You cannot select more than 25 topics
			Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
		
		
		
		
		
			
		
			
				
	
	
		
			406 lines
		
	
	
		
			14 KiB
		
	
	
	
		
			JavaScript
		
	
			
		
		
	
	
			406 lines
		
	
	
		
			14 KiB
		
	
	
	
		
			JavaScript
		
	
"use strict";
 | 
						|
(self["webpackChunk_JUPYTERLAB_CORE_OUTPUT"] = self["webpackChunk_JUPYTERLAB_CORE_OUTPUT"] || []).push([[2871],{
 | 
						|
 | 
						|
/***/ 72871:
 | 
						|
/***/ ((__unused_webpack_module, __webpack_exports__, __webpack_require__) => {
 | 
						|
 | 
						|
__webpack_require__.r(__webpack_exports__);
 | 
						|
/* harmony export */ __webpack_require__.d(__webpack_exports__, {
 | 
						|
/* harmony export */   cython: () => (/* binding */ cython),
 | 
						|
/* harmony export */   mkPython: () => (/* binding */ mkPython),
 | 
						|
/* harmony export */   python: () => (/* binding */ python)
 | 
						|
/* harmony export */ });
 | 
						|
function wordRegexp(words) {
 | 
						|
  return new RegExp("^((" + words.join(")|(") + "))\\b");
 | 
						|
}
 | 
						|
 | 
						|
var wordOperators = wordRegexp(["and", "or", "not", "is"]);
 | 
						|
var commonKeywords = ["as", "assert", "break", "class", "continue",
 | 
						|
                      "def", "del", "elif", "else", "except", "finally",
 | 
						|
                      "for", "from", "global", "if", "import",
 | 
						|
                      "lambda", "pass", "raise", "return",
 | 
						|
                      "try", "while", "with", "yield", "in", "False", "True"];
 | 
						|
var commonBuiltins = ["abs", "all", "any", "bin", "bool", "bytearray", "callable", "chr",
 | 
						|
                      "classmethod", "compile", "complex", "delattr", "dict", "dir", "divmod",
 | 
						|
                      "enumerate", "eval", "filter", "float", "format", "frozenset",
 | 
						|
                      "getattr", "globals", "hasattr", "hash", "help", "hex", "id",
 | 
						|
                      "input", "int", "isinstance", "issubclass", "iter", "len",
 | 
						|
                      "list", "locals", "map", "max", "memoryview", "min", "next",
 | 
						|
                      "object", "oct", "open", "ord", "pow", "property", "range",
 | 
						|
                      "repr", "reversed", "round", "set", "setattr", "slice",
 | 
						|
                      "sorted", "staticmethod", "str", "sum", "super", "tuple",
 | 
						|
                      "type", "vars", "zip", "__import__", "NotImplemented",
 | 
						|
                      "Ellipsis", "__debug__"];
 | 
						|
 | 
						|
function top(state) {
 | 
						|
  return state.scopes[state.scopes.length - 1];
 | 
						|
}
 | 
						|
 | 
						|
function mkPython(parserConf) {
 | 
						|
  var ERRORCLASS = "error";
 | 
						|
 | 
						|
  var delimiters = parserConf.delimiters || parserConf.singleDelimiters || /^[\(\)\[\]\{\}@,:`=;\.\\]/;
 | 
						|
  //               (Backwards-compatibility with old, cumbersome config system)
 | 
						|
  var operators = [parserConf.singleOperators, parserConf.doubleOperators, parserConf.doubleDelimiters, parserConf.tripleDelimiters,
 | 
						|
                   parserConf.operators || /^([-+*/%\/&|^]=?|[<>=]+|\/\/=?|\*\*=?|!=|[~!@]|\.\.\.)/]
 | 
						|
  for (var i = 0; i < operators.length; i++) if (!operators[i]) operators.splice(i--, 1)
