python (3.11.7)
e> d Z ddlZddlmZmZmZmZmZmZm Z m
Z
m Z m Z m
Z
mZmZm Z ddlmZ ddlmZ g dZ G d d Z G d d
e Z ed
e G d d e Z ed
e G d de Z ede G d d Z G d d Z e n
# e$ r e ZY nw xY w G d d Z dS )z+
csv.py - read/write/investigate CSV files
N)Error __version__writerreaderregister_dialectunregister_dialect get_dialect
list_dialectsfield_size_limit
QUOTE_MINIMAL QUOTE_ALLQUOTE_NONNUMERIC
QUOTE_NONE__doc__)Dialect)StringIO)r r
r r r r r excel excel_tabr r r r r r
Snifferr r
DictReader
DictWriter unix_dialectc B e Zd ZdZdZdZdZdZdZdZ dZ
dZ dZ d Z
d ZdS )r zDescribe a CSV dialect.
This must be subclassed (see csv.excel). Valid attributes are:
delimiter, quotechar, escapechar, doublequote, skipinitialspace,
lineterminator, quoting.
FNc \ | j t k rd| _ | d S )NT) __class__r _valid _validateselfs 7/BuggyBox/python/3.11.7/bootstrap/lib/python3.11/csv.py__init__zDialect.__init__* s, >W $ $ DK c t | d S # t $ r!}t t | d }~ww xY wN)_Dialect TypeErrorr str)r es r! r zDialect._validate/ sG TNNNNN A-- s
> 9>)__name__
__module__ __qualname__r _namer delimiter quotechar
escapechar doublequoteskipinitialspacelineterminatorquotingr" r r# r! r r sl
E
FIIJKNG
r# r c * e Zd ZdZdZdZdZdZdZe Z
dS )r z;Describe the usual properties of Excel-generated CSV files.,"TF
N) r* r+ r, r r. r/ r1 r2 r3 r r4 r5 r# r! r r 6 s3 EEIIK NGGGr# r c e Zd ZdZdZdS )r zEDescribe the usual properties of Excel-generated TAB-delimited files. N)r* r+ r, r r. r5 r# r! r r @ s OOIIIr# r z excel-tabc * e Zd ZdZdZdZdZdZdZe Z
dS )r z:Describe the usual properties of Unix-generated CSV files.r7 r8 TF
N) r* r+ r, r r. r/ r1 r2 r3 r
r4 r5 r# r! r r E s3 DDIIK NGGGr# r unixc \ e Zd Z ddZd Zed Zej d Zd ZdS ) r Nr c x || _ || _ || _ t ||g|R i || _ || _ d| _ d S Nr ) _fieldnamesrestkeyrestvalr dialectline_num)r f
fieldnamesrC rD rE argskwdss r! r" zDictReader.__init__Q sK % Q7$777$77
r# c | S r% r5 r s r! __iter__zDictReader.__iter__Z s r# c | j + t | j | _ n# t $ r Y nw xY w| j j | _ | j S r% )rB nextr
StopIterationrF r s r! rH zDictReader.fieldnames] sZ #
#' #4#4
,
s #
00c || _ d S r% )rB )r values r! rH zDictReader.fieldnamesg s r# c | j dk r| j t | j }| j j | _ |g k rt | j }|g k t t | j | }t
| j }t
| }||k r||d || j < n"||k r | j |d D ] }| j ||<
|S rA ) rF rH rN r dictziplenrC rD )r rowdlflrkeys r! __next__zDictReader.__next__k s =A OO4; ,
Riit{##C Rii T_c** + +
!
!
XX
77 !"##hAdlOO
"WWrss+
&
& #r# )NNNr ) r* r+ r, r" rL propertyrH setterr[ r5 r# r! r r P s AE X ! ! ! r# r c 2 e Zd Z d
dZd Zd Zd Zd Zd S ) r r raiser c || _ || _ | dvrt d|z || _ t ||g|R i || _ d S )N)r_ ignorez-extrasaction (%s) must be 'raise' or 'ignore')rH rD lower
ValueError extrasactionr )r rG rH rD rd rE rI rJ s r! r" zDictWriter.__init__ sr $ ': : : L + , - -
- ( Q7$777$77 r# c z t t | j | j }| | S r% )rS rT rH writerow)r headers r! writeheaderzDictWriter.writeheader s/ c$/4?;;<<}}V$$$r# c j dk rM j z
}|r/t dd d |D z fd j D S )Nr_ z(dict contains fields not in fieldnames: z, c , g | ]}t | S r5 )repr).0xs r!
