emitter.py 42 KB
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 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

# Emitter expects events obeying the following grammar:
# stream ::= STREAM-START document* STREAM-END
# document ::= DOCUMENT-START node DOCUMENT-END
# node ::= SCALAR | sequence | mapping
# sequence ::= SEQUENCE-START node* SEQUENCE-END
# mapping ::= MAPPING-START (node node)* MAPPING-END

__all__ = ['Emitter', 'EmitterError']

from .error import YAMLError
from .events import *

class EmitterError(YAMLError):
    pass

class ScalarAnalysis:
    def __init__(self, scalar, empty, multiline,
            allow_flow_plain, allow_block_plain,
            allow_single_quoted, allow_double_quoted,
            allow_block):
        self.scalar = scalar
        self.empty = empty
        self.multiline = multiline
        self.allow_flow_plain = allow_flow_plain
        self.allow_block_plain = allow_block_plain
        self.allow_single_quoted = allow_single_quoted
        self.allow_double_quoted = allow_double_quoted
        self.allow_block = allow_block

class Emitter:

    DEFAULT_TAG_PREFIXES = {
        '!' : '!',
        'tag:yaml.org,2002:' : '!!',
    }

    def __init__(self, stream, canonical=None, indent=None, width=None,
            allow_unicode=None, line_break=None):

        # The stream should have the methods `write` and possibly `flush`.
        self.stream = stream

        # Encoding can be overridden by STREAM-START.
        self.encoding = None

        # Emitter is a state machine with a stack of states to handle nested
        # structures.
        self.states = []
        self.state = self.expect_stream_start

        # Current event and the event queue.
        self.events = []
        self.event = None

        # The current indentation level and the stack of previous indents.
        self.indents = []
        self.indent = None

        # Flow level.
        self.flow_level = 0

        # Contexts.
        self.root_context = False
        self.sequence_context = False
        self.mapping_context = False
        self.simple_key_context = False

        # Characteristics of the last emitted character:
        #  - current position.
        #  - is it a whitespace?
        #  - is it an indention character
        #    (indentation space, '-', '?', or ':')?
        self.line = 0
        self.column = 0
        self.whitespace = True
        self.indention = True

        # Whether the document requires an explicit document indicator
        self.open_ended = False

        # Formatting details.
        self.canonical = canonical
        self.allow_unicode = allow_unicode
        self.best_indent = 2
        if indent and 1 < indent < 10:
            self.best_indent = indent
        self.best_width = 80
        if width and width > self.best_indent*2:
            self.best_width = width
        self.best_line_break = '\n'
        if line_break in ['\r', '\n', '\r\n']:
            self.best_line_break = line_break

        # Tag prefixes.
        self.tag_prefixes = None

        # Prepared anchor and tag.
        self.prepared_anchor = None
        self.prepared_tag = None

        # Scalar analysis and style.
        self.analysis = None
        self.style = None

    def dispose(self):
        # Reset the state attributes (to clear self-references)
        self.states = []
        self.state = None

    def emit(self, event):
        self.events.append(event)
        while not self.need_more_events():
            self.event = self.events.pop(0)
            self.state()
            self.event = None

    # In some cases, we wait for a few next events before emitting.

    def need_more_events(self):
        if not self.events:
            return True
        event = self.events[0]
        if isinstance(event, DocumentStartEvent):
            return self.need_events(1)
        elif isinstance(event, SequenceStartEvent):
            return self.need_events(2)
        elif isinstance(event, MappingStartEvent):
            return self.need_events(3)
        else:
            return False

    def need_events(self, count):
        level = 0
        for event in self.events[1:]:
            if isinstance(event, (DocumentStartEvent, CollectionStartEvent)):
                level += 1
            elif isinstance(event, (DocumentEndEvent, CollectionEndEvent)):
                level -= 1
            elif isinstance(event, StreamEndEvent):
                level = -1
            if level < 0:
                return False
        return (len(self.events) < count+1)

    def increase_indent(self, flow=False, indentless=False):
        self.indents.append(self.indent)
        if self.indent is None:
            if flow:
                self.indent = self.best_indent
            else:
                self.indent = 0
        elif not indentless:
            self.indent += self.best_indent

    # States.

    # Stream handlers.

    def expect_stream_start(self):
        if isinstance(self.event, StreamStartEvent):
            if self.event.encoding and not hasattr(self.stream, 'encoding'):
                self.encoding = self.event.encoding
            self.write_stream_start()
            self.state = self.expect_first_document_start
        else:
            raise EmitterError("expected StreamStartEvent, but got %s"
                    % self.event)

    def expect_nothing(self):
        raise EmitterError("expected nothing, but got %s" % self.event)