 | 
						|
 | 
						|
  var hangingIndent = parserConf.hangingIndent;
 | 
						|
 | 
						|
  var myKeywords = commonKeywords, myBuiltins = commonBuiltins;
 | 
						|
  if (parserConf.extra_keywords != undefined)
 | 
						|
    myKeywords = myKeywords.concat(parserConf.extra_keywords);
 | 
						|
 | 
						|
  if (parserConf.extra_builtins != undefined)
 | 
						|
    myBuiltins = myBuiltins.concat(parserConf.extra_builtins);
 | 
						|
 | 
						|
  var py3 = !(parserConf.version && Number(parserConf.version) < 3)
 | 
						|
  if (py3) {
 | 
						|
    // since http://legacy.python.org/dev/peps/pep-0465/ @ is also an operator
 | 
						|
    var identifiers = parserConf.identifiers|| /^[_A-Za-z\u00A1-\uFFFF][_A-Za-z0-9\u00A1-\uFFFF]*/;
 | 
						|
    myKeywords = myKeywords.concat(["nonlocal", "None", "aiter", "anext", "async", "await", "breakpoint", "match", "case"]);
 | 
						|
    myBuiltins = myBuiltins.concat(["ascii", "bytes", "exec", "print"]);
 | 
						|
    var stringPrefixes = new RegExp("^(([rbuf]|(br)|(rb)|(fr)|(rf))?('{3}|\"{3}|['\"]))", "i");
 | 
						|
  } else {
 | 
						|
    var identifiers = parserConf.identifiers|| /^[_A-Za-z][_A-Za-z0-9]*/;
 | 
						|
    myKeywords = myKeywords.concat(["exec", "print"]);
 | 
						|
    myBuiltins = myBuiltins.concat(["apply", "basestring", "buffer", "cmp", "coerce", "execfile",
 | 
						|
                                    "file", "intern", "long", "raw_input", "reduce", "reload",
 | 
						|
                                    "unichr", "unicode", "xrange", "None"]);
 | 
						|
    var stringPrefixes = new RegExp("^(([rubf]|(ur)|(br))?('{3}|\"{3}|['\"]))", "i");
 | 
						|
  }
 | 
						|
  var keywords = wordRegexp(myKeywords);
 | 
						|
  var builtins = wordRegexp(myBuiltins);
 | 
						|
 | 
						|
  // tokenizers
 | 
						|
  function tokenBase(stream, state) {
 | 
						|
    var sol = stream.sol() && state.lastToken != "\\"
 | 
						|
    if (sol) state.indent = stream.indentation()
 | 
						|
    // Handle scope changes
 | 
						|
    if (sol && top(state).type == "py") {
 | 
						|
      var scopeOffset = top(state).offset;
 | 
						|
      if (stream.eatSpace()) {
 | 
						|
        var lineOffset = stream.indentation();
 | 
						|
        if (lineOffset > scopeOffset)
 | 
						|
          pushPyScope(stream, state);
 | 
						|
        else if (lineOffset < scopeOffset && dedent(stream, state) && stream.peek() != "#")
 | 
						|
          state.errorToken = true;
 | 
						|
        return null;
 | 
						|
      } else {
 | 
						|
        var style = tokenBaseInner(stream, state);
 | 
						|
        if (scopeOffset > 0 && dedent(stream, state))
 | 
						|
          style += " " + ERRORCLASS;
 | 
						|
        return style;
 | 
						|
      }
 | 
						|
    }
 | 
						|
    return tokenBaseInner(stream, state);
 | 
						|
  }
 | 
						|
 | 
						|
  function tokenBaseInner(stream, state, inFormat) {
 | 
						|
    if (stream.eatSpace()) return null;
 | 
						|
 | 
						|
    // Handle Comments
 | 
						|
    if (!inFormat && stream.match(/^#.*/)) return "comment";
 | 
						|
 | 
						|
    // Handle Number Literals
 | 
						|
    if (stream.match(/^[0-9\.]/, false)) {
 | 
						|
      var floatLiteral = false;
 | 
						|
      // Floats
 | 
						|
