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

File Manager

Path: /opt/cloudlinux/venv/lib/python3.11/site-packages/lxml/html/

Viewing File: soupparser.py

"""External interface to the BeautifulSoup HTML parser.
"""

__all__ = ["fromstring", "parse", "convert_tree"]

import re
from lxml import etree, html

try:
    from bs4 import (
        BeautifulSoup, Tag, Comment, ProcessingInstruction, NavigableString,
        Declaration, Doctype)
    _DECLARATION_OR_DOCTYPE = (Declaration, Doctype)
except ImportError:
    from BeautifulSoup import (
        BeautifulSoup, Tag, Comment, ProcessingInstruction, NavigableString,
        Declaration)
    _DECLARATION_OR_DOCTYPE = Declaration


def fromstring(data, beautifulsoup=None, makeelement=None, **bsargs):
    """Parse a string of HTML data into an Element tree using the
    BeautifulSoup parser.

    Returns the root ``<html>`` Element of the tree.

    You can pass a different BeautifulSoup parser through the
    `beautifulsoup` keyword, and a diffent Element factory function
    through the `makeelement` keyword.  By default, the standard
    ``BeautifulSoup`` class and the default factory of `lxml.html` are
    used.
    """
    return _parse(data, beautifulsoup, makeelement, **bsargs)


def parse(file, beautifulsoup=None, makeelement=None, **bsargs):
    """Parse a file into an ElemenTree using the BeautifulSoup parser.

    You can pass a different BeautifulSoup parser through the
    `beautifulsoup` keyword, and a diffent Element factory function
    through the `makeelement` keyword.  By default, the standard
    ``BeautifulSoup`` class and the default factory of `lxml.html` are
    used.
    """
    if not hasattr(file, 'read'):
        file = open(file)
    root = _parse(file, beautifulsoup, makeelement, **bsargs)
    return etree.ElementTree(root)


def convert_tree(beautiful_soup_tree, makeelement=None):
    """Convert a BeautifulSoup tree to a list of Element trees.

    Returns a list instead of a single root Element to support
    HTML-like soup with more than one root element.

    You can pass a different Element factory through the `makeelement`
    keyword.
    """
    root = _convert_tree(beautiful_soup_tree, makeelement)
    children = root.getchildren()
    for child in children:
        root.remove(child)
    return children


# helpers

def _parse(source, beautifulsoup, makeelement, **bsargs):
    if beautifulsoup is None:
        beautifulsoup = BeautifulSoup
    if hasattr(beautifulsoup, "HTML_ENTITIES"):  # bs3
        if 'convertEntities' not in bsargs:
            bsargs['convertEntities'] = 'html'
    if hasattr(beautifulsoup, "DEFAULT_BUILDER_FEATURES"):  # bs4
        if 'features' not in bsargs:
            bsargs['features'] = 'html.parser'  # use Python html parser
    tree = beautifulsoup(source, **bsargs)
    root = _convert_tree(tree, makeelement)
    # from ET: wrap the document in a html root element, if necessary
    if len(root) == 1 and root[0].tag == "html":
        return root[0]
    root.tag = "html"
    return root


_parse_doctype_declaration = re.compile(
    r'(?:\s|[<!])*DOCTYPE\s*HTML'
    r'(?:\s+PUBLIC)?(?:\s+(\'[^\']*\'|"[^"]*"))?'
    r'(?:\s+(\'[^\']*\'|"[^"]*"))?',
    re.IGNORECASE).match


class _PseudoTag:
    # Minimal imitation of BeautifulSoup.Tag
    def __init__(self, contents):
        self.name = 'html'
        self.attrs = []
        self.contents = contents

    def __iter__(self):
        return self.contents.__iter__()


def _convert_tree(beautiful_soup_tree, makeelement):
    if makeelement is None:
        makeelement = html.html_parser.makeelement

    # Split the tree into three parts:
    # i) everything before the root element: document type
    # declaration, comments, processing instructions, whitespace
    # ii) the root(s),
    # iii) everything after the root: comments, processing
    # instructions, whitespace
    first_element_idx = last_element_idx = None
    html_root = declaration = None
    for i, e in enumerate(beautiful_soup_tree):
        if isinstance(e, Tag):
            if first_element_idx is None:
                first_element_idx = i
            last_element_idx = i
            if html_root is None and e.name and e.name.lower() == 'html':
                html_root = e
        elif declaration is None and isinstance(e, _DECLARATION_OR_DOCTYPE):
            declaration = e

    # For a nice, well-formatted document, the variable roots below is
    # a list consisting of a single <html> element. However, the document
    # may be a soup like '<meta><head><title>Hello</head><body>Hi
    # all<\p>'. In this example roots is a list containing meta, head
    # and body elements.
    if first_element_idx is None:
        pre_root = post_root = []
        roots = beautiful_soup_tree.contents
    else:
        pre_root = beautiful_soup_tree.contents[:first_element_idx]
        roots = beautiful_soup_tree.contents[first_element_idx:last_element_idx+1]
        post_root = beautiful_soup_tree.contents[last_element_idx+1:]