    # Document handlers.

    def expect_first_document_start(self):
        return self.expect_document_start(first=True)

    def expect_document_start(self, first=False):
        if isinstance(self.event, DocumentStartEvent):
            if (self.event.version or self.event.tags) and self.open_ended:
                self.write_indicator('...', True)
                self.write_indent()
            if self.event.version:
                version_text = self.prepare_version(self.event.version)
                self.write_version_directive(version_text)
            self.tag_prefixes = self.DEFAULT_TAG_PREFIXES.copy()
            if self.event.tags:
                handles = sorted(self.event.tags.keys())
                for handle in handles:
                    prefix = self.event.tags[handle]
                    self.tag_prefixes[prefix] = handle
                    handle_text = self.prepare_tag_handle(handle)
                    prefix_text = self.prepare_tag_prefix(prefix)
                    self.write_tag_directive(handle_text, prefix_text)
            implicit = (first and not self.event.explicit and not self.canonical
                    and not self.event.version and not self.event.tags
                    and not self.check_empty_document())
            if not implicit:
                self.write_indent()
                self.write_indicator('---', True)
                if self.canonical:
                    self.write_indent()
            self.state = self.expect_document_root
        elif isinstance(self.event, StreamEndEvent):
            if self.open_ended:
                self.write_indicator('...', True)
                self.write_indent()
            self.write_stream_end()
            self.state = self.expect_nothing
        else:
            raise EmitterError("expected DocumentStartEvent, but got %s"
                    % self.event)

    def expect_document_end(self):
        if isinstance(self.event, DocumentEndEvent):
            self.write_indent()
            if self.event.explicit:
                self.write_indicator('...', True)
                self.write_indent()
            self.flush_stream()
            self.state = self.expect_document_start
        else:
            raise EmitterError("expected DocumentEndEvent, but got %s"
                    % self.event)

    def expect_document_root(self):
        self.states.append(self.expect_document_end)
        self.expect_node(root=True)

    # Node handlers.

    def expect_node(self, root=False, sequence=False, mapping=False,
            simple_key=False):
        self.root_context = root
        self.sequence_context = sequence
        self.mapping_context = mapping
        self.simple_key_context = simple_key
        if isinstance(self.event, AliasEvent):
            self.expect_alias()
        elif isinstance(self.event, (ScalarEvent, CollectionStartEvent)):
            self.process_anchor('&')
            self.process_tag()
            if isinstance(self.event, ScalarEvent):
                self.expect_scalar()
            elif isinstance(self.event, SequenceStartEvent):
                if self.flow_level or self.canonical or self.event.flow_style   \
                        or self.check_empty_sequence():
                    self.expect_flow_sequence()
                else:
                    self.expect_block_sequence()
            elif isinstance(self.event, MappingStartEvent):
                if self.flow_level or self.canonical or self.event.flow_style   \
                        or self.check_empty_mapping():
                    self.expect_flow_mapping()
                else:
                    self.expect_block_mapping()
        else:
            raise EmitterError("expected NodeEvent, but got %s" % self.event)

    def expect_alias(self):
        if self.event.anchor is None:
            raise EmitterError("anchor is not specified for alias")
        self.process_anchor('*')
        self.state = self.states.pop()

    def expect_scalar(self):
        self.increase_indent(flow=True)
        self.process_scalar()
        self.indent = self.indents.pop()
        self.state = self.states.pop()

    # Flow sequence handlers.

    def expect_flow_sequence(self):
        self.write_indicator('[', True, whitespace=True)
        self.flow_level += 1
        self.increase_indent(flow=True)
        self.state = self.expect_first_flow_sequence_item

    def expect_first_flow_sequence_item(self):
        if isinstance(self.event, SequenceEndEvent):
            self.indent = self.indents.pop()
            self.flow_level -= 1
            self.write_indicator(']', False)
            self.state = self.states.pop()
        else:
            if self.canonical or self.column > self.best_width:
                self.write_indent()
            self.states.append(self.expect_flow_sequence_item)
            self.expect_node(sequence=True)

    def expect_flow_sequence_item(self):
        if isinstance(self.event, SequenceEndEvent):
            self.indent = self.indents.pop()
            self.flow_level -= 1
            if self.canonical:
                self.write_indicator(',', False)
                self.write_indent()
            self.write_indicator(']', False)
            self.state = self.states.pop()
        else:
            self.write_indicator(',', False)
            if self.canonical or self.column > self.best_width:
                self.write_indent()
            self.states.append(self.expect_flow_sequence_item)
            self.expect_node(sequence=True)

