ó BÜXQc@sdZddlZddlZddlmZddlmZeefZddlmZddl Z d„Z e d„Z d„Z ejd ƒZejd ƒZejd ƒZejd ƒZd „ZejdƒZd„Zd„Zd„Zd„ZedkreƒndS(s4Utility functions for HTTP header value parsing and construction. Copyright 1997-1998, Gisle Aas Copyright 2002-2006, John J. Lee This code is free software; you can redistribute it and/or modify it under the terms of the BSD or ZPL 2.1 licenses (see the file COPYING.txt included with the distribution). iÿÿÿÿN(t StringType(t UnicodeType(t http2timecCsOtjjtj|ƒdƒd}ddg}|rE|dg7}n||kS(Niis.htms.htmls.xhtml(tostpathtsplitextt_rfc3986turlsplit(turlt allow_xhtmltextt html_exts((s;/scratch/rashmi/Condor_Script/src/mechanize/_headersutil.pytis_html_file_extensions # cCsŽ|st||ƒSt|ƒ}t|ƒdkr>t||ƒS|d}|d}|d}dg}|r„|ddddg7}n||kS(sM ct_headers: Sequence of Content-Type headers url: Response URL iis text/htmls text/xhtmlstext/xmlsapplication/xmlsapplication/xhtml+xml(R tsplit_header_wordstlen(t ct_headersRR theaderst first_headertfirst_parametertctt html_types((s;/scratch/rashmi/Condor_Script/src/mechanize/_headersutil.pytis_htmls       cCs+|jdƒ\}}|j| |j|S(s)Return unmatched part of re.Match object.i(tspantstring(tmatchtstarttend((s;/scratch/rashmi/Condor_Script/src/mechanize/_headersutil.pyt unmatched4ss^\s*([^=\s;,]+)s&^\s*=\s*\"([^\"\\]*(?:\\.[^\"\\]*)*)\"s^\s*=\s*([^\s;,]*)s\\(.)c Cs¹t|ƒtkst‚g}x”|D]Œ}|}g}xa|rštj|ƒ}|rt|ƒ}|jdƒ}tj|ƒ}|rµt|ƒ}|jdƒ}tj d|ƒ}nEt j|ƒ}|rôt|ƒ}|jdƒ}|j ƒ}nd}|j ||fƒq:|jƒjdƒrT|jƒd}|rK|j |ƒng}q:tjdd|ƒ\}} | dks‘td|||f‚|}q:W|r%|j |ƒq%q%W|S( smParse header values into a list of lists containing key,value pairs. The function knows how to deal with ",", ";" and "=" as well as quoted values after "=". A list of space separated tokens are parsed as if they were separated by ";". If the header_values passed as argument contains multiple values, then they are treated as if they were a single value separated by comma ",". This means that this function is useful for parsing header fields that follow this syntax (BNF as from the HTTP/1.1 specification, but we relax the requirement for tokens). headers = #header header = (token | parameter) *( [";"] (token | parameter)) token = 1* separators = "(" | ")" | "<" | ">" | "@" | "," | ";" | ":" | "\" | <"> | "/" | "[" | "]" | "?" | "=" | "{" | "}" | SP | HT quoted-string = ( <"> *(qdtext | quoted-pair ) <"> ) qdtext = > quoted-pair = "\" CHAR parameter = attribute "=" value attribute = token value = token | quoted-string Each header is represented by a list of key/value pairs. The value for a simple token (not part of a parameter) is None. Syntactically incorrect headers will not necessarily be parsed as you would want. This is easier to describe with some examples: >>> split_header_words(['foo="bar"; port="80,81"; discard, bar=baz']) [[('foo', 'bar'), ('port', '80,81'), ('discard', None)], [('bar', 'baz')]] >>> split_header_words(['text/html; charset="iso-8859-1"']) [[('text/html', None), ('charset', 'iso-8859-1')]] >>> split_header_words([r'Basic realm="\"foo\bar\""']) [[('Basic', None), ('realm', '"foobar"')]] is\1t,s^[=\s;]*tis&split_header_words bug: '%s', '%s', %sN(ttypet STRING_TYPEStAssertionErrorttoken_retsearchRtgrouptquoted_value_ret escape_retsubtvalue_retrstriptNonetappendtlstript startswithtretsubn( t header_valuestresultttextt orig_texttpairstmtnametvaluetnon_junkt nr_junk_chars((s;/scratch/rashmi/Condor_Script/src/mechanize/_headersutil.pyR =sF-       s([\"\\])cCsÒg}x¼|D]´}g}x†|D]~\}}|dk r‘tjd|ƒsitjd|ƒ}d|}n|dkr~|}q‘d||f}n|j|ƒq W|r |jdj|ƒƒq q Wdj|ƒS(s²Do the inverse of the conversion done by split_header_words. Takes a list of lists of (key, value) pairs and produces a single header value. Attribute values are quoted if needed. >>> join_header_words([[("text/plain", None), ("charset", "iso-8859/1")]]) 'text/plain; charset="iso-8859/1"' >>> join_header_words([[("text/plain", None)], [("charset", "iso-8859/1")]]) 'text/plain, charset="iso-8859/1"' s^\w+$s\\\1s"%s"s%s=%ss; s, N(R)R-R"tjoin_escape_reR&R*tjoin(tlistsRR3tattrtktv((s;/scratch/rashmi/Condor_Script/src/mechanize/_headersutil.pytjoin_header_words“s      cCs<|jdƒr|d}n|jdƒr8|d }n|S(Nt"iiÿÿÿÿ(R,tendswith(R1((s;/scratch/rashmi/Condor_Script/src/mechanize/_headersutil.pyt strip_quotes¯s   c Csxd}g}xe|D]]}g}t}tjd|ƒ}x tt|ƒƒD]ö}||}|jƒ}|d krxqJnd |kr”|d} } n'tjd |d ƒ\} } | jƒ} |d kr-| jƒ} | |krè| } n| dkr t | ƒ} t }n| dkr-t t | ƒƒ} q-n|j | | fƒqJW|r|s`|j dƒn|j |ƒqqW|S(s5Ad-hoc parser for Netscape protocol cookie-attributes. The old Netscape cookie format for Set-Cookie can for instance contain an unquoted "," in the expires field, so we have to use this ad-hoc parser instead of split_header_words. XXX This may not make the best possible effort to parse all the crap that Netscape Cookie headers contain. Ronald Tschalar's HTTPClient parser is probably better, so could do worse than following that if this ever gives any trouble. Currently, this is also used for parsing RFC 2109 cookies. texpirestdomainRtsecuretversiontportsmax-ages;\s*Rt=s\s*=\s*iit0(sexpiressdomainspathssecuresversionsportsmax-ageN(sversionRI( tFalseR-tsplittrangeRR(R)R+tlowerRBtTrueRR*( t ns_headerst known_attrsR0t ns_headerR3t version_settparamstiitparamR=R>tlc((s;/scratch/rashmi/Condor_Script/src/mechanize/_headersutil.pytparse_ns_headers¶s>              cCs%ddl}ddl}|j|ƒS(Niÿÿÿÿ(tdoctestt _headersutilttestmod(RXRY((s;/scratch/rashmi/Condor_Script/src/mechanize/_headersutil.pyt_testìst__main__(t__doc__RR-ttypesRRRt_utilRRR RJRRtcompileR!R$R'R%R R9R?RBRWR[t__name__(((s;/scratch/rashmi/Condor_Script/src/mechanize/_headersutil.pyt s*      U   6