gE dZddlmZddlZddlZddlZddlZddlmZdZ dZ dZ ejd Z ejd jd d d ddZejdjdddZejd ejd ejd ejdejdejdejdejddZejdZejdZGddej(Zd$d%dZd&dZd'dZd(dZd)dZd*d Zd+d,d!Zd+d,d"Zd-d#Zy).a An implementation of `urlparse` that provides URL validation and normalization as described by RFC3986. We rely on this implementation rather than the one in Python's stdlib, because: * It provides more complete URL validation. * It properly differentiates between an empty querystring and an absent querystring, to distinguish URLs with a trailing '?'. * It handles scheme, hostname, port, and path normalization. * It supports IDNA hostnames, normalizing them to their encoded form. * The API supports passing individual components, as well as the complete URL string. Previously we relied on the excellent `rfc3986` package to handle URL parsing and validation, but this module provides a simpler alternative, with less indirection required. ) annotationsN) InvalidURLizBABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-._~z !$&'()*+,;=z%[A-Fa-f0-9]{2}z(?:(?P{scheme}):)?(?://(?P{authority}))?(?P{path})(?:\?(?P{query}))?(?:#(?P{fragment}))?z([a-zA-Z][a-zA-Z0-9+.-]*)?z[^/?#]*z[^?#]*z[^#]*z.*scheme authoritypathqueryfragmentzA(?:(?P{userinfo})@)?(?P{host}):?(?P{port})?z(\[.*\]|[^:@]*))userinfohostportz[^@]*z(\[.*\]|[^:]*))rrr r r r r rz ^[0-9]+\.[0-9]+\.[0-9]+\.[0-9]+$z^\[.*\]$ceZdZUded<ded<ded<ded<ded<ded <ded <edd Zedd Zdd ZddZy) ParseResultstrrr r int | Nonerr str | Noner r cdj|jr|jdndd|jvrd|jdn |j|jd|jgSdgS)N@:[])joinr r rselfs D/var/www/openai/venv/lib/python3.12/site-packages/httpx/_urlparse.pyrzParseResult.authorityjszww'+}}4==/#"$'499$4!DII;a $))#'99#8!DII;   ?A   cdjd|jvrd|jdn |j|jd|jgSdgS)Nrrrr)rr rrs rnetloczParseResult.netloctsdww$'499$4!DII;a $))#'99#8!DII;   ?A   rc |s|S|j|j|j|j|jd}|j |t di|S)Nrr)rrr r r updateurlparse)rkwargsdefaultss r copy_withzParseResult.copy_with}sRKkkIIZZ   'h''rc|j}dj|jr|jdnd|rd|nd|j|jd|jnd|j d|j gSdgS)Nrr//?#)rrrr r r )rrs r__str__zParseResult.__str__sNN ww%)[[4;;-q!b$-"YK 2 $(JJ$:!DJJ< '+}}'@!DMM?#    GI    rN)returnr)r%rr-r) __name__ __module__ __qualname____annotations__propertyrr r'r,rrrrasX KM I  I       (  rrc Nt|tkDr tdtd|Dr7t d|D}|j |}d|d|d}t|d|vr%|d}t |tr t|n||d<d|vr0|jdxsd }|jd \|d <}|d<d |vsd |vrLt|jd d xsd }t|jd d xsd } | r|d | n||d<d|vr7|jdxsd } | jd\|d<} |d<| sd|d<d |vrD|jd xsd } d | vr+| jdr| jds d| d|d <|jD]\} }| t|tkDrtd| dtd|Dr:t d|D}|j |}d| d|d|d}t|t | j#|rtd| dt$j'|}|J|j)}|jd|dxsd }|jd|dxsd }|jd|dxsd }|jd|d}|jd|d}t*j'|}|J|j)}|jd|dxsd }|jd |d xsd } |jd|d}|j-}t|t.d z }t1| }t3||}|d k7}|d k7xs |d k7xs|du}t5|||!|s|r t7|}d"}t|t.|zd#z } |dnt|t.|zd$z }!|dnt|t.|zd%z }"t9||||| |!|"S)&Nz URL too longc3bK|]'}|jxr|j )ywNisascii isprintable.0chars r zurlparse..s+ E4<<> 4$"2"2"44 4-/c3dK|](}|js|jr%|*ywr6r7r:s rr=zurlparse..s$VSTDLLN4CSCSCUDS 000z.Invalid non-printable ASCII character in URL, z at position .rr rrr usernamepasswordr raw_pathr*r r rrzURL component 'z ' too longc3bK|]'}|jxr|j )ywr6r7r:s rr=zurlparse..s+O4<<><$*:*:*<&<<r>c3dK|](}|js|jr%|*ywr6r7r:s rr=zurlparse..s'%*TdllnTEUEUEWDUr@z-Invalid non-printable ASCII character in URL z component, zInvalid URL component ''rrr safe) has_scheme has_authorityz`{}%|^\"z:/[]@z:/?[]@z:/?#[]@)lenMAX_URL_LENGTHranynextfind isinstanceintrpop partitionquoteget startswithendswithitemsCOMPONENT_REGEX fullmatch URL_REGEXmatch groupdictAUTHORITY_REGEXlower SUB_DELIMS encode_hostnormalize_port validate_pathnormalize_pathr)#urlr%r<idxerrorrr _rBrCrD seperatorr keyvalue url_matchurl_dictrrr r r authority_matchauthority_dictr parsed_schemeparsed_userinfo parsed_host parsed_portrJrK WHATWG_SAFE parsed_path parsed_queryparsed_fragments# rr$r$s  3x. (( E EEVSVVhhtnVzV3J39r:J39r:9Az8*5xzV::j)/R5=5G5G5L2v 6'?"F7Ozz&!'R $; 4s9K a[F6N lln U  5zN* ?3%z!BCCOOO%*jj&CC5 hmC53!''