PNG  IHDRxsBIT|d pHYs+tEXtSoftwarewww.inkscape.org<,tEXtComment File Manager

File Manager

Path: /opt/alt/python27/lib/python2.7/site-packages/pip/_vendor/html5lib/

Viewing File: serializer.py

from __future__ import absolute_import, division, unicode_literals
from pip._vendor.six import text_type

import re

from codecs import register_error, xmlcharrefreplace_errors

from .constants import voidElements, booleanAttributes, spaceCharacters
from .constants import rcdataElements, entities, xmlEntities
from . import treewalkers, _utils
from xml.sax.saxutils import escape

_quoteAttributeSpecChars = "".join(spaceCharacters) + "\"'=<>`"
_quoteAttributeSpec = re.compile("[" + _quoteAttributeSpecChars + "]")
_quoteAttributeLegacy = re.compile("[" + _quoteAttributeSpecChars +
                                   "\x00\x01\x02\x03\x04\x05\x06\x07\x08\t\n"
                                   "\x0b\x0c\r\x0e\x0f\x10\x11\x12\x13\x14\x15"
                                   "\x16\x17\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f"
                                   "\x20\x2f\x60\xa0\u1680\u180e\u180f\u2000"
                                   "\u2001\u2002\u2003\u2004\u2005\u2006\u2007"
                                   "\u2008\u2009\u200a\u2028\u2029\u202f\u205f"
                                   "\u3000]")


_encode_entity_map = {}
_is_ucs4 = len("\U0010FFFF") == 1
for k, v in list(entities.items()):
    # skip multi-character entities
    if ((_is_ucs4 and len(v) > 1) or
            (not _is_ucs4 and len(v) > 2)):
        continue
    if v != "&":
        if len(v) == 2:
            v = _utils.surrogatePairToCodepoint(v)
        else:
            v = ord(v)
        if v not in _encode_entity_map or k.islower():
            # prefer &lt; over &LT; and similarly for &amp;, &gt;, etc.
            _encode_entity_map[v] = k


def htmlentityreplace_errors(exc):
    if isinstance(exc, (UnicodeEncodeError, UnicodeTranslateError)):
        res = []
        codepoints = []
        skip = False
        for i, c in enumerate(exc.object[exc.start:exc.end]):
            if skip:
                skip = False
                continue
            index = i + exc.start
            if _utils.isSurrogatePair(exc.object[index:min([exc.end, index + 2])]):
                codepoint = _utils.surrogatePairToCodepoint(exc.object[index:index + 2])
                skip = True
            else:
                codepoint = ord(c)
            codepoints.append(codepoint)
        for cp in codepoints:
            e = _encode_entity_map.get(cp)
            if e:
                res.append("&")
                res.append(e)
                if not e.endswith(";"):
                    res.append(";")
            else:
                res.append("&#x%s;" % (hex(cp)[2:]))
        return ("".join(res), exc.end)
    else:
        return xmlcharrefreplace_errors(exc)


register_error("htmlentityreplace", htmlentityreplace_errors)


def serialize(input, tree="etree", encoding=None, **serializer_opts):
    """Serializes the input token stream using the specified treewalker

    :arg input: the token stream to serialize

    :arg tree: the treewalker to use

    :arg encoding: the encoding to use

    :arg serializer_opts: any options to pass to the
        :py:class:`html5lib.serializer.HTMLSerializer` that gets created

    :returns: the tree serialized as a string

    Example:

    >>> from html5lib.html5parser import parse
    >>> from html5lib.serializer import serialize
    >>> token_stream = parse('<html><body><p>Hi!</p></body></html>')
    >>> serialize(token_stream, omit_optional_tags=False)
    '<html><head></head><body><p>Hi!</p></body></html>'

    """
    # XXX: Should we cache this?
    walker = treewalkers.getTreeWalker(tree)
    s = HTMLSerializer(**serializer_opts)
    return s.render(walker(input), encoding)


class HTMLSerializer(object):

    # attribute quoting options
    quote_attr_values = "legacy"  # be secure by default
    quote_char = '"'
    use_best_quote_char = True