    # Flow mapping handlers.

    def expect_flow_mapping(self):
        self.write_indicator('{', True, whitespace=True)
        self.flow_level += 1
        self.increase_indent(flow=True)
        self.state = self.expect_first_flow_mapping_key

    def expect_first_flow_mapping_key(self):
        if isinstance(self.event, MappingEndEvent):
            self.indent = self.indents.pop()
            self.flow_level -= 1
            self.write_indicator('}', False)
            self.state = self.states.pop()
        else:
            if self.canonical or self.column > self.best_width:
                self.write_indent()
            if not self.canonical and self.check_simple_key():
                self.states.append(self.expect_flow_mapping_simple_value)
                self.expect_node(mapping=True, simple_key=True)
            else:
                self.write_indicator('?', True)
                self.states.append(self.expect_flow_mapping_value)
                self.expect_node(mapping=True)

    def expect_flow_mapping_key(self):
        if isinstance(self.event, MappingEndEvent):
            self.indent = self.indents.pop()
            self.flow_level -= 1
            if self.canonical:
                self.write_indicator(',', False)
                self.write_indent()
            self.write_indicator('}', False)
            self.state = self.states.pop()
        else:
            self.write_indicator(',', False)
            if self.canonical or self.column > self.best_width:
                self.write_indent()
            if not self.canonical and self.check_simple_key():
                self.states.append(self.expect_flow_mapping_simple_value)
                self.expect_node(mapping=True, simple_key=True)
            else:
                self.write_indicator('?', True)
                self.states.append(self.expect_flow_mapping_value)
                self.expect_node(mapping=True)

    def expect_flow_mapping_simple_value(self):
        self.write_indicator(':', False)
        self.states.append(self.expect_flow_mapping_key)
        self.expect_node(mapping=True)

    def expect_flow_mapping_value(self):
        if self.canonical or self.column > self.best_width:
            self.write_indent()
        self.write_indicator(':', True)
        self.states.append(self.expect_flow_mapping_key)
        self.expect_node(mapping=True)

    # Block sequence handlers.

    def expect_block_sequence(self):
        indentless = (self.mapping_context and not self.indention)
        self.increase_indent(flow=False, indentless=indentless)
        self.state = self.expect_first_block_sequence_item

    def expect_first_block_sequence_item(self):
        return self.expect_block_sequence_item(first=True)

    def expect_block_sequence_item(self, first=False):
        if not first and isinstance(self.event, SequenceEndEvent):
            self.indent = self.indents.pop()
            self.state = self.states.pop()
        else:
            self.write_indent()
            self.write_indicator('-', True, indention=True)
            self.states.append(self.expect_block_sequence_item)
            self.expect_node(sequence=True)

    # Block mapping handlers.

    def expect_block_mapping(self):
        self.increase_indent(flow=False)
        self.state = self.expect_first_block_mapping_key

    def expect_first_block_mapping_key(self):
        return self.expect_block_mapping_key(first=True)

    def expect_block_mapping_key(self, first=False):
        if not first and isinstance(self.event, MappingEndEvent):
            self.indent = self.indents.pop()
            self.state = self.states.pop()
        else:
            self.write_indent()
            if self.check_simple_key():
                self.states.append(self.expect_block_mapping_simple_value)
                self.expect_node(mapping=True, simple_key=True)
            else:
                self.write_indicator('?', True, indention=True)
                self.states.append(self.expect_block_mapping_value)
                self.expect_node(mapping=True)

    def expect_block_mapping_simple_value(self):
        self.write_indicator(':', False)
        self.states.append(self.expect_block_mapping_key)
        self.expect_node(mapping=True)

    def expect_block_mapping_value(self):
        self.write_indent()
        self.write_indicator(':', True, indention=True)
        self.states.append(self.expect_block_mapping_key)
        self.expect_node(mapping=True)

