idJddlZddlZddlmZddlmZmZddlmZm Z m Z m Z m Z ddl mZmZmZmZddlmZmZmZmZddlmZdd lmZmZdd lmZmZmZm Z m!Z!m"Z"m#Z#ej$d Z%ej&Z'e'(ej)d  d"de*de+de+de,de e e-de e e-de.de.defdZ/ d"de de+de+de,de e e-de e e-de.de.defdZ0 d"ddde+de+de,de e e-de e e-de.de.defd Z1 d#ddde+de+de,de e e-de e e-de.defd!Z2dS)$N)PathLike)basenamesplitext)AnyBinaryIOListOptionalSet)coherence_ratioencoding_languagesmb_encoding_languagesmerge_coherence_ratios)IANA_SUPPORTEDTOO_BIG_SEQUENCETOO_SMALL_SEQUENCETRACE) mess_ratio) CharsetMatchCharsetMatches)any_specified_encodingcut_sequence_chunks iana_nameidentify_sig_or_bom is_cp_similaris_multi_byte_encodingshould_strip_sig_or_bomcharset_normalizerz)%(asctime)s | %(levelname)s | %(message)s皙?TF sequencessteps chunk_size threshold cp_isolation cp_exclusionpreemptive_behaviourexplainreturnc t|ttfs/tdt ||rJt j}t tt tt|} | dkrt d|rEt tt |p tjt#t%|dddgdgS|At td d |d |D}ng}|At td d |d |D}ng}| ||zkr't td||| d}| }|dkr| |z |krt+| |z }t|t,k} t|t.k} | r4t td| n5| r3t td| g} |rt1|nd} | 6| | t td| t5}g}g}d}d}d}t#}t7|\}}|D| |t tdt||| dd| vr| d| t8zD]}|r||vr |r||vr||vr||d}||k}|ot=|}|dvr$|s"t td|l t?|}n8#t@tBf$r$t td|YwxYw | rS|durOtE|dur|dt+dn#|t|t+d|n,tE|dur|n|t|d|}nx#tFtHf$rd}t|tHs/t td|tE|||Yd}~d}~wwxYwd}|D]}tK||rd}n|r$t td||tM|sdnt|| t+| |z }|o|duot|| k} | r!t td|t+t|dz }!tO|!d}!d}"d}#g}$g}% tQ||||||||| D]Y}&|$|&|%tS|&||%d |kr|"dz }"|"|!ks|r|durnZnJ#tF$r=}t td!|tE||!}"d}#Yd}~nd}~wwxYw|#s| r|s |t+d"d*|d#$n\#tF$rO}t td%|tE|||Yd}~d}~wwxYw|%rtW|%t|%z nd}'|'|ks|"|!kr}||t td&||"tY|'d'zd()|dd| fvr*|#s(t%|||dg|}(|| kr|(}n |dkr|(}n|(}t td*|tY|'d'zd()|st[|})nt]|})|)rAt td+|tE|)g}*|dkrB|$D]?}&t_|&d,|)rd-|)nd}+|*|+@ta|*},|,r4t td.|,||t%|||'||,||| ddfvrt|'d,krnt d/||r9t tt |t#||gcS||krnt d0||r9t tt |t#||gcSt|dkr|s|s|r t td1|r6t d2|j1||n{|r||r|r|j2|j2ks|0t d3||n1|r/t d4|||rDt d5|3j1t|dz nt d6|r9t tt ||S)7ae Given a raw bytes sequence, return the best possibles charset usable to render str objects. If there is no results, it is a strong indicator that the source is binary/not text. By default, the process will extract 5 blocs of 512o each to assess the mess and coherence of a given sequence. And will give up a particular code page after 20% of measured mess. Those criteria are customizable at will. The preemptive behavior DOES NOT replace the traditional detection workflow, it prioritize a particular code page but never take it for granted. Can improve the performance. You may want to focus your attention to some code page or/and not others, use cp_isolation and cp_exclusion for that purpose. This function will strip the SIG in the payload/sequence every time except on UTF-16, UTF-32. By default the library does not setup any handler other than the NullHandler, if you choose to set the 'explain' toggle to True it will alter the logger configuration to add a StreamHandler that is suitable for debugging. Custom logging format and handler can be set manually. z4Expected object of type bytes or bytearray, got: {0}rzzfrom_bytes..["DDD "e,,DDDzacp_exclusion is set. use this flag for debugging purpose. limited list of encoding excluded : %s.c.g|]}t|dSr/r0r1s r4r5zfrom_bytes..fr6r7z^override steps (%i) and chunk_size (%i) as content does not fit (%i byte(s) given) parameters.r z>Trying to detect encoding from a tiny portion of ({}) byte(s).zIUsing lazy str decoding because the payload is quite large, ({}) byte(s).z@Detected declarative mark in sequence. Priority +1 given for %s.zIDetected a SIG or BOM mark on first %i byte(s). Priority +1 given for %s.ascii>utf_16utf_32z[Encoding %s wont be tested as-is because it require a BOM. Will try some sub-encoder LE/BE.z2Encoding %s does not provide an IncrementalDecodergA)encodingz9Code page %s does not fit given bytes sequence at ALL. %sTzW%s is deemed too similar to code page %s and was consider unsuited already. Continuing!zpCode page %s is a multi byte encoding table and it appear that at least one character was encoded using n-bytes.zaLazyStr Loading: After MD chunk decode, code page %s does not fit given bytes sequence at ALL. %sgj@strict)errorsz^LazyStr Loading: After final lookup, code page %s does not fit given bytes sequence at ALL. %szc%s was excluded because of initial