    # tag syntax options
    omit_optional_tags = True
    minimize_boolean_attributes = True
    use_trailing_solidus = False
    space_before_trailing_solidus = True

    # escaping options
    escape_lt_in_attrs = False
    escape_rcdata = False
    resolve_entities = True

    # miscellaneous options
    alphabetical_attributes = False
    inject_meta_charset = True
    strip_whitespace = False
    sanitize = False

    options = ("quote_attr_values", "quote_char", "use_best_quote_char",
               "omit_optional_tags", "minimize_boolean_attributes",
               "use_trailing_solidus", "space_before_trailing_solidus",
               "escape_lt_in_attrs", "escape_rcdata", "resolve_entities",
               "alphabetical_attributes", "inject_meta_charset",
               "strip_whitespace", "sanitize")

    def __init__(self, **kwargs):
        """Initialize HTMLSerializer

        :arg inject_meta_charset: Whether or not to inject the meta charset.

            Defaults to ``True``.

        :arg quote_attr_values: Whether to quote attribute values that don't
            require quoting per legacy browser behavior (``"legacy"``), when
            required by the standard (``"spec"``), or always (``"always"``).

            Defaults to ``"legacy"``.

        :arg quote_char: Use given quote character for attribute quoting.

            Defaults to ``"`` which will use double quotes unless attribute
            value contains a double quote, in which case single quotes are
            used.

        :arg escape_lt_in_attrs: Whether or not to escape ``<`` in attribute
            values.

            Defaults to ``False``.

        :arg escape_rcdata: Whether to escape characters that need to be
            escaped within normal elements within rcdata elements such as
            style.

            Defaults to ``False``.

        :arg resolve_entities: Whether to resolve named character entities that
            appear in the source tree. The XML predefined entities &lt; &gt;
            &amp; &quot; &apos; are unaffected by this setting.

            Defaults to ``True``.

        :arg strip_whitespace: Whether to remove semantically meaningless
            whitespace. (This compresses all whitespace to a single space
            except within ``pre``.)

            Defaults to ``False``.

        :arg minimize_boolean_attributes: Shortens boolean attributes to give
            just the attribute value, for example::

              <input disabled="disabled">

            becomes::

              <input disabled>

            Defaults to ``True``.

        :arg use_trailing_solidus: Includes a close-tag slash at the end of the
            start tag of void elements (empty elements whose end tag is
            forbidden). E.g. ``<hr/>``.

            Defaults to ``False``.

        :arg space_before_trailing_solidus: Places a space immediately before
            the closing slash in a tag using a trailing solidus. E.g.
            ``<hr />``. Requires ``use_trailing_solidus=True``.

            Defaults to ``True``.

        :arg sanitize: Strip all unsafe or unknown constructs from output.
            See :py:class:`html5lib.filters.sanitizer.Filter`.

            Defaults to ``False``.

        :arg omit_optional_tags: Omit start/end tags that are optional.

            Defaults to ``True``.

        :arg alphabetical_attributes: Reorder attributes to be in alphabetical order.

            Defaults to ``False``.

        """
        unexpected_args = frozenset(kwargs) - frozenset(self.options)
        if len(unexpected_args) > 0:
            raise TypeError("__init__() got an unexpected keyword argument '%s'" % next(iter(unexpected_args)))
        if 'quote_char' in kwargs:
            self.use_best_quote_char = False
        for attr in self.options:
            setattr(self, attr, kwargs.get(attr, getattr(self, attr)))
        self.errors = []
        self.strict = False

    def encode(self, string):
        assert(isinstance(string, text_type))
        if self.encoding:
            return string.encode(self.encoding, "htmlentityreplace")
        else:
            return string

    def encodeStrict(self, string):
        assert(isinstance(string, text_type))
        if self.encoding:
            return string.encode(self.encoding, "strict")
        else:
            return string

    def serialize(self, treewalker, encoding=None):
        # pylint:disable=too-many-nested-blocks
        self.encoding = encoding
        in_cdata = False
        self.errors = []