    # Checkers.

    def check_empty_sequence(self):
        return (isinstance(self.event, SequenceStartEvent) and self.events
                and isinstance(self.events[0], SequenceEndEvent))

    def check_empty_mapping(self):
        return (isinstance(self.event, MappingStartEvent) and self.events
                and isinstance(self.events[0], MappingEndEvent))

    def check_empty_document(self):
        if not isinstance(self.event, DocumentStartEvent) or not self.events:
            return False
        event = self.events[0]
        return (isinstance(event, ScalarEvent) and event.anchor is None
                and event.tag is None and event.implicit and event.value == '')

    def check_simple_key(self):
        length = 0
        if isinstance(self.event, NodeEvent) and self.event.anchor is not None:
            if self.prepared_anchor is None:
                self.prepared_anchor = self.prepare_anchor(self.event.anchor)
            length += len(self.prepared_anchor)
        if isinstance(self.event, (ScalarEvent, CollectionStartEvent))  \
                and self.event.tag is not None:
            if self.prepared_tag is None:
                self.prepared_tag = self.prepare_tag(self.event.tag)
            length += len(self.prepared_tag)
        if isinstance(self.event, ScalarEvent):
            if self.analysis is None:
                self.analysis = self.analyze_scalar(self.event.value)
            length += len(self.analysis.scalar)
        return (length < 128 and (isinstance(self.event, AliasEvent)
            or (isinstance(self.event, ScalarEvent)
                    and not self.analysis.empty and not self.analysis.multiline)
            or self.check_empty_sequence() or self.check_empty_mapping()))

    # Anchor, Tag, and Scalar processors.

    def process_anchor(self, indicator):
        if self.event.anchor is None:
            self.prepared_anchor = None
            return
        if self.prepared_anchor is None:
            self.prepared_anchor = self.prepare_anchor(self.event.anchor)
        if self.prepared_anchor:
            self.write_indicator(indicator+self.prepared_anchor, True)
        self.prepared_anchor = None

    def process_tag(self):
        tag = self.event.tag
        if isinstance(self.event, ScalarEvent):
            if self.style is None:
                self.style = self.choose_scalar_style()
            if ((not self.canonical or tag is None) and
                ((self.style == '' and self.event.implicit[0])
                        or (self.style != '' and self.event.implicit[1]))):
                self.prepared_tag = None
                return
            if self.event.implicit[0] and tag is None:
                tag = '!'
                self.prepared_tag = None
        else:
            if (not self.canonical or tag is None) and self.event.implicit:
                self.prepared_tag = None
                return
        if tag is None:
            raise EmitterError("tag is not specified")
        if self.prepared_tag is None:
            self.prepared_tag = self.prepare_tag(tag)
        if self.prepared_tag:
            self.write_indicator(self.prepared_tag, True)
        self.prepared_tag = None

    def choose_scalar_style(self):
        if self.analysis is None:
            self.analysis = self.analyze_scalar(self.event.value)
        if self.event.style == '"' or self.canonical:
            return '"'
        if not self.event.style and self.event.implicit[0]:
            if (not (self.simple_key_context and
                    (self.analysis.empty or self.analysis.multiline))
                and (self.flow_level and self.analysis.allow_flow_plain
                    or (not self.flow_level and self.analysis.allow_block_plain))):
                return ''
        if self.event.style and self.event.style in '|>':
            if (not self.flow_level and not self.simple_key_context
                    and self.analysis.allow_block):
                return self.event.style
        if not self.event.style or self.event.style == '\'':
            if (self.analysis.allow_single_quoted and
                    not (self.simple_key_context and self.analysis.multiline)):
                return '\''
        return '"'

    def process_scalar(self):
        if self.analysis is None:
            self.analysis = self.analyze_scalar(self.event.value)
        if self.style is None:
            self.style = self.choose_scalar_style()
        split = (not self.simple_key_context)
        #if self.analysis.multiline and split    \
        #        and (not self.style or self.style in '\'\"'):
        #    self.write_indent()
        if self.style == '"':
            self.write_double_quoted(self.analysis.scalar, split)
        elif self.style == '\'':
            self.write_single_quoted(self.analysis.scalar, split)
        elif self.style == '>':
            self.write_folded(self.analysis.scalar)
        elif self.style == '|':
            self.write_literal(self.analysis.scalar)
        else:
            self.write_plain(self.analysis.scalar, split)
        self.analysis = None
        self.style = None