      if (stream.match(/^[\d_]*\.\d+(e[\+\-]?\d+)?/i)) { floatLiteral = true; }
 | 
						|
      if (stream.match(/^[\d_]+\.\d*/)) { floatLiteral = true; }
 | 
						|
      if (stream.match(/^\.\d+/)) { floatLiteral = true; }
 | 
						|
      if (floatLiteral) {
 | 
						|
        // Float literals may be "imaginary"
 | 
						|
        stream.eat(/J/i);
 | 
						|
        return "number";
 | 
						|
      }
 | 
						|
      // Integers
 | 
						|
      var intLiteral = false;
 | 
						|
      // Hex
 | 
						|
      if (stream.match(/^0x[0-9a-f_]+/i)) intLiteral = true;
 | 
						|
      // Binary
 | 
						|
      if (stream.match(/^0b[01_]+/i)) intLiteral = true;
 | 
						|
      // Octal
 | 
						|
      if (stream.match(/^0o[0-7_]+/i)) intLiteral = true;
 | 
						|
      // Decimal
 | 
						|
      if (stream.match(/^[1-9][\d_]*(e[\+\-]?[\d_]+)?/)) {
 | 
						|
        // Decimal literals may be "imaginary"
 | 
						|
        stream.eat(/J/i);
 | 
						|
        // TODO - Can you have imaginary longs?
 | 
						|
        intLiteral = true;
 | 
						|
      }
 | 
						|
      // Zero by itself with no other piece of number.
 | 
						|
      if (stream.match(/^0(?![\dx])/i)) intLiteral = true;
 | 
						|
      if (intLiteral) {
 | 
						|
        // Integer literals may be "long"
 | 
						|
        stream.eat(/L/i);
 | 
						|
        return "number";
 | 
						|
      }
 | 
						|
    }
 | 
						|
 | 
						|
    // Handle Strings
 | 
						|
    if (stream.match(stringPrefixes)) {
 | 
						|
      var isFmtString = stream.current().toLowerCase().indexOf('f') !== -1;
 | 
						|
      if (!isFmtString) {
 | 
						|
        state.tokenize = tokenStringFactory(stream.current(), state.tokenize);
 | 
						|
        return state.tokenize(stream, state);
 | 
						|
      } else {
 | 
						|
        state.tokenize = formatStringFactory(stream.current(), state.tokenize);
 | 
						|
        return state.tokenize(stream, state);
 | 
						|
      }
 | 
						|
    }
 | 
						|
 | 
						|
    for (var i = 0; i < operators.length; i++)
 | 
						|
      if (stream.match(operators[i])) return "operator"
 | 
						|
 | 
						|
    if (stream.match(delimiters)) return "punctuation";
 | 
						|
 | 
						|
    if (state.lastToken == "." && stream.match(identifiers))
 | 
						|
      return "property";
 | 
						|
 | 
						|
    if (stream.match(keywords) || stream.match(wordOperators))
 | 
						|
      return "keyword";
 | 
						|
 | 
						|
    if (stream.match(builtins))
 | 
						|
      return "builtin";
 | 
						|
 | 
						|
    if (stream.match(/^(self|cls)\b/))
 | 
						|
      return "self";
 | 
						|
 | 
						|
    if (stream.match(identifiers)) {
 | 
						|
      if (state.lastToken == "def" || state.lastToken == "class")
 | 
						|
        return "def";
 | 
						|
      return "variable";
 | 
						|
    }
 | 
						|
 | 
						|
    // Handle non-detected items
 | 
						|
    stream.next();
 | 
						|
    return inFormat ? null :ERRORCLASS;
 | 
						|
  }
 | 
						|
 | 
						|
  function formatStringFactory(delimiter, tokenOuter) {
 | 
						|
    while ("rubf".indexOf(delimiter.charAt(0).toLowerCase()) >= 0)
 | 
						|
      delimiter = delimiter.substr(1);
 | 
						|
 | 
						|
    var singleline = delimiter.length == 1;
 | 