        if encoding and self.inject_meta_charset:
            from .filters.inject_meta_charset import Filter
            treewalker = Filter(treewalker, encoding)
        # Alphabetical attributes is here under the assumption that none of
        # the later filters add or change order of attributes; it needs to be
        # before the sanitizer so escaped elements come out correctly
        if self.alphabetical_attributes:
            from .filters.alphabeticalattributes import Filter
            treewalker = Filter(treewalker)
        # WhitespaceFilter should be used before OptionalTagFilter
        # for maximum efficiently of this latter filter
        if self.strip_whitespace:
            from .filters.whitespace import Filter
            treewalker = Filter(treewalker)
        if self.sanitize:
            from .filters.sanitizer import Filter
            treewalker = Filter(treewalker)
        if self.omit_optional_tags:
            from .filters.optionaltags import Filter
            treewalker = Filter(treewalker)

        for token in treewalker:
            type = token["type"]
            if type == "Doctype":
                doctype = "<!DOCTYPE %s" % token["name"]

                if token["publicId"]:
                    doctype += ' PUBLIC "%s"' % token["publicId"]
                elif token["systemId"]:
                    doctype += " SYSTEM"
                if token["systemId"]:
                    if token["systemId"].find('"') >= 0:
                        if token["systemId"].find("'") >= 0:
                            self.serializeError("System identifier contains both single and double quote characters")
                        quote_char = "'"
                    else:
                        quote_char = '"'
                    doctype += " %s%s%s" % (quote_char, token["systemId"], quote_char)

                doctype += ">"
                yield self.encodeStrict(doctype)

            elif type in ("Characters", "SpaceCharacters"):
                if type == "SpaceCharacters" or in_cdata:
                    if in_cdata and token["data"].find("</") >= 0:
                        self.serializeError("Unexpected </ in CDATA")
                    yield self.encode(token["data"])
                else:
                    yield self.encode(escape(token["data"]))

            elif type in ("StartTag", "EmptyTag"):
                name = token["name"]
                yield self.encodeStrict("<%s" % name)
                if name in rcdataElements and not self.escape_rcdata:
                    in_cdata = True
                elif in_cdata:
                    self.serializeError("Unexpected child element of a CDATA element")
                for (_, attr_name), attr_value in token["data"].items():
                    # TODO: Add namespace support here
                    k = attr_name
                    v = attr_value
                    yield self.encodeStrict(' ')

                    yield self.encodeStrict(k)
                    if not self.minimize_boolean_attributes or \
                        (k not in booleanAttributes.get(name, tuple()) and
                         k not in booleanAttributes.get("", tuple())):
                        yield self.encodeStrict("=")
                        if self.quote_attr_values == "always" or len(v) == 0:
                            quote_attr = True
                        elif self.quote_attr_values == "spec":
                            quote_attr = _quoteAttributeSpec.search(v) is not None
                        elif self.quote_attr_values == "legacy":
                            quote_attr = _quoteAttributeLegacy.search(v) is not None
                        else:
                            raise ValueError("quote_attr_values must be one of: "
                                             "'always', 'spec', or 'legacy'")
                        v = v.replace("&", "&amp;")
                        if self.escape_lt_in_attrs:
                            v = v.replace("<", "&lt;")
                        if quote_attr:
                            quote_char = self.quote_char
                            if self.use_best_quote_char:
                                if "'" in v and '"' not in v:
                                    quote_char = '"'
                                elif '"' in v and "'" not in v:
                                    quote_char = "'"
                            if quote_char == "'":
                                v = v.replace("'", "&#39;")
                            else:
                                v = v.replace('"', "&quot;")
                            yield self.encodeStrict(quote_char)
                            yield self.encode(v)
                            yield self.encodeStrict(quote_char)
                        else:
                            yield self.encode(v)
                if name in voidElements and self.use_trailing_solidus:
                    if self.space_before_trailing_solidus:
                        yield self.encodeStrict(" /")
                    else:
                        yield self.encodeStrict("/")
                yield self.encode(">")

            elif type == "EndTag":
                name = token["name"]
                if name in rcdataElements:
                    in_cdata = False
                elif in_cdata:
                    self.serializeError("Unexpected child element of a CDATA element")
                yield self.encodeStrict("</%s>" % name)

            elif type == "Comment":
                data = token["data"]
                if data.find("--") >= 0:
                    self.serializeError("Comment contains --")
                yield self.encodeStrict("<!--%s-->" % token["data"])

            elif type == "Entity":
                name = token["name"]
                key = name + ";"
                if key not in entities:
                    self.serializeError("Entity %s not recognized" % name)
                if self.resolve_entities and key not in xmlEntities:
                    data = entities[key]
                else:
                    data = "&%s;" % name
                yield self.encodeStrict(data)

            else:
                self.serializeError(token["data"])

    def render(self, treewalker, encoding=None):
        """Serializes the stream from the treewalker into a string