    # Analyzers.

    def prepare_version(self, version):
        major, minor = version
        if major != 1:
            raise EmitterError("unsupported YAML version: %d.%d" % (major, minor))
        return '%d.%d' % (major, minor)

    def prepare_tag_handle(self, handle):
        if not handle:
            raise EmitterError("tag handle must not be empty")
        if handle[0] != '!' or handle[-1] != '!':
            raise EmitterError("tag handle must start and end with '!': %r" % handle)
        for ch in handle[1:-1]:
            if not ('0' <= ch <= '9' or 'A' <= ch <= 'Z' or 'a' <= ch <= 'z'    \
                    or ch in '-_'):
                raise EmitterError("invalid character %r in the tag handle: %r"
                        % (ch, handle))
        return handle

    def prepare_tag_prefix(self, prefix):
        if not prefix:
            raise EmitterError("tag prefix must not be empty")
        chunks = []
        start = end = 0
        if prefix[0] == '!':
            end = 1
        while end < len(prefix):
            ch = prefix[end]
            if '0' <= ch <= '9' or 'A' <= ch <= 'Z' or 'a' <= ch <= 'z' \
                    or ch in '-;/?!:@&=+$,_.~*\'()[]':
                end += 1
            else:
                if start < end:
                    chunks.append(prefix[start:end])
                start = end = end+1
                data = ch.encode('utf-8')
                for ch in data:
                    chunks.append('%%%02X' % ord(ch))
        if start < end:
            chunks.append(prefix[start:end])
        return ''.join(chunks)

    def prepare_tag(self, tag):
        if not tag:
            raise EmitterError("tag must not be empty")
        if tag == '!':
            return tag
        handle = None
        suffix = tag
        prefixes = sorted(self.tag_prefixes.keys())
        for prefix in prefixes:
            if tag.startswith(prefix)   \
                    and (prefix == '!' or len(prefix) < len(tag)):
                handle = self.tag_prefixes[prefix]
                suffix = tag[len(prefix):]
        chunks = []
        start = end = 0
        while end < len(suffix):
            ch = suffix[end]
            if '0' <= ch <= '9' or 'A' <= ch <= 'Z' or 'a' <= ch <= 'z' \
                    or ch in '-;/?:@&=+$,_.~*\'()[]'   \
                    or (ch == '!' and handle != '!'):
                end += 1
            else:
                if start < end:
                    chunks.append(suffix[start:end])
                start = end = end+1
                data = ch.encode('utf-8')
                for ch in data:
                    chunks.append('%%%02X' % ch)
        if start < end:
            chunks.append(suffix[start:end])
        suffix_text = ''.join(chunks)
        if handle:
            return '%s%s' % (handle, suffix_text)
        else:
            return '!<%s>' % suffix_text

    def prepare_anchor(self, anchor):
        if not anchor:
            raise EmitterError("anchor must not be empty")
        for ch in anchor:
            if not ('0' <= ch <= '9' or 'A' <= ch <= 'Z' or 'a' <= ch <= 'z'    \
                    or ch in '-_'):
                raise EmitterError("invalid character %r in the anchor: %r"
                        % (ch, anchor))
        return anchor

    def analyze_scalar(self, scalar):

        # Empty scalar is a special case.
        if not scalar:
            return ScalarAnalysis(scalar=scalar, empty=True, multiline=False,
                    allow_flow_plain=False, allow_block_plain=True,
                    allow_single_quoted=True, allow_double_quoted=True,
                    allow_block=False)

        # Indicators and special characters.
        block_indicators = False
        flow_indicators = False
        line_breaks = False
        special_characters = False

        # Important whitespace combinations.
        leading_space = False
        leading_break = False
        trailing_space = False
        trailing_break = False
        break_space = False
        space_break = False

        # Check document indicators.
        if scalar.startswith('---') or scalar.startswith('...'):
            block_indicators = True
            flow_indicators = True

        # First character or preceded by a whitespace.
        preceded_by_whitespace = True

        # Last character or followed by a whitespace.
        followed_by_whitespace = (len(scalar) == 1 or
                scalar[1] in '\0 \t\r\n\x85\u2028\u2029')

        # The previous character is a space.
        previous_space = False

        # The previous character is a break.
        previous_break = False

        index = 0
        while index < len(scalar):
            ch = scalar[index]

            # Check for indicators.
            if index == 0:
                # Leading indicators are special characters.
                if ch in '#,[]{}&*!|>\'\"%@`':
                    flow_indicators = True
                    block_indicators = True
                if ch in '?:':
                    flow_indicators = True
                    if followed_by_whitespace:
                        block_indicators = True
                if ch == '-' and followed_by_whitespace:
                    flow_indicators = True
                    block_indicators = True
            else:
                # Some indicators cannot appear within a scalar as well.
                if ch in ',?[]{}':
                    flow_indicators = True
                if ch == ':':
                    flow_indicators = True
                    if followed_by_whitespace:
                        block_indicators = True
                if ch == '#' and preceded_by_whitespace:
                    flow_indicators = True
                    block_indicators = True