chaos probing. Gave up %i time(s). Computed mean chaos is %f %%.d)ndigitsz=%s passed initial chaos probing. Mean measured chaos is %f %%z&{} should target any language(s) of {}g?,z We detected language {} using {}z.Encoding detection: %s is most likely the one.zoEncoding detection: %s is most likely the one as we detected a BOM or SIG within the beginning of the sequence.zONothing got out of the detection process. Using ASCII/UTF-8/Specified fallback.z7Encoding detection: %s will be used as a fallback matchz:Encoding detection: utf_8 will be used as a fallback matchz:Encoding detection: ascii will be used as a fallback matchz]Encoding detection: Found %s as plausible (best-candidate) for content. With %i alternatives.z=Encoding detection: Unable to determine any suitable charset.)4 isinstance bytearraybytes TypeErrorformattypeloggerlevel addHandlerexplain_handlersetLevelrlendebug removeHandlerloggingWARNINGrrlogjoinintrrrappendsetrraddrrModuleNotFoundError ImportErrorstrUnicodeDecodeError LookupErrorrrangemaxrrdecodesumroundr rr rr< fingerprintbest)-r"r#r$r%r&r'r(r)previous_logger_levellengthis_too_small_sequenceis_too_large_sequenceprioritized_encodingsspecified_encodingtestedtested_but_hard_failuretested_but_soft_failurefallback_ascii fallback_u8fallback_specifiedresults sig_encoding sig_payload encoding_ianadecoded_payloadbom_or_sig_availablestrip_sig_or_bomis_multi_byte_decoderesimilar_soft_failure_testencoding_soft_failedr_multi_byte_bonusmax_chunk_gave_upearly_stop_countlazy_str_hard_failure md_chunks md_ratioschunkmean_mess_ratiofallback_entrytarget_languages cd_ratioschunk_languagescd_ratios_mergeds- r4 from_bytesr#s<8 i)U!3 4 4  B I IY     %+\/***i..F {{ STTT  F   1 1 1 OO1DW_ E E E|IwUBPRSSTUUU  5 IIl # #    ED|DDD    6 IIl # #    ED|DDD   *u$%%  l         qyyVe^j00%(( "%i..3E"E"%i..4D"D    L S S         W ^ ^     (*.BKy)))t%$$%7888  N    uuF)+)+-1N*.K15,..G 3I > >L+$$\222  W          )))+++$$W---.?E<E<  M==   M\99  F " "  =!!!)-%1]%B!5" :Q ; ;  0 0 09M 0 JJm     *@*O*O ! !#[1    JJD    H   $ )>%)G)G'500kD k**"3{#3#3c$ii#?@* #&'500I"3{#3#3#5#56* ### #K0   a--  O!FF  $ * *= 9 9 9 HHHH +0!$;   ],@AA ,0)  %  JJi$      ) ?AAs;/?/?       " .t+ .O$$v-    JJ-    "%SWWq[!1!1 1155 ! %!   ),$ %      '''  E9!=!=>>>R=I--$)$$(999(:-=-F-FE! ) ) ) JJsA     1 $( ! ! ! ! ! ! )& % *  #d))++&--mH-MMMM%    t!FF  (..}=== EN!VY#i..!@!@SV i ' '+;?P+P+P # * *= 9 9 9 JJ0 o+Q777    '74F!GGG-H".}iO""!$666)7&&"g--%3NN"0K   K  /C' 3 3 3    % D*<]*K*K  4]CC    JJ8??!3'7#8#8     G # #" 2 2"13>N X)9 : : :TX##  11111)<<   JJ299$m     $      0'7C C C#%% LL@-    7$$_555 5666!7=#9":;; ; ; ; L ( ( LL1     7$$_555 5666!7=#9":;; ; ; ; ) 7||q  . ,>  JJa     + LLI"+    NN- . . . .  ++34#4 +~/III' LLU V V V NN; ' ' ' '  + LLU V V V NN> * * *V k LLNN # LL1      TUUU/_----... Ns^P&&2QQBS!!U2AUUA1Z22 [9<3[44[9,\00 ^ :A^^ fpc Rt||||||||S)z Same thing than the function from_bytes but using a file pointer that is already ready. Will not close the file pointer. )rread)rr#r$r%r&r'r(r)s r4from_fprs6       r7pathz PathLike[Any]c t|d5}t||||||||cdddS#1swxYwYdS)z Same thing than the function from_bytes but with one extra step. Opening and reading given file path in binary mode. Can raise IOError. rbN)openr) rr#r$r%r&r'r(r)rs r4 from_pathrs dD    R                             s 488c tjdtt|||||||}t |}t t |} t|dkr"td || } | dxxd| j zz cc<td t||d| d5} | | dddn #1swxYwY| S) zi Take a (text-based) file path and try to create another file next to it, this time using UTF-8. z2normalize is deprecated and will be removed in 3.0rz;Unable to normalize "{}", no encoding charset seems to fit.-z{}r-wbN)warningswarnDeprecationWarningrrlistrrQIOErrorrJrgr<rr^replacerWwriteoutput) rr#r$r%r&r'r(rtfilenametarget_extensionsresultrs r4 normalizers M<   G~~HXh//00 7||q I P P     \\^^FaC&/11  CII%%h8I0J0JKKLLd  "  !!!""""""""""""""" Ms(EE E )rr r!NNTF)rr r!NNT)3rTrosros.pathrrtypingrrrr r cdr r rrconstantrrrrmdrmodelsrrutilsrrrrrrr getLoggerrL StreamHandlerrO setFormatter FormatterrHrXfloatr^boolrrrrr7r4rs &&&&&&&&55555555555555 RQQQQQQQQQQQ00000000  / 0 0'''))GABB(,(,!%   49%  49% H(,(,!%   49%  49% 8(,(,!%       49%  49%        :(,(,!%.. . .. . 49% . 49% ........r7