#3'11%8 #:3%q!ABB)%.$I   ""$HZZ(8"4 5 ;F ;(=>D"I ::fhv. / 52D JJw 1 2Ezz*hz&:;H&++I6O  && &$..0Nzz*nZ&@AGRH ::fnV4 5 ;D ::fnV4 5D  M  S0@AO"4(K,T6:K"$J2M!2Mk6M$:]K]d# KT [(@7(JKK = 5zK7(B C   8*{":Y"F G  rct|sytj|r tj||St j|r tj|dd|ddS|jr$d}t|jt|zS tj|jjdS#tj$rt d|wxYw#tj$rt d|wxYw#tj$rt d |wxYw) NrzInvalid IPv4 address: rzInvalid IPv6 address: z"`{}%|\rHasciizInvalid IDNA hostname: )IPv4_STYLE_HOSTNAMEr] ipaddress IPv4AddressAddressValueErrorrIPv6_STYLE_HOSTNAME IPv6Addressr8rUr`raidnaencodedecode IDNAError)r rus rrbrb/s2   " "4 (  @  ! !$ '  " "4 ( @  ! !$q* -Abz  ! TZZ\ [(@AA={{4::<(//88=** @5dX>? ? @** @5dX>? ? @ >>=24(;<<=s#C C01D "C-0"D"D7c||dk(ry t|}ddddddj|}||k(ry|S#t$rtd|wxYw)NrzInvalid port: Pi)ftphttphttpswswss)rR ValueErrorrrV)rr port_as_int default_ports rrcrc^sv |trz4$i rCr#NRRLl"  4>$2334s 4A c|r|r|jds td|s<|s9|jdr td|jdr tdyyy)z Path validation rules that depend on if the URL contains a scheme or authority component. See https://datatracker.ietf.org/doc/html/rfc3986.html#section-3.3 /z7For absolute URLs, path must be empty or begin with '/'r)z3Relative URLs cannot have a path starting with '//'rz2Relative URLs cannot have a path starting with ':'N)rWr)r rJrKs rrdrdysh ,VW W m ??4 RS S ??3 QR R ,:rcd|vr|S|jd}d|vrd|vr|Sg}|D]9}|dk(r |dk(r|s|dgk7s|j)|j|;dj|S)z Drop "." and ".." segments from a URL path. For example: normalize_path("/path/./to/somewhere/..") == "/path/to" rArz..r)splitrSappendr)r componentsoutput components rreres $ CJ *Z!7 F    $ &RD. MM) $  88F rcpdj|jdDcgc]}d|d c}Scc}w)Nrzutf-8%02X)rr)stringbytes rPERCENTrs6 77v}}W/EF/EtaSzN/EF GGFs 3ct|z}|j|s|Sdj|Dcgc]}||vr|n t|c}Scc}w)z1 Use percent-encoding to quote a string. r)UNRESERVED_CHARACTERSrstriprr)rrINON_ESCAPED_CHARSr<s rpercent_encodedrsZ.4 ==* + 77JPQ&$** =&Q QsA cg}d}tjt|D]l}|j|j }}|j d}||k7r!|||}|j t|||j ||}n|t|k7r!||d} |j t| |dj|S)a Use percent-encoding to quote a string, omitting existing '%xx' escape sequences. See: https://www.rfc-editor.org/rfc/rfc3986#section-2.1 * `string`: The string to be percent-escaped. * `safe`: A string containing characters that may be treated as safe, and do not need to be escaped. Unreserved characters are always treated as safe. See: https://www.rfc-editor.org/rfc/rfc3986#section-2.3 rrHNr) refinditerPERCENT_ENCODED_REGEXstartendgrouprrrLr) rrIpartscurrent_positionr]start_position end_position matched_text leading_text trailing_texts rrUrUs E2F;',{{}eiik {{1~ - -!"2>BL LLDA B  \"'<3v;&/01  _]>? 775>rc dj|Dcgc]#\}}t|ddzt|dz%c}}Scc}}w)am We can use a much simpler version of the stdlib urlencode here because we don't need to handle a bunch of different typing cases, such as bytes vs str. https://github.com/python/cpython/blob/b2f7b2ef0b5421e01efb8c7bee2ef95d3bab77eb/Lib/urllib/parse.py#L926 Note that we use '%20' encoding for spaces. and '%2F for '/'. This is slightly different than `requests`, but is the behaviour that browsers use. See - https://github.com/encode/httpx/issues/2536 - https://github.com/encode/httpx/issues/2721 - https://docs.python.org/3/library/urllib.parse.html#urllib.parse.urlencode &rrH=)rr)rYkvs r urlencodersS 88 1 AB '# -0K K   s(A r")rfrr%rr-r)r rr-r)rzstr | int | Nonerrr-r)r rrJboolrKrr-None)r rr-r)rrr-r)r)rrrIrr-r)rYzlist[tuple[str, str]]r-r)__doc__ __future__rr}rtypingr _exceptionsrrMrracompilerformatr\r_rZr|r NamedTuplerr$rbrcrdrerrrUrr3rrrs$# #I " #45 BJJ *  f+     &"**T f     bjj56I& BJJx RZZ  4  7# BJJ) * BJJt  !bjj!DE bjj-4 &##4 nTn,=^6S2>H Br