<listcomp>z,DictWriter._dict_to_list.<locals>.<listcomp> s -L-L-L!d1gg-L-L-Lr# c 3 N K | ] } |j V d S r% )getrD )rl rZ rowdictr s r! <genexpr>z+DictWriter._dict_to_list.<locals>.<genexpr> s3 JJ3 C ..JJJJJJr# )rd keysrH rc join)r rq wrong_fieldss `` r!
_dict_to_listzDictWriter._dict_to_list s ' '"<<>>DO;L
O !K#'99-L-L|-L-L-L#M#M"N O O OJJJJJ$/JJJJr# c \ | j | | S r% )r rf rv )r rq s r! rf zDictWriter.writerow s& {##D$6$6w$?$?@@@r# c \ | j t | j | S r% )r writerowsmaprv )r rowdictss r! ry zDictWriter.writerows s% {$$S);X%F%FGGGr# N)r r_ r )r* r+ r, r" rh rv rf ry r5 r# r! r r su ?F 8 8 8 8% % %K K KA A AH H H H Hr# r c 2 e Zd ZdZd ZddZd Zd Zd ZdS ) r ze
"Sniffs" the format of a CSV file (i.e. delimiter, quotechar)
Returns a Dialect object.
c g d| _ d S )N)r7 r; ; :) preferredr s r! r" zSniffer.__init__ s 333r# Nc | || \ }}}}|s| || \ }}|st d G d dt }||_ ||_ |pd|_ ||_ |S )zI
Returns a dialect (or None) corresponding to the sample
z Could not determine delimiterc e Zd ZdZdZeZdS ) Sniffer.sniff.<locals>.dialectsniffedr9 N)r* r+ r, r- r3 r r4 r5 r# r! rE r s E #N#GGGr# rE r8 )_guess_quote_and_delimiter_guess_delimiterr r r1 r. r/ r2 )r sample
delimitersr/ r1 r. r2 rE s r! sniffz
Sniffer.sniff s 226:FF < ; +; L*.*?*?@J+L +L 'I' 9788 8 $ $ $ $ $g $ $ $ * % % ,#3 r# c J g }dD ]G}t j |t j t j z }| | }|r nH|sdS i }i }d}|j } |D ]}
| d dz
} |
| } | r | | d dz || < | d dz
} |
| } n# t $ r Y Tw xY w| r"|| |v r | | d dz || < | d dz
} n# t $ r Y w xY w|
| r|dz
}t ||j }
|r+t ||j }|| |k }|d
k rd }nd }d}t j d t j | |
d
z t j }|
| rd}nd}|
|||fS )a
Looks for text enclosed between two identical quotes
(the probable quotechar) which are preceded and followed
by the same character (the probable delimiter).
For example:
,'some text',
The quote with the most wins, same with the delimiter.
If there is no quotechar the delimiter can't be determined
this way.
)zI(?P<delim>[^\w\n"\'])(?P<space> ?)(?P<quote>["\']).*?(?P=quote)(?P=delim)zG(?:^|\n)(?P<quote>["\']).*?(?P=quote)(?P<delim>[^\w\n"\'])(?P<space> ?)zG(?P<delim>[^\w\n"\'])(?P<space> ?)(?P<quote>["\']).*?(?P=quote)(?:$|\n)z-(?:^|\n)(?P<quote>["\']).*?(?P=quote)(?:$|\n))r FNr r quote delimNspacerZ r= r z]((%(delim)s)|^)\W*%(quote)s[^%(delim)s\n]*%(quote)s[^%(delim)s\n]*%(quote)s\W*((%(delim)s)|$))r r TF) recompileDOTALL MULTILINEfindall
groupindexrp KeyErrormaxescapesearch)r datar matchesrestrregexpquotesdelimsspacesr mnrZ r/ r r2 dq_regexpr1 s r! r z"Sniffer._guess_quote_and_delimiter sQ H E Zry2<'?@@F nnT**G
(''&
A7#a'AA$C
5 $jja 0 01 4s
w'!+d
5
*cZ.?.? $jja 0 01 4s
w'!+
t
! FJ/// !FJ///E %e} 6 }} E J (* %(8(8) L L MNPl\ \ D ! ! KK K;/?@@s$ B((
B54B5 C))
C65C6c . t t d| d }d t d D }t dt | }d}i }i }i }d|}
} | t | k r|dz
}|| |
D ]S} |D ]N} | | i }
| | }|
|d dz |
|< |
|| < OT| D ]} t || }t | dk r|d d dk rOt | dk rjt |d || < | || || d || d t d
|D z
f|| < |d || < | }t t ||z t | }d }d }t | dk r^||k rX|D ]7\ }}|d dk r&|d dk r|d |z |k r |||v r|||< 8|d
z }t | dk r||k Xt | dk rft | d }|d | |d d|z k }||fS |
} |
|z
}
| t | k |sdS t | dk ra| j D ]Y}|| v rA|d | |d d|z k }||fc S Zd | D }| |d d }|d | |d d|z k }||fS )a
The delimiter /should/ occur the same number of times on
each row. However, due to malformed data, it may not. We don't want
an all or nothing approach, so we allow for small variations in this
number.
