cVHdZdZddlZddlZddlmZ ddlmZmZddlmZddlmZddlmZmZmZmZmZmZmZ dd lmZdd lmZid d ddddddddddddddddddd d!d"d#d$dd%d&d'd(d)d*Z d~d+e d,eed-ee d.e fd/Z!d0e d.e fd1Z"ejd2Z#ejd3Z$dd4e d5e%d6e d7e&d.e f d8Z'dd4e d5e%d6e d.e fd9Z(dd4e d5e%d6e d.e fd:Z)d4e d5e%d.ee e ffd;Z*dd0e de+Z,d?Z-d@Z.dAZ/eje/Z0dBZ1dCZ2ejdDe1dEe2dFZ3ejdGe1dHe2dIe1dJZ4dKdLdMdNdOdPZ5dQdKdRdSdTdUZ6de7e0e3fd0e dVee e%fdWeede fdXe dYedZed.ee7e%ffd[Z8ejd\Z9dd]ed0e d^e d.e fd_Z:d`Z;daZdcdkdldmdndodpdqdrds Z?d~dtee duee d.e fdvZ@d~dwe dtee duee d.e fdxZAdydzd{d|ZBejeBfd}ZDy#e $rWe ejjddZe ejjddjdZYLwxYw#e $rd ZY5wxYw)aSome text manipulation utility functions. :group text formatting: normalize_text, normalize_paragraph, pretty_match,unquote, colorize_ansi :group text manipulation: searchall, splitstrip :sort: text formatting, text manipulation :type ANSI_STYLES: dict(str) :var ANSI_STYLES: dictionary mapping style identifier to ANSI terminal code :type ANSI_COLORS: dict(str) :var ANSI_COLORS: dictionary mapping color identifier to ANSI terminal code :type ANSI_PREFIX: str :var ANSI_PREFIX: ANSI terminal code notifying the start of an ANSI escape sequence :type ANSI_END: str :var ANSI_END: ANSI terminal code notifying the end of an ANSI escape sequence :type ANSI_RESET: str :var ANSI_RESET: ANSI terminal code resetting format defined by a previous ANSI escape sequence zrestructuredtext enN)PatternMatch)warn) normalize)AnyOptionalTupleListCallableDictUnion)linesep )callable_deprecated¡!ułlu⁄/ÆAE©z(c)«"æae®z(r)uœoeuŒOEØOøo»ßssu–-u’'ustringignorenonascii substitutereturnc|tdtd|rd}g}|ddD]} t|}|j|dj|S#t$r/t d|d}t |dk\r| t d |}Y[wxYw) aVreplace diacritical characters with their corresponding ascii characters Convert the unicode string to its long normalized form (unicode character will be transform into several characters) and keep the first one only. The normal form KD (NFKD) will apply the compatibility decomposition, i.e. replace all compatibility characters with their equivalents. :type substitute: str :param substitute: replacement character to use if decomposition fails :see: Another project about ASCII transliterations of Unicode text http://pypi.python.org/pypi/Unidecode NzDignorenonascii is deprecated, use substitute named parameter instead) stacklevelrNFKDrz*can't deal with non-ascii based characters) rDeprecationWarningMANUAL_UNICODE_MAPKeyError _uninormalizeord ValueErrorappendjoin)r)r*r+resletter replacements :/usr/lib/python3/dist-packages/logilab/common/textutils.py unormalizer>\s"! R  J C!*   ),V4K ;   773< )'7:K;4'%$%QRR(  )s A5B B stringc>|s|S|ddvr|dd}|ddvr|dd}|S)zremove optional quotes (simple or double) from the string :type string: str or unicode :param string: an optionally quoted string :rtype: str or unicode :return: the unquoted string (or the input string if it wasn't quoted) rz"'N)r?s r=unquoterDs?   ayE bzU Mz ? ? z\s+textline_lenindentrestc|rt}nt}g}tj|D]}|j ||||t |t j |S)anormalize a text to display it with a maximum line size and optionally arbitrary indentation. Line jumps are normalized but blank lines are kept. The indentation string may be used to insert a comment (#) or a quoting (>) mark for instance. :type text: str or unicode :param text: the input text to normalize :type line_len: int :param line_len: expected maximum line's length, default to 80 :type indent: str or unicode :param indent: optional string to use as indentation :rtype: str or unicode :return: the input text normalized to fit on lines with a maximized size inferior to `line_len`, and optionally prefixed by an indentation string )normalize_rest_paragraphnormalize_paragraph_BLANKLINES_RGXsplitr8rr9)rFrGrHrInormpresults r=normalize_textrQs]* (# F%%d+5 eD(F345 1 7 7 ??rEctjd|}|t|z }g}|r4t|j |\}}|j ||z|r4t j|S)anormalize a text to display it with a maximum line size and