U dd> @sdZddlZddlZddlZddlZddlmZdZdZdZ e dZ e d j d d d d ddZ e dj ddddZe d e d e d e d e de de de ddZe dZe dZGdddejZd3eejeedddZeedddZejejeefeejed d!d"Zeeedd#d$d%Zeed&d'd(Zeed)d*d+Zd4eeed-d.d/Zej ej!eefed0d1d2Z"dS)5a An implementation of `urlparse` that provides URL validation and normalization as described by RFC3986. We rely on this implementation rather than the one in Python's stdlib, because: * It provides more complete URL validation. * It properly differentiates between an empty querystring and an absent querystring, to distinguish URLs with a trailing '?'. * It handles scheme, hostname, port, and path normalization. * It supports IDNA hostnames, normalizing them to their encoded form. * The API supports passing individual components, as well as the complete URL string. Previously we relied on the excellent `rfc3986` package to handle URL parsing and validation, but this module provides a simpler alternative, with less indirection required. N) InvalidURLizBABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-._~z !$&'()*+,;=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[^@]*z(\[.*\]|[^:]*))userinfohostport)rrrrr r r r z^[0-9]+.[0-9]+.[0-9]+.[0-9]+$z^\[.*\]$c@seZdZUeed<eed<eed<ejeed<eed<ejeed<ejeed<eedd d Z eedd d Z ejedd ddZ edddZ dS) ParseResultrr r r rrr )returncCsRd|jr|jdndd|jkr0d|jdn|j|jdk rJd|jndgS)N@:[])joinr r r selfr3/tmp/pip-unpacked-wheel-fkuvgui5/httpx/_urlparse.pyrgs zParseResult.authoritycCs>dd|jkrd|jdn|j|jdk r6d|jndgS)Nrrrr)rr r rrrrnetlocqs zParseResult.netloc)kwargsrcKs6|s|S|j|j|j|j|jd}||td|S)Nrr)r)rrrrr updateurlparse)rrdefaultsrrr copy_withzs zParseResult.copy_withcCsh|j}d|jr|jdnd|r,d|nd|j|jdk rHd|jnd|jdk r`d|jndgS)Nrr//?#)rrrrrr )rrrrr__str__szParseResult.__str__N) __name__ __module__ __qualname__str__annotations__typingOptionalintpropertyrrrr"rrrrr ^s  r r)urlrrcKst|tkrtdtdd|Dr.tdd|krX|d}t|trPt|n||d<d|kr|dpld}|d\|d <}|d<d |ksd |krt |d dpd}t |d dpd}|r|d|n||d <d |kr|d pd}|d\|d<}|d<|sd|d<d |krb| d p.d} d| krb| drR| dsbd| d|d <| D]v\} } | dk rjt| tkrtd| dtdd| Drtd| dt| | sjtd| dqjt|} | dk st| } | d| dpd}| d| dp,d}| d| dpBd}| d| d}| d| d}t|}|dk s|t|}| d |d pd}| d |d pd} | d|d}|}t |tdd}t| }t||}|dk}|dkp |dkp |dk }t|||d|r*t|}t |tdd}|dkrHdnt |tdd}|dkrfdnt |tdd}t|||||||S) Nz URL too longcss |]}|o| VqdSNisascii isprintable.0charrrr szurlparse..z,Invalid non-printable ASCII character in URLr rrrr usernamepasswordr raw_pathr rrrrzURL component 'z ' too longcss |]}|o| VqdSr-r.r1rrrr4sz8Invalid non-printable ASCII character in URL component ''zInvalid URL component 'rrr safe) has_scheme has_authorityz:@/z/?)lenMAX_URL_LENGTHrany isinstancer*r&pop partitionquoteget startswithendswithitemsCOMPONENT_REGEX fullmatch URL_REGEXmatchAssertionError groupdictAUTHORITY_REGEXlower SUB_DELIMS encode_hostnormalize_port validate_pathnormalize_pathr )r,rr r_r5r6r7Z seperatorr keyvalueZ url_matchZurl_dictrrrrr Zauthority_matchZauthority_dictr Z parsed_schemeZparsed_userinfoZ parsed_hostZ parsed_portr;r<Z parsed_pathZ parsed_queryZparsed_fragmentrrrrs   "     r)r rcCs|sdSt|rDzt|Wntjk r>tdYnX|St|rzt|ddWntjk rtdYnX|ddS|rt | t dSzt | dWSt jk rtdYnXdS) NrzInvalid IPv4 addressrzInvalid IPv6 addressr9asciizInvalid IDNA hostname)IPv4_STYLE_HOSTNAMErK ipaddress IPv4AddressAddressValueErrorrIPv6_STYLE_HOSTNAME IPv6Addressr/rCrOrPidnaencodedecode IDNAError)r rrrrQs(   rQ)r rrcCsd|dks|dkrdSz t|}Wntk r<tdYnXdddddd|}||kr`dS|S)Nrz Invalid portPi)ftphttphttpswswss)r* ValueErrorrrD)r rZ port_as_int default_portrrrrRCs  rR)rr;r<rcCsH|r|rD|dsDtdn(|dr.td|drD|sDtddS)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 '/'rzFURLs with no authority component cannot have a path starting with '//'rzBURLs with no scheme component cannot have a path starting with ':'N)rEr)rr;r<rrrrS`s  rS)rrcCsV|d}g}|D]8}|dkr q|dkr@|rJ|dgkrJ|q||qd|S)z Drop "." and ".." segments from a URL path. For example: normalize_path("/path/./to/somewhere/..") == "/path/to" rm.z..r)splitrAappendr)r componentsoutput componentrrrrTzs   rT)r3rcCsddd|dDS)a Replace every character in a string with the percent-encoded representation. Characters outside the ASCII range are represented with their a percent-encoded representation of their UTF-8 byte sequence. For example: percent_encode(" ") == "%20" rcSsg|]}d|dqS)%Z02xr)r2byterrr sz"percent_encode..zutf-8)rraupper)r3rrrpercent_encodes rxrm)stringr:rcs@t||dtt|kr(d7dfdd|DS)Nrtrcs g|]}|kr|nt|qSr)rxr1ZNON_ESCAPED_CHARSrrrvszquote..)UNRESERVED_CHARACTERScountr=PERCENT_ENCODED_REGEXfindallr)ryr:rrzrrCs rC)rGrcCsddd|DS)N&cSs$g|]\}}t|dt|qS)=)rC)r2kvrrrrvszurlencode..)r)rGrrr urlencodes r)r)rm)#__doc__r[rer(r` _exceptionsrr>r{rPcompiler}formatrJrNrHrZr^ NamedTupler r&r)rrQUnionr*rRboolrSrTrxrCListTuplerrrrrsh    7/