    # Reorganize so that there is one <html> root...
    if html_root is not None:
        # ... use existing one if possible, ...
        i = roots.index(html_root)
        html_root.contents = roots[:i] + html_root.contents + roots[i+1:]
    else:
        # ... otherwise create a new one.
        html_root = _PseudoTag(roots)

    convert_node = _init_node_converters(makeelement)

    # Process pre_root
    res_root = convert_node(html_root)
    prev = res_root
    for e in reversed(pre_root):
        converted = convert_node(e)
        if converted is not None:
            prev.addprevious(converted)
            prev = converted

    # ditto for post_root
    prev = res_root
    for e in post_root:
        converted = convert_node(e)
        if converted is not None:
            prev.addnext(converted)
            prev = converted

    if declaration is not None:
        try:
            # bs4 provides full Doctype string
            doctype_string = declaration.output_ready()
        except AttributeError:
            doctype_string = declaration.string

        match = _parse_doctype_declaration(doctype_string)
        if not match:
            # Something is wrong if we end up in here. Since soupparser should
            # tolerate errors, do not raise Exception, just let it pass.
            pass
        else:
            external_id, sys_uri = match.groups()
            docinfo = res_root.getroottree().docinfo
            # strip quotes and update DOCTYPE values (any of None, '', '...')
            docinfo.public_id = external_id and external_id[1:-1]
            docinfo.system_url = sys_uri and sys_uri[1:-1]

    return res_root


def _init_node_converters(makeelement):
    converters = {}
    ordered_node_types = []

    def converter(*types):
        def add(handler):
            for t in types:
                converters[t] = handler
                ordered_node_types.append(t)
            return handler
        return add

    def find_best_converter(node):
        for t in ordered_node_types:
            if isinstance(node, t):
                return converters[t]
        return None

    def convert_node(bs_node, parent=None):
        # duplicated in convert_tag() below
        try:
            handler = converters[type(bs_node)]
        except KeyError:
            handler = converters[type(bs_node)] = find_best_converter(bs_node)
        if handler is None:
            return None
        return handler(bs_node, parent)

    def map_attrs(bs_attrs):
        if isinstance(bs_attrs, dict):  # bs4
            attribs = {}
            for k, v in bs_attrs.items():
                if isinstance(v, list):
                    v = " ".join(v)
                attribs[k] = unescape(v)
        else:
            attribs = dict((k, unescape(v)) for k, v in bs_attrs)
        return attribs

    def append_text(parent, text):
        if len(parent) == 0:
            parent.text = (parent.text or '') + text
        else:
            parent[-1].tail = (parent[-1].tail or '') + text

    # converters are tried in order of their definition

    @converter(Tag, _PseudoTag)
    def convert_tag(bs_node, parent):
        attrs = bs_node.attrs
        if parent is not None:
            attribs = map_attrs(attrs) if attrs else None
            res = etree.SubElement(parent, bs_node.name, attrib=attribs)
        else:
            attribs = map_attrs(attrs) if attrs else {}
            res = makeelement(bs_node.name, attrib=attribs)

        for child in bs_node:
            # avoid double recursion by inlining convert_node(), see above
            try:
                handler = converters[type(child)]
            except KeyError:
                pass
            else:
                if handler is not None:
                    handler(child, res)
                continue
            convert_node(child, res)
        return res

    @converter(Comment)
    def convert_comment(bs_node, parent):
        res = html.HtmlComment(bs_node)
        if parent is not None:
            parent.append(res)
        return res

    @converter(ProcessingInstruction)
    def convert_pi(bs_node, parent):
        if bs_node.endswith('?'):
            # The PI is of XML style (<?as df?>) but BeautifulSoup
            # interpreted it as being SGML style (<?as df>). Fix.
            bs_node = bs_node[:-1]
        res = etree.ProcessingInstruction(*bs_node.split(' ', 1))
        if parent is not None:
            parent.append(res)
        return res

    @converter(NavigableString)
    def convert_text(bs_node, parent):
        if parent is not None:
            append_text(parent, unescape(bs_node))
        return None

    return convert_node


# copied from ET's ElementSoup

try:
    from html.entities import name2codepoint  # Python 3
except ImportError:
    from htmlentitydefs import name2codepoint


handle_entities = re.compile(r"&(\w+);").sub


try:
    unichr
except NameError:
    # Python 3
    unichr = chr


def unescape(string):
    if not string:
        return ''
    # work around oddities in BeautifulSoup's entity handling
    def unescape_entity(m):
        try:
            return unichr(name2codepoint[m.group(1)])
        except KeyError:
            return m.group(0)  # use as is
    return handle_entities(unescape_entity, string)
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`