add_subparsers(help="Select update mode")

    parser_tags = subparsers.add_parser("langtags",
                                        help="Update language-subtag-registry")
    parser_tags.add_argument("--url",
                             metavar="URL",
                             default="https://www.iana.org/assignments/language-subtag-registry",
                             type=EnsureHttps,
                             help="Download url for language-subtag-registry.txt (default: %(default)s)")
    parser_tags.add_argument("--out",
                             default="IntlData.js",
                             help="Output file (default: %(default)s)")
    parser_tags.add_argument("file",
                             nargs="?",
                             help="Local language-subtag-registry.txt file, if omitted uses <URL>")
    parser_tags.set_defaults(func=updateLangTags)

    parser_tz = subparsers.add_parser("tzdata", help="Update tzdata")
    parser_tz.add_argument("--tz",
                           help="Local tzdata directory or file, if omitted downloads tzdata "
                                "distribution from https://www.iana.org/time-zones/")
    # ICU doesn't include the backzone file by default, but we still like to
    # use the backzone time zone names to avoid user confusion. This does lead
    # to formatting "historic" dates (pre-1970 era) with the wrong time zone,
    # but that's probably acceptable for now.
    parser_tz.add_argument("--ignore-backzone",
                           action="store_true",
                           help="Ignore tzdata's 'backzone' file. Can be enabled to generate more "
                                "accurate time zone canonicalization reflecting the actual time "
                                "zones as used by ICU.")
    parser_tz.add_argument("--out",
                           default="IntlTimeZoneData.h",
                           help="Output file (default: %(default)s)")
    parser_tz.set_defaults(func=partial(updateTzdata, topsrcdir))


    parser_currency = subparsers.add_parser("currency", help="Update currency digits mapping")
    parser_currency.add_argument("--url",
                                 metavar="URL",
                                 default="https://www.currency-iso.org/dam/downloads/lists/list_one.xml",
                                 type=EnsureHttps,
                                 help="Download url for the currency & funds code list (default: "
                                      "%(default)s)")
    parser_currency.add_argument("--out",
                                 default="IntlCurrency.js",
                                 help="Output file (default: %(default)s)")
    parser_currency.add_argument("file",
                                 nargs="?",
                                 help="Local currency code list file, if omitted uses <URL>")
    parser_currency.set_defaults(func=partial(updateCurrency, topsrcdir))

    args = parser.parse_args()
    args.func(args)