1) build a table of the frequency of each character on every line.
2) build a table of frequencies of this frequency (meta-frequency?),
e.g. 'x occurred 5 times in 10 rows, 6 times in 1000 rows,
7 times in 2 rows'
3) use the mode of the meta-frequency to determine the /expected/
frequency for that character
4) find out how often the character actually meets that goal
5) the character that best meets its goal is the delimiter
For performance reasons, the data is evaluated in chunks, so it can
try and evaluate the smallest portion of the data possible, evaluating
additional chunks as necessary.
Nr= c , g | ]}t | S r5 )chr)rl cs r! rn z,Sniffer._guess_delimiter.<locals>.<listcomp>- s ,,,AQ,,,r#
r r c | d S )Nr r5 )rm s r! <lambda>z*Sniffer._guess_delimiter.<locals>.<lambda>G s
1Q4 r# r c 3 & K | ] }|d V
dS )r Nr5 )rl items r! rr z+Sniffer._guess_delimiter.<locals>.<genexpr>L s& (C(CTa(C(C(C(C(C(Cr# g ?g?g{Gz?z%c )r r c g | ] \ }}||f
S r5 r5 )rl kvs r! rn z,Sniffer._guess_delimiter.<locals>.<listcomp>v s 3335Aa!A333r# )listfiltersplitrangeminrU rp countrs itemsr removesumfloatr sort)r r r ascii chunkLength iteration
charFrequencymodesr startendlinechar
metaFrequencyfreqr modeListtotal consistency thresholdr r r r2 rW s r! r zSniffer._guess_delimiter sf & F4D!1!12233,,s,,, "c$ii((
sc$ii NI U3Y
8
8 ! 8 8D$1$5$5dB$?$?M ::d++D*7*;*;D!*D*Dq*HM$'*7M$''
8 &**,,
+
+ ]4 0 6 6 8 899u::??uQx{a'7'7 u::>>"%e "@"@"@E$K LLt---#(;q>5;q>%((C(CU(C(C(C%C%C4D #EE$KK #((E$KK {{}}H#kI 5s4yyAABBE KIf++""{i'?'? $ * *DAqtaxxAaD1HH qT%ZK 7 7 ' /1
??()F1It#
f++""{i'?'? 6{{a V[[]]++A.$(GMM%$8$8$(GMM%%-$@$@%A /00 E ; Cc c$ii f 7 v;;??^
1
1
%%(,Q
a(8(8(,Q
eai(@(@)A$ /0000 & 43FLLNN333
b ! GMM% 0 0 GMM%%- 8 8 9'((r# c b t t | | | }t | }t | }i }t | D ]}d ||< d}|D ]}|dk r n|dz
}t | |k r#t
| D ]e} t }
|
|| n,# t t f$ r t || }
Y nw xY w|
|| k r|| |
|| < b|| = fd} | D ]}\ } } t | t d k r%t || | k r| dz
} D| dz } J | || | dz } b# t t f$ r | dz
} Y zw xY w| dk S )Nr r )r r r rN rU r r rs complexrc
OverflowErrorr typer' )
r r rdrrg columns columnTypesicheckedrV colthisType hasHeadercolTypes
r!
has_headerzSniffer.has_header s Xf%%tzz&'9'9::cf++ w 66A$Q - -C|| qLG3xx7""K , , . .//
-
-"- HSX&&&&"M2 - - - "3s8}}HHH- {3 ///"3'/+3 C(( ( ,
-$ ' - - / / # #LCG}}Q''vc{##w.. NII NII#GF3K((( NII #I. # # # NIII#
1} s$ :C &C54C58FF('F(r% ) r* r+ r, r r" r r r r r5 r# r! r r sy 4 4 4
:IA IA IAXd) d) d)N= = = = = r# r ) r r _csvr r r r r r r r
r r r
r r r r& ior __all__r r r r r r NameErrorr r r5 r# r! <module>r s
% $ $ $ $ $ > G % i ( ( ( 7 & & &/ / / / / / / /dH H H H H H H H< GG GGGW W W W W W W W W W s -B0 0B:9B:
Browse
Usage
Versions