            # Check for line breaks, special, and unicode characters.
            if ch in '\n\x85\u2028\u2029':
                line_breaks = True
            if not (ch == '\n' or '\x20' <= ch <= '\x7E'):
                if (ch == '\x85' or '\xA0' <= ch <= '\uD7FF'
                        or '\uE000' <= ch <= '\uFFFD'
                        or '\U00010000' <= ch < '\U0010ffff') and ch != '\uFEFF':
                    unicode_characters = True
                    if not self.allow_unicode:
                        special_characters = True
                else:
                    special_characters = True

            # Detect important whitespace combinations.
            if ch == ' ':
                if index == 0:
                    leading_space = True
                if index == len(scalar)-1:
                    trailing_space = True
                if previous_break:
                    break_space = True
                previous_space = True
                previous_break = False
            elif ch in '\n\x85\u2028\u2029':
                if index == 0:
                    leading_break = True
                if index == len(scalar)-1:
                    trailing_break = True
                if previous_space:
                    space_break = True
                previous_space = False
                previous_break = True
            else:
                previous_space = False
                previous_break = False

            # Prepare for the next character.
            index += 1
            preceded_by_whitespace = (ch in '\0 \t\r\n\x85\u2028\u2029')
            followed_by_whitespace = (index+1 >= len(scalar) or
                    scalar[index+1] in '\0 \t\r\n\x85\u2028\u2029')

        # Let's decide what styles are allowed.
        allow_flow_plain = True
        allow_block_plain = True
        allow_single_quoted = True
        allow_double_quoted = True
        allow_block = True

        # Leading and trailing whitespaces are bad for plain scalars.
        if (leading_space or leading_break
                or trailing_space or trailing_break):
            allow_flow_plain = allow_block_plain = False

        # We do not permit trailing spaces for block scalars.
        if trailing_space:
            allow_block = False

        # Spaces at the beginning of a new line are only acceptable for block
        # scalars.
        if break_space:
            allow_flow_plain = allow_block_plain = allow_single_quoted = False

        # Spaces followed by breaks, as well as special character are only
        # allowed for double quoted scalars.
        if space_break or special_characters:
            allow_flow_plain = allow_block_plain =  \
            allow_single_quoted = allow_block = False

        # Although the plain scalar writer supports breaks, we never emit
        # multiline plain scalars.
        if line_breaks:
            allow_flow_plain = allow_block_plain = False

        # Flow indicators are forbidden for flow plain scalars.
        if flow_indicators:
            allow_flow_plain = False

        # Block indicators are forbidden for block plain scalars.
        if block_indicators:
            allow_block_plain = False

        return ScalarAnalysis(scalar=scalar,
                empty=False, multiline=line_breaks,
                allow_flow_plain=allow_flow_plain,
                allow_block_plain=allow_block_plain,
                allow_single_quoted=allow_single_quoted,
                allow_double_quoted=allow_double_quoted,
                allow_block=allow_block)

    # Writers.

    def flush_stream(self):
        if hasattr(self.stream, 'flush'):
            self.stream.flush()

    def write_stream_start(self):
        # Write BOM if needed.
        if self.encoding and self.encoding.startswith('utf-16'):
            self.stream.write('\uFEFF'.encode(self.encoding))

    def write_stream_end(self):
        self.flush_stream()

    def write_indicator(self, indicator, need_whitespace,
            whitespace=False, indention=False):
        if self.whitespace or not need_whitespace:
            data = indicator
        else:
            data = ' '+indicator
        self.whitespace = whitespace
        self.indention = self.indention and indention
        self.column += len(data)
        self.open_ended = False
        if self.encoding:
            data = data.encode(self.encoding)
        self.stream.write(data)

    def write_indent(self):
        indent = self.indent or 0
        if not self.indention or self.column > indent   \
                or (self.column == indent and not self.whitespace):
            self.write_line_break()
        if self.column < indent:
            self.whitespace = True
            data = ' '*(indent-self.column)
            self.column = indent
            if self.encoding:
                data = data.encode(self.encoding)
            self.stream.write(data)

    def write_line_break(self, data=None):
        if data is None:
            data = self.best_line_break
        self.whitespace = True
        self.indention = True
        self.line += 1
        self.column = 0
        if self.encoding:
            data = data.encode(self.encoding)
        self.stream.write(data)

    def write_version_directive(self, version_text):
        data = '%%YAML %s' % version_text
        if self.encoding:
            data = data.encode(self.encoding)
        self.stream.write(data)
        self.write_line_break()

    def write_tag_directive(self, handle_text, prefix_text):
        data = '%%TAG %s %s' % (handle_text, prefix_text)
        if self.encoding:
            data = data.encode(self.encoding)
        self.stream.write(data)
        self.write_line_break()