        :arg treewalker: the treewalker to serialize

        :arg encoding: the string encoding to use

        :returns: the serialized tree

        Example:

        >>> from html5lib import parse, getTreeWalker
        >>> from html5lib.serializer import HTMLSerializer
        >>> token_stream = parse('<html><body>Hi!</body></html>')
        >>> walker = getTreeWalker('etree')
        >>> serializer = HTMLSerializer(omit_optional_tags=False)
        >>> serializer.render(walker(token_stream))
        '<html><head></head><body>Hi!</body></html>'

        """
        if encoding:
            return b"".join(list(self.serialize(treewalker, encoding)))
        else:
            return "".join(list(self.serialize(treewalker)))

    def serializeError(self, data="XXX ERROR MESSAGE NEEDED"):
        # XXX The idea is to make data mandatory.
        self.errors.append(data)
        if self.strict:
            raise SerializeError


class SerializeError(Exception):
    """Error in serialized tree"""
    pass
b IDATxytVսϓ22 A@IR :hCiZ[v*E:WũZA ^dQeQ @ !jZ'>gsV仿$|?g)&x-EIENT ;@xT.i%-X}SvS5.r/UHz^_$-W"w)Ɗ/@Z &IoX P$K}JzX:;` &, ŋui,e6mX ԵrKb1ԗ)DADADADADADADADADADADADADADADADADADADADADADADADADADADADADADADADADADADADADADADADADADA݀!I*]R;I2$eZ#ORZSrr6mteffu*((Pu'v{DIߔ4^pIm'77WEEE;vƎ4-$]'RI{\I&G :IHJ DWBB=\WR޽m o$K(V9ABB.}jѢv`^?IOȅ} ڶmG}T#FJ`56$-ھ}FI&v;0(h;Б38CӧOWf!;A i:F_m9s&|q%=#wZprrrla A &P\\СC[A#! {olF} `E2}MK/vV)i{4BffV\|ۭX`b@kɶ@%i$K z5zhmX[IXZ` 'b%$r5M4º/l ԃߖxhʔ)[@=} K6IM}^5k㏷݆z ΗÿO:gdGBmyT/@+Vɶ纽z񕏵l.y޴it뭷zV0[Y^>Wsqs}\/@$(T7f.InݺiR$푔n.~?H))\ZRW'Mo~v Ov6oԃxz! S,&xm/yɞԟ?'uaSѽb,8GלKboi&3t7Y,)JJ c[nzӳdE&KsZLӄ I?@&%ӟ۶mSMMњ0iؐSZ,|J+N ~,0A0!5%Q-YQQa3}$_vVrf9f?S8`zDADADADADADADADADAdqP,تmMmg1V?rSI꒟]u|l RCyEf٢9 jURbztѰ!m5~tGj2DhG*{H9)꒟ר3:(+3\?