						|
    var OUTCLASS = "string";
 | 
						|
 | 
						|
    function tokenNestedExpr(depth) {
 | 
						|
      return function(stream, state) {
 | 
						|
        var inner = tokenBaseInner(stream, state, true)
 | 
						|
        if (inner == "punctuation") {
 | 
						|
          if (stream.current() == "{") {
 | 
						|
            state.tokenize = tokenNestedExpr(depth + 1)
 | 
						|
          } else if (stream.current() == "}") {
 | 
						|
            if (depth > 1) state.tokenize = tokenNestedExpr(depth - 1)
 | 
						|
            else state.tokenize = tokenString
 | 
						|
          }
 | 
						|
        }
 | 
						|
        return inner
 | 
						|
      }
 | 
						|
    }
 | 
						|
 | 
						|
    function tokenString(stream, state) {
 | 
						|
      while (!stream.eol()) {
 | 
						|
        stream.eatWhile(/[^'"\{\}\\]/);
 | 
						|
        if (stream.eat("\\")) {
 | 
						|
          stream.next();
 | 
						|
          if (singleline && stream.eol())
 | 
						|
            return OUTCLASS;
 | 
						|
        } else if (stream.match(delimiter)) {
 | 
						|
          state.tokenize = tokenOuter;
 | 
						|
          return OUTCLASS;
 | 
						|
        } else if (stream.match('{{')) {
 | 
						|
          // ignore {{ in f-str
 | 
						|
          return OUTCLASS;
 | 
						|
        } else if (stream.match('{', false)) {
 | 
						|
          // switch to nested mode
 | 
						|
          state.tokenize = tokenNestedExpr(0)
 | 
						|
          if (stream.current()) return OUTCLASS;
 | 
						|
          else return state.tokenize(stream, state)
 | 
						|
        } else if (stream.match('}}')) {
 | 
						|
          return OUTCLASS;
 | 
						|
        } else if (stream.match('}')) {
 | 
						|
          // single } in f-string is an error
 | 
						|
          return ERRORCLASS;
 | 
						|
        } else {
 | 
						|
          stream.eat(/['"]/);
 | 
						|
        }
 | 
						|
      }
 | 
						|
      if (singleline) {
 | 
						|
        if (parserConf.singleLineStringErrors)
 | 
						|
          return ERRORCLASS;
 | 
						|
        else
 | 
						|
          state.tokenize = tokenOuter;
 | 
						|
      }
 | 
						|
      return OUTCLASS;
 | 
						|
    }
 | 
						|
    tokenString.isString = true;
 | 
						|
    return tokenString;
 | 
						|
  }
 | 
						|
 | 
						|
  function tokenStringFactory(delimiter, tokenOuter) {
 | 
						|
    while ("rubf".indexOf(delimiter.charAt(0).toLowerCase()) >= 0)
 | 
						|
      delimiter = delimiter.substr(1);
 | 
						|
 | 
						|
    var singleline = delimiter.length == 1;
 | 
						|
    var OUTCLASS = "string";
 | 
						|
 | 
						|
    function tokenString(stream, state) {
 | 
						|
      while (!stream.eol()) {
 | 
						|
        stream.eatWhile(/[^'"\\]/);
 | 
						|
        if (stream.eat("\\")) {
 | 
						|
          stream.next();
 | 
						|
          if (singleline && stream.eol())
 | 
						|
            return OUTCLASS;
 | 
						|
        } else if (stream.match(delimiter)) {
 | 
						|
          state.tokenize = tokenOuter;
 | 
						|
          return OUTCLASS;
 | 
						|
        } else {
 | 
						|
          stream.eat(/['"]/);
 | 
						|
        }
 | 
						|
      }
 | 
						|
      if (singleline) {
 | 
						|
        if (parserConf.singleLineStringErrors)