    # Scalar streams.

    def write_single_quoted(self, text, split=True):
        self.write_indicator('\'', True)
        spaces = False
        breaks = False
        start = end = 0
        while end <= len(text):
            ch = None
            if end < len(text):
                ch = text[end]
            if spaces:
                if ch is None or ch != ' ':
                    if start+1 == end and self.column > self.best_width and split   \
                            and start != 0 and end != len(text):
                        self.write_indent()
                    else:
                        data = text[start:end]
                        self.column += len(data)
                        if self.encoding:
                            data = data.encode(self.encoding)
                        self.stream.write(data)
                    start = end
            elif breaks:
                if ch is None or ch not in '\n\x85\u2028\u2029':
                    if text[start] == '\n':
                        self.write_line_break()
                    for br in text[start:end]:
                        if br == '\n':
                            self.write_line_break()
                        else:
                            self.write_line_break(br)
                    self.write_indent()
                    start = end
            else:
                if ch is None or ch in ' \n\x85\u2028\u2029' or ch == '\'':
                    if start < end:
                        data = text[start:end]
                        self.column += len(data)
                        if self.encoding:
                            data = data.encode(self.encoding)
                        self.stream.write(data)
                        start = end
            if ch == '\'':
                data = '\'\''
                self.column += 2
                if self.encoding:
                    data = data.encode(self.encoding)
                self.stream.write(data)
                start = end + 1
            if ch is not None:
                spaces = (ch == ' ')
                breaks = (ch in '\n\x85\u2028\u2029')
            end += 1
        self.write_indicator('\'', False)

    ESCAPE_REPLACEMENTS = {
        '\0':       '0',
        '\x07':     'a',
        '\x08':     'b',
        '\x09':     't',
        '\x0A':     'n',
        '\x0B':     'v',
        '\x0C':     'f',
        '\x0D':     'r',
        '\x1B':     'e',
        '\"':       '\"',
        '\\':       '\\',
        '\x85':     'N',
        '\xA0':     '_',
        '\u2028':   'L',
        '\u2029':   'P',
    }

    def write_double_quoted(self, text, split=True):
        self.write_indicator('"', True)
        start = end = 0
        while end <= len(text):
            ch = None
            if end < len(text):
                ch = text[end]
            if ch is None or ch in '"\\\x85\u2028\u2029\uFEFF' \
                    or not ('\x20' <= ch <= '\x7E'
                        or (self.allow_unicode
                            and ('\xA0' <= ch <= '\uD7FF'
                                or '\uE000' <= ch <= '\uFFFD'))):
                if start < end:
                    data = text[start:end]
                    self.column += len(data)
                    if self.encoding:
                        data = data.encode(self.encoding)
                    self.stream.write(data)
                    start = end
                if ch is not None:
                    if ch in self.ESCAPE_REPLACEMENTS:
                        data = '\\'+self.ESCAPE_REPLACEMENTS[ch]
                    elif ch <= '\xFF':
                        data = '\\x%02X' % ord(ch)
                    elif ch <= '\uFFFF':
                        data = '\\u%04X' % ord(ch)
                    else:
                        data = '\\U%08X' % ord(ch)
                    self.column += len(data)
                    if self.encoding:
                        data = data.encode(self.encoding)
                    self.stream.write(data)
                    start = end+1
            if 0 < end < len(text)-1 and (ch == ' ' or start >= end)    \
                    and self.column+(end-start) > self.best_width and split:
                data = text[start:end]+'\\'
                if start < end:
                    start = end
                self.column += len(data)
                if self.encoding:
                    data = data.encode(self.encoding)
                self.stream.write(data)
                self.write_indent()
                self.whitespace = False
                self.indention = False
                if text[start] == ' ':
                    data = '\\'
                    self.column += len(data)
                    if self.encoding:
                        data = data.encode(self.encoding)
                    self.stream.write(data)
            end += 1
        self.write_indicator('"', False)

