U
|
M ±d8' ã @ sÐ d dl Z d dlZd dlmZ d dlmZmZmZmZm Z d dl
|
m Z d dl m Z mZ d dlmZ d dlmZ d dlmZ d dlmZ d d lmZ deeed d d Z dee e e dddZ e dkrÌe dS )é N)Údumps)ÚabspathÚbasenameÚdirnameÚjoinÚrealpath)Úpython_version)ÚListÚOptional)Úunidata_version)Úfrom_fp)ÚCliDetectionResult)Ú __version__Úyes)ÚquestionÚdefaultÚreturnc C s dddddd}|dkr d}n(|dkr,d}n|dkr:d }n t d
|
| tj | | ¡ t ¡ }|dk rx|d krx|| S ||kr|| S tj d ¡ qFdS ) aÝ Ask a yes/no question via input() and return their answer.
|
|
"question" is a string that is presented to the user.
|
"default" is the presumed answer if the user just hits <Enter>.
|
It must be "yes" (the default), "no" or None (meaning
|
an answer is required of the user).
|
|
The "answer" return value is True for "yes" or False for "no".
|
|
Credit goes to (c) https://stackoverflow.com/questions/3041986/apt-command-line-interface-like-yes-no-input
|
TF)r ÚyZyeÚnoÚnNz [y/n] r z [Y/n] r z [y/N] z invalid default answer: '%s'Ú z3Please respond with 'yes' or 'no' (or 'y' or 'n').
|
)Ú
|
ValueErrorÚsysÚstdoutÚwriteÚinputÚlower)r r ZvalidÚpromptÚchoice© r úXd:\z\workplace\vscode\pyvenv\venv\Lib\site-packages\charset_normalizer/cli/normalizer.pyÚ query_yes_no s
|
r! )Úargvr c
|
s¾ t jdd}|jdt d¡ddd |jdd d
|
d d d d |jddd
|
d ddd |jddd
|
d ddd |jddd
|
d ddd |jdd d
|
d d d d |jd d d
|
d d!d"d |jd#d$d%d&td'd(d) |jd*d+d, tt tt j
|
¡ d-¡rèd.nd/¡d0d1 | | ¡}|jd2kr(|jd kr(td3tjd4 d5S |jd2krR|jd krRtd6tjd4 d5S |jd7k sj|jd8kr|td9tjd4 d5S g }|jD ]Ät|j|jd:}| ¡ d;krtd< j|jd8k rÊd=nd>¡tjd4 | tt jd;g g d?g d d8d7d;d2 ¡ n2| tt j j j fd@dA j D j j! j" j# j$d;d2 ¡ t%|d5kr¼|j&r¼|D ]V krd| tt jj j fdBdAj D j j!j"j#j$d;d ¡ qd|jd2kr6 j 'dC¡d2krtdD j¡tjd4 j(d kr )¡ qt*t+j}t,t+j}| -dE¡}|jd kr^| .dF j ¡ j(d kr )¡ n<|jd krt/dG j¡dHd krj(d kr )¡ qzHt0|dE 0|¡|dI _1t2|dI j1dJdKdL}| 3t4 ¡ W 5 Q R X W nR t5k
|
r4 } z2tt4| tjd4 j(d kr )¡ W Y ¢ dMS d;} ~ X Y nX j(d kr )¡ q|j6d krtt7t%|d5krzdNdA |D n|dI j8d2dOdP n*|jD ]"tdQ 0fdRdA|D ¡ qdIS )Sz
|
CLI assistant using ARGV and ArgumentParser
|
:param argv:
|
:return: 0 if everything is fine, anything else equal trouble
|
zvThe Real First Universal Charset Detector. Discover originating encoding used on text file. Normalize text to unicode.)Ú descriptionÚfilesÚrbú+zFile(s) to be analysed)ÚtypeÚnargsÚhelpz-vz --verboseÚ
|
store_trueFÚverbosezjDisplay complementary information about file if any. Stdout will contain logs about the detection process.)Úactionr Údestr) z-az--with-alternativeÚ alternativeszIOutput complementary possibilities if any. Top-level JSON WILL be a list.z-nz --normalizeÚ normalizezLPermit to normalize input file. If not set, program does not write anything.z-mz --minimalÚminimalzBOnly output the charset detected to STDOUT. Disabling JSON output.z-rz --replaceÚreplacezGReplace file when trying to normalize it instead of creating a new one.z-fz--forceÚforcezHReplace file without asking if you are sure, use this flag with caution.z-tz --thresholdÚstoregÉ?Ú thresholdzUDefine a custom maximum amount of chaos allowed in decoded content. 0. <= chaos <= 1.)r, r r' r- r) z --versionÚversionz;Charset-Normalizer {} - Python {} - Unicode {} - SpeedUp {}z.pyZOFFÚONz"Show version information and exit.)r, r5 r) Tz.Use --replace in addition of --normalize only.)Úfileé z*Use --force in addition of --replace only.g g ð?z---threshold VALUE should be between 0. AND 1.)r4 ÚexplainNz4Unable to identify originating encoding for "{}". {}z-Maybe try increasing maximum amount of chaos.r ÚUnknownc s g | ]}| j kr|qS r ©Úencoding©Ú.0Úcp)Ú
|
best_guessr r Ú
|
<listcomp>¹ s
|
þz cli_detect.<locals>.<listcomp>c s g | ]}| j kr|qS r r; r= )Úelr r rA Ð s
|
þÚutfzJ"{}" file does not need to be normalized, as it already came from unicode.Ú.éÿÿÿÿz0Are you sure to normalize "{}" by replacing it ?r r Úwzutf-8r; é c S s g | ]
|
}|j qS r )Ú__dict__©r> rB r r r rA s é )Ú ensure_asciiÚindentz, c s& g | ] }|j t jkr|jp d qS )Z undefined)Úpathr Únamer< rI )Úmy_filer r rA s þ)9ÚargparseÚArgumentParserÚ add_argumentÚFileTypeÚfloatÚformatr r r Ú md_moduleÚ__file__r ÚendswithÚ
|
parse_argsr1 r/ Úprintr Ústderrr2 r4 r$ r r+ ÚbestrN Úappendr r r< Zencoding_aliasesZcould_be_from_charsetÚlanguageZ alphabetsÚbomZ percent_chaosZpercent_coherenceÚlenr. Ú
|
startswithÚclosedÚcloser r r ÚsplitÚinsertr! r Z unicode_pathÚopenr ÚstrÚIOErrorr0 r rH )
|
r" ÚparserÚargsZx_ÚmatchesZdir_pathÚ file_nameZo_ÚfpÚer )r@ rB rO r Ú
|
cli_detect0 sª ÿ ÿú úúúúúù ü÷
|
|
ÿüù õÿ
|
þñÿ
|
|
þñÿ ÿü
|
|
ÿÿüúþ
|
$ýÿ
|
|
þÿÿ
|
ro Ú__main__)r )N) rP r Újsonr Úos.pathr r r r r Úplatformr Útypingr r
|
Ú unicodedatar Zcharset_normalizer.mdZmdrV Zcharset_normalizerr Zcharset_normalizer.modelsr Zcharset_normalizer.versionr rg Úboolr! Úintro Ú__name__r r r r Ú<module> s ! x
|
