a W_#@sDdZddlZddlZddlmZedZedejejBej BZ edej ej BejBejBZ edejZGdd d eZGd d d eZd2d dZd3ddZd4ddZd5ddZddZddZddZddZddZiZdd Zd!d"Zd6d#d$Z d7d&d'Z!Gd(d)d)Z"d*d+Z#d,d-Z$d.d/Z%Gd0d1d1eZ&dS)8z pygments.util ~~~~~~~~~~~~~ Utility functions. :copyright: Copyright 2006-2020 by the Pygments team, see AUTHORS. :license: BSD, see LICENSE for details. N) TextIOWrapperz[/\\ ]z (<\?.*?\?>)?\s* ]*> z<(.+?)(\s.*?)?>.*?z\s*<\?xml[^>]*\?>c@seZdZdZdS) ClassNotFoundzCRaised if one of the lookup functions didn't find a matching class.N)__name__ __module__ __qualname____doc__rr//usr/lib/python3/dist-packages/pygments/util.pyr!src@s eZdZdS) OptionErrorN)rrrrrrr r %sr FcCs@|||}|r|}||vrInvalid value %r for option %s; you must give an integer value)r r TypeErrorr ValueErrorrrrr get_int_optGs     r cCsH|||}t|tr|St|ttfr4t|Std||fdS)Nz9Invalid type %r for option %s; you must give a list value)r rrsplitlisttupler )rrrvalrrr get_list_optUs  r%cCsP|js dSg}|jD]$}|r<|d|qqBqd|S)N )rstrip splitlinesappendr lstrip)objreslinerrr docstring_headlineasr/csfdd}j|_t|S)zAReturn a static text analyser function that returns float values.c s^z |}Wnty YdS0|s*dSztdtdt|WSttfyXYdS0dS)Ngg?) Exceptionminmaxfloatrr)textrvfrr text_analyseos  z%make_analysator..text_analyse)r staticmethod)r7r8rr6r make_analysatorms r:cCs|d}|dkr$|d|}n|}|drz(ddt|ddDd}WntyrYd S0td |tj }| |durd Sd S) aCheck if the given regular expression matches the last part of the shebang if one exists. >>> from pygments.util import shebang_matches >>> shebang_matches('#!/usr/bin/env python', r'python(2\.\d)?') True >>> shebang_matches('#!/usr/bin/python2.4', r'python(2\.\d)?') True >>> shebang_matches('#!/usr/bin/python-ruby', r'python(2\.\d)?') False >>> shebang_matches('#!/usr/bin/python/ruby', r'python(2\.\d)?') False >>> shebang_matches('#!/usr/bin/startsomethingwith python', ... r'python(2\.\d)?') True It also checks for common windows executable file extensions:: >>> shebang_matches('#!C:\\Python2.4\\Python.exe', r'python(2\.\d)?') True Parameters (``'-f'`` or ``'--foo'`` are ignored so ``'perl'`` does the same as ``'perl -e'``) Note that this method automatically searches the whole string (eg: the regular expression is wrapped in ``'^$'``)  rNz#!cSsg|]}|r|ds|qS)-) startswith).0xrrr sz#shebang_matches..Fz^%s(\.(exe|cmd|bat|bin))?$T) findr r= split_path_rer!r( IndexErrorrecompile IGNORECASEsearch)r4regexindexZ first_linefoundrrr shebang_matches~s   rMcCs<t|}|durdS|d}t|tj|duS)zCheck if the doctype matches a regular expression (if present). Note that this method only checks the first part of a DOCTYPE. eg: 'html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"' NFrA)doctype_lookup_rerIgrouprFrGImatchr()r4rJmZdoctyperrr doctype_matchess   rScCs t|dS)z3Check if the file looks like it has a html doctype.Zhtml)rS)r4rrr html_doctype_matchessrTcCsrt|rdSt|}z t|WStylt|}|durFYdSt|dddu}|t|<|YS0dS)z2Check if a doctype exists or if we have some tags.TNi) xml_decl_rerQhash_looks_like_xml_cacheKeyErrorrNrItag_re)r4keyrRr5rrr looks_like_xmls    r[cCsd|d?d|d@fS)zoGiven a unicode character code with length greater than 16 bits, return the two 16 bit surrogate pair. i iir)crrr surrogatepairsr^c Csg}d|d}d|dd}|||d|rT|D]}|||dq:n8|D]2}t|d}|||dd|d dqX||d d |S) z)Formats a sequence of strings for output.r'z = (,"NrB)r;)r*reprr ) Zvar_nameseqrawZ indent_levellinesZ base_indentZ inner_indentirrrr format_liness  $rkrcCs>g}t}|D]*}||vs||vr$q||||q|S)za Returns a list with duplicates removed from the iterable `it`. Order is preserved. )setr*add)itZ already_seenZlstseenrirrr duplicates_removeds  rpc@seZdZdZddZdS)FuturezGeneric class to defer some work. Handled specially in RegexLexerMeta, to support regex string construction at first use. cCstdSN)NotImplementedErrorselfrrr r sz Future.getN)rrrrr rrrr rqsrqc Csz|d}|dfWStyzz&ddl}|}|}||fWYSttfyt|d}|dfYYS0Yn0dS)zDecode *text* with guessed encoding. First try UTF-8; this should fail for non-UTF-8 encodings. Then try the preferred locale encoding. Fall back to latin-1, which always works. zutf-8rNlatin1)decodeUnicodeDecodeErrorlocalegetpreferredencoding LookupError)r4ryZ prefencodingrrr guess_decodes    r|cCsBt|ddr:z||j}Wnty.Yn 0||jfSt|S)zDecode *text* coming from terminal *term*. First try the terminal encoding, if given. Then try UTF-8. Then try the preferred locale encoding. Fall back to latin-1, which always works. encodingN)getattrrwr}rxr|)r4termrrr guess_decode_from_terminals   rcCs"t|ddr|jSddl}|S)z7Return our best guess of encoding for the given *term*.r}Nr)r~r}ryrz)rryrrr terminal_encoding,s rc@seZdZddZdS)UnclosingTextIOWrappercCs |dSrr)flushrtrrr close6szUnclosingTextIOWrapper.closeN)rrrrrrrr r4sr)NF)N)N)N)Fr)r)'rrFsysiorrGrDDOTALL MULTILINEVERBOSErNUNICODErHrYrPrUrrr0r rrr r%r/r:rMrSrTrWr[r^rkrprqr|rrrrrrr s@         -