/;TUݭʴ~S6lڧUJ*i$d(#=Yݺd{,p|3B))q:vN0Y.jkק6;SɶVzHJJЀ-utѹսk>QUU\޲~]fFnK?&ߡ5b=z9)^|u_k-[y%ZNU6 7Mi:]ۦtk[n X(e6Bb."8cۭ|~teuuw|ήI-5"~Uk;ZicEmN/:]M> cQ^uiƞ??Ңpc#TUU3UakNwA`:Y_V-8.KKfRitv޲* 9S6ֿj,ՃNOMߤ]z^fOh|<>@Å5 _/Iu?{SY4hK/2]4%it5q]GGe2%iR| W&f*^]??vq[LgE_3f}Fxu~}qd-ږFxu~I N>\;͗O֊:̗WJ@BhW=y|GgwܷH_NY?)Tdi'?խwhlmQi !SUUsw4kӺe4rfxu-[nHtMFj}H_u~w>)oV}(T'ebʒv3_[+vn@Ȭ\S}ot}w=kHFnxg S 0eޢm~l}uqZfFoZuuEg `zt~? b;t%>WTkķh[2eG8LIWx,^\thrl^Ϊ{=dž<}qV@ ⠨Wy^LF_>0UkDuʫuCs$)Iv:IK;6ֲ4{^6եm+l3>݆uM 9u?>Zc }g~qhKwڭeFMM~pМuqǿz6Tb@8@Y|jx](^]gf}M"tG -w.@vOqh~/HII`S[l.6nØXL9vUcOoB\xoǤ'T&IǍQw_wpv[kmO{w~>#=P1Pɞa-we:iǏlHo׈꒟f9SzH?+shk%Fs:qVhqY`jvO'ρ?PyX3lх]˾uV{ݞ]1,MzYNW~̈́ joYn}ȚF߾׮mS]F z+EDxm/d{F{-W-4wY듏:??_gPf ^3ecg ҵs8R2מz@TANGj)}CNi/R~}c:5{!ZHӋӾ6}T]G]7W6^n 9*,YqOZj:P?Q DFL|?-^.Ɵ7}fFh׶xe2Pscz1&5\cn[=Vn[ĶE鎀uˌd3GII k;lNmشOuuRVfBE]ۣeӶu :X-[(er4~LHi6:Ѻ@ԅrST0trk%$Č0ez" *z"T/X9|8.C5Feg}CQ%͞ˣJvL/?j^h&9xF`њZ(&yF&Iݻfg#W;3^{Wo^4'vV[[K';+mӍִ]AC@W?1^{එyh +^]fm~iԵ]AB@WTk̏t uR?l.OIHiYyԶ]Aˀ7c:q}ힽaf6Z~қm(+sK4{^6}T*UUu]n.:kx{:2 _m=sAߤU@?Z-Vކеz왍Nэ{|5 pڶn b p-@sPg]0G7fy-M{GCF'%{4`=$-Ge\ eU:m+Zt'WjO!OAF@ik&t݆ϥ_ e}=]"Wz_.͜E3leWFih|t-wZۍ-uw=6YN{6|} |*={Ѽn.S.z1zjۻTH]흾 DuDvmvK.`V]yY~sI@t?/ϓ. m&["+P?MzovVЫG3-GRR[(!!\_,^%?v@ҵő m`Y)tem8GMx.))A]Y i`ViW`?^~!S#^+ѽGZj?Vģ0.))A꨷lzL*]OXrY`DBBLOj{-MH'ii-ϰ ok7^ )쭡b]UXSְmռY|5*cֽk0B7镹%ڽP#8nȎq}mJr23_>lE5$iwui+ H~F`IjƵ@q \ @#qG0".0" l`„.0! ,AQHN6qzkKJ#o;`Xv2>,tێJJ7Z/*A .@fفjMzkg @TvZH3Zxu6Ra'%O?/dQ5xYkU]Rֽkق@DaS^RSּ5|BeHNN͘p HvcYcC5:y #`οb;z2.!kr}gUWkyZn=f Pvsn3p~;4p˚=ē~NmI] ¾ 0lH[_L hsh_ғߤc_њec)g7VIZ5yrgk̞W#IjӪv>՞y睝M8[|]\շ8M6%|@PZڨI-m>=k='aiRo-x?>Q.}`Ȏ:Wsmu u > .@,&;+!!˱tﭧDQwRW\vF\~Q7>spYw$%A~;~}6¾ g&if_=j,v+UL1(tWake:@Ș>j$Gq2t7S?vL|]u/ .