M[0h# f d dl mZ d dlmZ d dlmZmZmZ d dlm Z m
Z
m Z m Z G d de Z
dS ) )
CharSetProber)CodingStateMachine)LanguageFilter ProbingState MachineState) HZ_SM_MODELISO2022CN_SM_MODELISO2022JP_SM_MODELISO2022KR_SM_MODELc f e Zd ZdZd fd Z fdZed Zed Zd Z d Z
xZ S )
EscCharSetProberz
This CharSetProber uses a "code scheme" approach for detecting encodings,
whereby easily recognizable escape or shift sequences are relied on to
identify these encodings.
Nc t t | | g | _ | j t
j z rX| j t t | j t t | j t
j z r,| j t t | j t
j
z r,| j t t d | _ d | _ d | _ d | _ | d S )N) lang_filter)superr
__init__ coding_smr r CHINESE_SIMPLIFIEDappendr r r JAPANESEr
KOREANr active_sm_count_detected_charset_detected_language_statereset)selfr __class__s /builddir/build/BUILDROOT/alt-python311-pip-21.3.1-4.el9.x86_64/opt/alt/python311/lib/python3.11/site-packages/pip/_vendor/chardet/escprober.pyr zEscCharSetProber.__init__* s
%%..;.GGG n ? ? J N ! !"4["A"A B B B N ! !"45G"H"H I I I n 5 5 J N ! !"45G"H"H I I I n 3 3 J N ! !"45G"H"H I I I # !% "&
c t t | | j D ] }|sd|_ | !t | j | _ d | _ d | _ d S )NT) r r
r r activelenr r r )r r r s r r zEscCharSetProber.reset: s
%%++--- I
#I OO "4> 2 2 !% "& r c | j S Nr r s r charset_namez EscCharSetProber.charset_nameE s %%r c | j S r$ )r r& s r languagezEscCharSetProber.languageI s &&r c | j rdS dS )NgGz?g r% r& s r get_confidencez EscCharSetProber.get_confidenceM s ! 44r c |D ]}| j D ]}|r|j s | | }|t j k r?d|_ | xj dz c_ | j dk r t j | _ | j c c S p|t j
k rAt j | _ | | _
|j | _ | j c c S | j S )NFr )r r!
next_stater ERRORr r NOT_MEr stateITS_MEFOUND_ITget_coding_state_machiner r) r )r byte_strcr coding_states r feedzEscCharSetProber.feedS s & &A !^
&
& (8 ( 3 3A 6 6 <#555',I$((A-((+q00&2&9 #z))))) 1 "\%888"."7DK-6-O-O-Q-QD*.7.@D+ :%%%%% 9
& zr r$ ) __name__
__module__ __qualname____doc__r r propertyr' r) r+ r8
__classcell__)r s @r r
r
# s ' ' ' ' ' & & X& ' ' X' r r
N)
charsetproberr codingstatemachiner enumsr r r escsmr r r
r r
r r rD s 8 ) ( ( ( ( ( 2 2 2 2 2 2 = = = = = = = = = =( ( ( ( ( ( ( ( ( ( ( (B B B B B} B B B B Br