optionally arbitrary indentation. Line jumps are normalized. The indentation string may be used top insert a comment mark for instance. :type text: str or unicode :param text: the input text to normalize :type line_len: int :param line_len: expected maximum line's length, default to 80 :type indent: str or unicode :param indent: optional string to use as indentation :rtype: str or unicode :return: the input text normalized to fit on lines with a maximized size inferior to `line_len`, and optionally prefixed by an indentation string  )_NORM_SPACES_RGXsublen splittextstripr8rr9)rFrGrHlinesalines r=rLrLsh*   T *D#f+%H E  h7 t Ve^$  << rEcd}g}|t|z }|jD]}|tjd|j z}d}t||kDr>t ||\}}|j ||z|r|dz}d}nd}t||kDr>|s{|j ||j ztj|S)anormalize a ReST text to display it with a maximum line size and optionally arbitrary indentation. Line jumps are normalized. The indentation string may be used top insert a comment mark for instance. :type text: str or unicode :param text: the input text to normalize :type line_len: int :param line_len: expected maximum line's length, default to 80 :type indent: str or unicode :param indent: optional string to use as indentation :rtype: str or unicode :return: the input text normalized to fit on lines with a maximized size inferior to `line_len`, and optionally prefixed by an indentation string rrS) rV splitlinesrTrUrXrWr8rr9)rFrGrHtoreportrYlines r=rKrKs*H E#f+%H! 0*..sDJJLAA$i("&tX6ND( LL$ '#~$i("  LL$**,. / 0 << rEcdt||kr|dfStt|dz |}|dkDr||dk7r|dz}|dkDr ||dk7r|dk(rGtt||}t||kDr$||dk7r|dz }t||kDr ||dk7r|d|||dzdjfS)zsplit the given text on space according to the given max line size return a 2-uple: * a line <= line_len if possible * the rest of the text which has to be reported on another line rrArrSN)rVminrX)rFrGposs r=rWrWs 4yHRx c$i!mX &C 'd3i3& q 'd3i3& ax#d)X&$i#o$s)s"2 1HC$i#o$s)s"2 :tC!GI,,. ..rEsepc|j|Dcgc]#}|js|j%c}Scc}w)areturn a list of stripped string by splitting the string given as argument on `sep` (',' by default). Empty string are discarded. >>> splitstrip('a, b, c , 4,,') ['a', 'b', 'c', '4'] >>> splitstrip('a') ['a'] >>> :type string: str or unicode :param string: a csv line :type sep: str or unicode :param sep: field separator, default to the comma (',') :rtype: str or unicode :return: the unquoted string (or the input string if it wasn't quoted) )rNrX)r?rbwords r= splitstripres/&&,\\#%6 GT$**,DJJL GG Gs??z%get_csv is deprecated, use splitstripcd|vr!|jdjddStj|jtjS)zreturn the latest component of a string containing either an url of the form :// or a local file system path z://rrA)rstriprsplitosprNrb) url_or_paths r=split_url_or_pathrk+sH !!#&--c155 99[''0 11rEcbi}|s|S|jD]w}|j}|s|jdr(|jddDcgc]}|jc}\}}||vr ||j |s|||<y|Scc}w#t $r |||g||<YwxYw)amparse multilines text containing simple 'key=value' lines and return a dict of {'key': 'value'}. When the same key is encountered multiple time, value is turned into a list containing all values. >>> d = text_to_dict('''multiple=1 ... multiple= 2 ... single =3 ... ''') >>> d['single'] '3' >>> d['multiple'] ['1', '2'] #=rA)r\rX startswithrNr8AttributeError)rFr:r^wkeyvalues r= text_to_dictrt4s C  ! !zz| ,-1ZZQ-?@!'')@JCcz1HOOE*!C ! JA&1 #C%0CH1sB5BB.-B.z(\s|,)+z#-?(([0-9]+\.[0-9]*)|((0x?)?[0-9]+))z [a-zA-Z]+z (?Pz )(?Pz)?z^((z)(z))*(z)?$rAiii@l)bkbmbgbtbg-C6?