o j4@sdZddlmZddlZgdZGdddZGdddeZGd d d eZGd d d eZGd ddeZ GdddeZ GdddeZ dddZ dddZ dS)a Parser for parsing a regular expression. Take a string representing a regular expression and return the root node of its parse tree. usage:: root_node = parse_regex('(hello|world)') Remarks: - The regex parser processes multiline, it ignores all whitespace and supports multiple named groups with the same name and #-style comments. Limitations: - Lookahead is not supported. ) annotationsN)RepeatVariableRegex Lookaheadtokenize_regex parse_regexc@s$eZdZdZd ddZd dd Zd S) NodezT Base class for all the grammar nodes. (You don't initialize this one.) other_nodereturn NodeSequencecC t||gSN)r selfr r/root/aizidognhua/tmp/workspace/projects/ec89d86c-575f-41c9-af57-ac45cbdbf775/venv/lib/python3.10/site-packages/prompt_toolkit/contrib/regular_languages/regex_parser.py__add__& z Node.__add__AnyNodecCr r)rrrrr__or__)rz Node.__or__Nr r r r r r r r)__name__ __module__ __qualname____doc__rrrrrrr s r c@.eZdZdZdddZdd d Zdd dZdS)rz Union operation (OR operation) between several grammars. You don't initialize this yourself, but it's a result of a "Grammar1 | Grammar2" operation. children list[Node]r NonecC ||_dSrrrrrrr__init__4 zAnyNode.__init__r r cCt|j|gSr)rrrrrrr7zAnyNode.__or__strcC|jjd|jdSN() __class__rrrrrr__repr__:zAnyNode.__repr__Nrrr r rr r()rrrrr$rr0rrrrr-s   rc@r)r z Concatenation operation of several grammars. You don't initialize this yourself, but it's a result of a "Grammar1 + Grammar2" operation. rrr r cCr!rr"r#rrrr$Dr%zNodeSequence.__init__r r cCr&r)r rrrrrrGr'zNodeSequence.__add__r(cCr)r*r-r/rrrr0Jr1zNodeSequence.__repr__Nr2rr3)rrrrr$rr0rrrrr >s   r c@s$eZdZdZd ddZd dd Zd S) rz Regular expression. regexr(r r cCst|||_dSr)recompiler4)rr4rrrr$Ss  zRegex.__init__cCs|jjd|jdS)Nz(/z/))r.rr4r/rrrr0Xr1zRegex.__repr__N)r4r(r r r3rrrrr$r0rrrrrNs rc@s&eZdZdZddd d Zdd d ZdS)rz Lookahead expression. F childnoder negativeboolr r cC||_||_dSr)r8r9)rr8r9rrrr$a zLookahead.__init__r(cCr)r*r.rr8r/rrrr0er1zLookahead.__repr__N)F)r8r r9r:r r r3r7rrrrr\s rc@s&eZdZdZddd d Zdd d Zd S)ra Mark a variable in the regular grammar. This will be translated into a named group. Each variable can have his own completer, validator, etc.. :param childnode: The grammar which is wrapped inside this variable. :param varname: String. r8r varnamer(r r cCr;r)r8r?)rr8r?rrrr$rr<zVariable.__init__cCs|jjd|jd|jdS)N (childnode=z , varname=r,)r.rr8r?r/rrrr0vszVariable.__repr__N)r>)r8r r?r(r r r3r7rrrrris rc@s(eZdZ   ddddZdddZdS)rrNTr8r min_repeatint max_repeat int | Nonegreedyr:r r cCs||_||_||_||_dSr)r8rArCrE)rr8rArCrErrrr${s zRepeat.__init__r(cCr))Nr@r,r=r/rrrr0r1zRepeat.__repr__)rNT) r8r rArBrCrDrEr:r r r3)rrrr$r0rrrrrzs   rinputr(r list[str]cCsjtdtj}g}|r3||}|r-|d|||d}}|s,||ntd|s |S)z Takes a string, representing a regular expression as input, and tokenizes it. :param input: string, representing a regular expression. :returns: List of tokens. a^( \(\?P\<[a-zA-Z0-9_-]+\> | # Start of named group. \(\?#[^)]*\) | # Comment \(\?= | # Start of lookahead assertion \(\?! | # Start of negative lookahead assertion \(\?<= | # If preceded by. \(\?< | # If not preceded by. \(?: | # Start of group. (non capturing.) \( | # Start of group. \(?[iLmsux] | # Flags. \(?P=[a-zA-Z]+\) | # Back reference to named group \) | # End of group. \{[^{}]*\} | # Repetition \*\? | \+\? | \?\?\ | # Non greedy repetition. \* | \+ | \? | # Repetition \#.*\n | # Comment \\. | # Character group. \[ ( [^\]\\] | \\.)* \] | [^(){}] | . )NzCould not tokenize input regex.)r5r6VERBOSEmatchendisspaceappend Exception)rFptokensmtokenrrrrs  "  r regex_tokenscsNdg|dddddd dfd d }td kr%td |S)zN Takes a list of tokens from the tokenizer, and returns a parse tree. r,Nlstrr r cSst|dkr |dSt|S)z7Turn list into sequence when it contains several items.r)lenr )rTrrrwraps zparse_regex..wrapcsggd#fdd }rو}|dr)t|ddd}|n|d vr<|d k}td|d d<n|d vrP|d k}tdd|dd<n|dvrqgkr`tdt|dk}tddd|dd<nf|dkr}gnZ|dvrnO|dkrtddn@|dkrtddn1|dkr|S|drn$|drt|d|d rt|d!|rnt |std")$Nr r cs0gkrStfddDS)Ncsg|]}|qSrr).0i)rWrr szGparse_regex.._parse..wrapped_result..)rLrr)or_listresultrWrrwrapped_results z3parse_regex.._parse..wrapped_resultz(?P<rS)r?)*z*?r_)rE)+z+?r`rU)rArE)?z??zNothing to repeat.rar)rArCrE|)r+z(?:z(?!T)r9z(?=Fr,#{z#-style repetition not yet supportedz(?z not supportedzExpecting ')' tokenr r ) pop startswithrrLrrMreprrrKr)r]tvariablerE_parserOrW)r[r\rrlsV        7zparse_regex.._parserzUnmatched parentheses.)rTrr r re)rVrM)rRr\rrkrrs D r)rFr(r rG)rRrGr r )r __future__rr5__all__r rr rrrrrrrrrrs     5