(0E6Mk6hiۺzښOrifޱxm/Gx> Lal%%~{lBsR4*}{0Z/tNIɚpV^#Lf:u@k#RSu =S^ZyuR/.@n&΃z~B=0eg뺆#,Þ[B/?H uUf7y Wy}Bwegל`Wh(||`l`.;Ws?V@"c:iɍL֯PGv6zctM̠':wuW;d=;EveD}9J@B(0iհ bvP1{\P&G7D޴Iy_$-Qjm~Yrr&]CDv%bh|Yzni_ˆR;kg}nJOIIwyuL}{ЌNj}:+3Y?:WJ/N+Rzd=hb;dj͒suݔ@NKMԄ jqzC5@y°hL m;*5ezᕏ=ep XL n?מ:r`۵tŤZ|1v`V뽧_csج'ߤ%oTuumk%%%h)uy]Nk[n 'b2 l.=͜E%gf$[c;s:V-͞WߤWh-j7]4=F-X]>ZLSi[Y*We;Zan(ӇW|e(HNNP5[= r4tP &0<pc#`vTNV GFqvTi*Tyam$ߏWyE*VJKMTfFw>'$-ؽ.Ho.8c"@DADADADADADADADADA~j*֘,N;Pi3599h=goضLgiJ5փy~}&Zd9p֚ e:|hL``b/d9p? fgg+%%hMgXosج, ΩOl0Zh=xdjLmhݻoO[g_l,8a]٭+ӧ0$I]c]:粹:Teꢢ"5a^Kgh,&= =՟^߶“ߢE ܹS J}I%:8 IDAT~,9/ʃPW'Mo}zNƍ쨓zPbNZ~^z=4mswg;5 Y~SVMRXUյڱRf?s:w ;6H:ºi5-maM&O3;1IKeamZh͛7+##v+c ~u~ca]GnF'ټL~PPPbn voC4R,ӟgg %hq}@#M4IÇ Oy^xMZx ) yOw@HkN˖-Sǎmb]X@n+i͖!++K3gd\$mt$^YfJ\8PRF)77Wא!Cl$i:@@_oG I{$# 8磌ŋ91A (Im7֭>}ߴJq7ޗt^ -[ԩSj*}%]&' -ɓ'ꫯVzzvB#;a 7@GxI{j޼ƌ.LÇWBB7`O"I$/@R @eee@۷>}0,ɒ2$53Xs|cS~rpTYYY} kHc %&k.], @ADADADADADADADADA@lT<%''*Lo^={رc5h %$+CnܸQ3fҥK}vUVVs9G R,_{xˇ3o߾;TTTd}馛]uuuG~iԩ@4bnvmvfϞ /Peeeq}}za I~,誫{UWW뮻}_~YƍSMMMYχ֝waw\ďcxꩧtEƍկ_?۷5@u?1kNׯWzz/wy>}zj3 k(ٺuq_Zvf̘:~ ABQ&r|!%KҥKgԞ={<_X-z !CyFUUz~ ABQIIIjݺW$UXXDٳZ~ ABQƍecW$<(~<RSSvZujjjԧOZQu@4 8m&&&jԩg$ď1h ͟?_{768@g =@`)))5o6m3)ѣƌJ;wҿUTT /KZR{~a=@0o<*狔iFɶ[ˎ;T]]OX@?K.ۈxN pppppppppppppppppPfl߾] ,{ァk۶mڿo5BTӦMӴiӴ|r DB2e|An!Dy'tkΝ[A $***t5' "!駟oaDnΝ:t֭[gDШQ06qD;@ x M6v(PiizmZ4ew"@̴ixf [~-Fٱc&IZ2|n!?$@{[HTɏ#@hȎI# _m(F /6Z3z'\r,r!;w2Z3j=~GY7"I$iI.p_"?pN`y DD?: _  Gÿab7J !Bx@0 Bo cG@`1C[@0G @`0C_u V1 aCX>W ` | `!<S `"<. `#c`?cAC4 ?c p#~@0?:08&_MQ1J h#?/`7;I  q 7a wQ A 1 Hp !#<8/#@1Ul7=S=K.4Z?E_$i@!1!E4?`P_  @Bă10#: "aU,xbFY1 [n|n #'vEH:`xb #vD4Y hi.i&EΖv#O H4IŶ}:Ikh @tZRF#(tXҙzZ ?I3l7q@õ|ۍ1,GpuY Ꮿ@hJv#xxk$ v#9 5 }_$c S#=+"K{F*m7`#%H:NRSp6I?sIՖ{Ap$I$I:QRv2$Z @UJ*$]<FO4IENDB`