    def determine_block_hints(self, text):
        hints = ''
        if text:
            if text[0] in ' \n\x85\u2028\u2029':
                hints += str(self.best_indent)
            if text[-1] not in '\n\x85\u2028\u2029':
                hints += '-'
            elif len(text) == 1 or text[-2] in '\n\x85\u2028\u2029':
                hints += '+'
        return hints

    def write_folded(self, text):
        hints = self.determine_block_hints(text)
        self.write_indicator('>'+hints, True)
        if hints[-1:] == '+':
            self.open_ended = True
        self.write_line_break()
        leading_space = True
        spaces = False
        breaks = True
        start = end = 0
        while end <= len(text):
            ch = None
            if end < len(text):
                ch = text[end]
            if breaks:
                if ch is None or ch not in '\n\x85\u2028\u2029':
                    if not leading_space and ch is not None and ch != ' '   \
                            and text[start] == '\n':
                        self.write_line_break()
                    leading_space = (ch == ' ')
                    for br in text[start:end]:
                        if br == '\n':
                            self.write_line_break()
                        else:
                            self.write_line_break(br)
                    if ch is not None:
                        self.write_indent()
                    start = end
            elif spaces:
                if ch != ' ':
                    if start+1 == end and self.column > self.best_width:
                        self.write_indent()
                    else:
                        data = text[start:end]
                        self.column += len(data)
                        if self.encoding:
                            data = data.encode(self.encoding)
                        self.stream.write(data)
                    start = end
            else:
                if ch is None or ch in ' \n\x85\u2028\u2029':
                    data = text[start:end]
                    self.column += len(data)
                    if self.encoding:
                        data = data.encode(self.encoding)
                    self.stream.write(data)
                    if ch is None:
                        self.write_line_break()
                    start = end
            if ch is not None:
                breaks = (ch in '\n\x85\u2028\u2029')
                spaces = (ch == ' ')
            end += 1

    def write_literal(self, text):
        hints = self.determine_block_hints(text)
        self.write_indicator('|'+hints, True)
        if hints[-1:] == '+':
            self.open_ended = True
        self.write_line_break()
        breaks = True
        start = end = 0
        while end <= len(text):
            ch = None
            if end < len(text):
                ch = text[end]
            if breaks:
                if ch is None or ch not in '\n\x85\u2028\u2029':
                    for br in text[start:end]:
                        if br == '\n':
                            self.write_line_break()
                        else:
                            self.write_line_break(br)
                    if ch is not None:
                        self.write_indent()
                    start = end
            else:
                if ch is None or ch in '\n\x85\u2028\u2029':
                    data = text[start:end]
                    if self.encoding:
                        data = data.encode(self.encoding)
                    self.stream.write(data)
                    if ch is None:
                        self.write_line_break()
                    start = end
            if ch is not None:
                breaks = (ch in '\n\x85\u2028\u2029')
            end += 1

    def write_plain(self, text, split=True):
        if self.root_context:
            self.open_ended = True
        if not text:
            return
        if not self.whitespace:
            data = ' '
            self.column += len(data)
            if self.encoding:
                data = data.encode(self.encoding)
            self.stream.write(data)
        self.whitespace = False
        self.indention = False
        spaces = False
        breaks = False
        start = end = 0
        while end <= len(text):
            ch = None
            if end < len(text):
                ch = text[end]
            if spaces:
                if ch != ' ':
                    if start+1 == end and self.column > self.best_width and split:
                        self.write_indent()
                        self.whitespace = False
                        self.indention = False
                    else:
                        data = text[start:end]
                        self.column += len(data)
                        if self.encoding:
                            data = data.encode(self.encoding)
                        self.stream.write(data)
                    start = end
            elif breaks:
                if ch not in '\n\x85\u2028\u2029':
                    if text[start] == '\n':
                        self.write_line_break()
                    for br in text[start:end]:
                        if br == '\n':
                            self.write_line_break()
                        else:
                            self.write_line_break(br)
                    self.write_indent()
                    self.whitespace = False
                    self.indention = False
                    start = end
            else:
                if ch is None or ch in ' \n\x85\u2028\u2029':
                    data = text[start:end]
                    self.column += len(data)
                    if self.encoding:
                        data = data.encode(self.encoding)
                    self.stream.write(data)
                    start = end
            if ch is not None:
                spaces = (ch == ' ')
                breaks = (ch in '\n\x85\u2028\u2029')
            end += 1