<iiQ)mssr`hdunitsinterfinal blank_reg value_regc ||}tjd|}|rtj|st d|zg}|j |D]Z}|j } | d| jd} } || } |  | || jz} |j| \|t|S#t$r)t d| dt|jwxYw)amParse the string applying the units defined in units (e.g.: "1.5m",{'m',60} -> 80). :type string: str or unicode :param string: the string to parse :type units: dict (or any object with __getitem__ using basestring key) :param units: a dict mapping a unit string repr to its value :type inter: type :param inter: used to parse every intermediate value (need __sum__) :type blank_reg: regexp :param blank_reg: should match every blank char to ignore. :type value_reg: regexp with "value" and optional "unit" group :param value_reg: match a value and it's unit into the rzInvalid unit string: %r.rsunitz invalid unit z. valid units are ) _BLANK_RErU_VALIDATION_REmatchr7finditer groupdictgetlowerr4listkeysr8sum) r?rrrrrfstringvaluesrdiclitrrss r= apply_unitsrls4 }mmB'G ,,W53f<== F##G, ooL#''&/Tc    etzz|,,  e  V  e $PTUZU_U_UaPb!cdd es C2C3z | +| runderline_charct|j}|j}tjt|}|j td|}|dk(rd}g}n|d|g}|t tz }||z }d|z|||z zz}|jt|} | dk(r(||d}|j||j|nM|| t tzd}||| }|j||j||j|t j|jS)areturn a string with the match location underlined: >>> import re >>> print(pretty_match(re.search('mange', 'il mange du bacon'), 'il mange du bacon')) il mange du bacon ^^^^^ >>> :type match: _sre.SRE_match :param match: object returned by re.match, re.search or re.finditer :type string: str or unicode :param string: the string on which the regular expression has been applied to obtain the `match` object :type underline_char: str or unicode :param underline_char: character to use to underline the matched section, default to the carret '^' :rtype: str or unicode :return: the original string with an inserted line to underline the match location rrBNrS) startend _LINE_RGXrUrrfindrVfindr8r9rg) rr?rrrstart_line_posrPoffset underline end_line_poss r= pretty_matchrs(6 KKME ))+C ]]7F +F\\'1e4N.)*#g,& ^ #Ff ~u==I;;w,Lr( f i \CL023|4 f i  c <<  & & ((rEz[mz0134579)resetbolditalicrblinkinversestrike3031323334353637) rblackredgreenyellowbluemagentacyanwhitecolorstylec@g}|r*t|}|D]}|jt||rM|jr%|j ddg|j|n|jt ||rt dj|ztzSy)areturn ansi escape code corresponding to color and style :type color: str or None :param color: the color name (see `ANSI_COLORS` for available values) or the color number when 256 colors are available :type style: str or None :param style: style string (see `ANSI_COLORS` for available values). To get several style effects at the same time, use a coma as separator. :raise KeyError: if an unexistent color or style identifier is given :rtype: str :return: the built escape code 38r;r) rer8 ANSI_STYLESisdigitextend ANSI_COLORS ANSI_PREFIXr9ANSI_END)rr ansi_code style_attrseffects r=_get_ansi_coders$I  ' ! 2F   [0 1 2 ==?   dC[ )   U #   [/ 0SXXi008;; rEmsgcF|||St||}|r ||tS|S)aCcolorize message by wrapping it with ansi escape codes :type msg: str or unicode :param msg: the message string to colorize :type color: str or None :param color: the color identifier (see `ANSI_COLORS` for available values) :type style: str or None :param style: style string (see `ANSI_COLORS` for available values). To get several style effects at the same time, use a coma as separator. :raise KeyError: if an unexistent color or style identifier is given :rtype: str or unicode :return: the ansi escaped string )r ANSI_RESET)rrr escape_codes r= colorize_ansirs3* }  .K&Z88 JrErrr) separatorremoveaddc|D]}|dddvr|jt||d)|ddk(r|jt||dP|ddk(r|jt||dw|ddd k(r|jt||d|ddd k(r|jt||d|j|y) N)--- +++ rrr'r+rrr)writer)rYoutrr^s r=diff_colorize_ansir2s  8' ' IImD% *<= > !W^ IImD%/: ; !W^ IImD%,7 8 "1X  IImD% *<= > "1X  IImD% *<= > IIdO rE)NN)PrF)rr),)^)E__doc__ __docformat__sysreos.pathpathrirr ImportErrortype sre_compilecompilerwarningsr unicodedatarr5typingrr r r r r rosrlogilab.common.deprecationrr3strr>rDrMrTintboolrQrLrKrWreget_csvrkrt _BLANK_UREr __VALUE_URE __UNITS_URE _VALUE_REr BYTE_UNITS TIME_UNITSfloatrrrrrrrrrr DIFF_STYLEstdoutrrCrEr=rs$4&  :!2DDD; C c c D   E   C  D E d d C C C D c  c!*UY$ $"*3-$DLSM$$NCC$"**\*2::f%@@@#@$@[^@>cSsC<&3&#&C&QT&R/C/3/5c?/&HsHHtCyH, G E Fz R2@ BJJz " 4  BJJ [Q R [+{[\           *."" * * S>* 4% &*  *  *  * 5#:*Z BJJ} % 5)5)s5)C5)#5)t              (3-x}PSDs8C= Y\<"U7 C #&**J }:2>>))"a01G ''A.44R8 9E:Gs#H I, AI)(I),I76I7