 | 
						|
          return ERRORCLASS;
 | 
						|
        else
 | 
						|
          state.tokenize = tokenOuter;
 | 
						|
      }
 | 
						|
      return OUTCLASS;
 | 
						|
    }
 | 
						|
    tokenString.isString = true;
 | 
						|
    return tokenString;
 | 
						|
  }
 | 
						|
 | 
						|
  function pushPyScope(stream, state) {
 | 
						|
    while (top(state).type != "py") state.scopes.pop()
 | 
						|
    state.scopes.push({offset: top(state).offset + stream.indentUnit,
 | 
						|
                       type: "py",
 | 
						|
                       align: null})
 | 
						|
  }
 | 
						|
 | 
						|
  function pushBracketScope(stream, state, type) {
 | 
						|
    var align = stream.match(/^[\s\[\{\(]*(?:#|$)/, false) ? null : stream.column() + 1
 | 
						|
    state.scopes.push({offset: state.indent + (hangingIndent || stream.indentUnit),
 | 
						|
                       type: type,
 | 
						|
                       align: align})
 | 
						|
  }
 | 
						|
 | 
						|
  function dedent(stream, state) {
 | 
						|
    var indented = stream.indentation();
 | 
						|
    while (state.scopes.length > 1 && top(state).offset > indented) {
 | 
						|
      if (top(state).type != "py") return true;
 | 
						|
      state.scopes.pop();
 | 
						|
    }
 | 
						|
    return top(state).offset != indented;
 | 
						|
  }
 | 
						|
 | 
						|
  function tokenLexer(stream, state) {
 | 
						|
    if (stream.sol()) {
 | 
						|
      state.beginningOfLine = true;
 | 
						|
      state.dedent = false;
 | 
						|
    }
 | 
						|
 | 
						|
    var style = state.tokenize(stream, state);
 | 
						|
    var current = stream.current();
 | 
						|
 | 
						|
    // Handle decorators
 | 
						|
    if (state.beginningOfLine && current == "@")
 | 
						|
      return stream.match(identifiers, false) ? "meta" : py3 ? "operator" : ERRORCLASS;
 | 
						|
 | 
						|
    if (/\S/.test(current)) state.beginningOfLine = false;
 | 
						|
 | 
						|
    if ((style == "variable" || style == "builtin")
 | 
						|
        && state.lastToken == "meta")
 | 
						|
      style = "meta";
 | 
						|
 | 
						|
    // Handle scope changes.
 | 
						|
    if (current == "pass" || current == "return")
 | 
						|
      state.dedent = true;
 | 
						|
 | 
						|
    if (current == "lambda") state.lambda = true;
 | 
						|
    if (current == ":" && !state.lambda && top(state).type == "py" && stream.match(/^\s*(?:#|$)/, false))
 | 
						|
      pushPyScope(stream, state);
 | 
						|
 | 
						|
    if (current.length == 1 && !/string|comment/.test(style)) {
 | 
						|
      var delimiter_index = "[({".indexOf(current);
 | 
						|
      if (delimiter_index != -1)
 | 
						|
        pushBracketScope(stream, state, "])}".slice(delimiter_index, delimiter_index+1));
 | 
						|
 | 
						|
      delimiter_index = "])}".indexOf(current);
 | 
						|
      if (delimiter_index != -1) {
 | 
						|
        if (top(state).type == current) state.indent = state.scopes.pop().offset - (hangingIndent || stream.indentUnit)
 | 
						|
        else return ERRORCLASS;
 | 
						|
      }
 | 
						|
    }
 | 
						|
    if (state.dedent && stream.eol() && top(state).type == "py" && state.scopes.length > 1)
 | 
						|
      state.scopes.pop();
 | 
						|
 | 
						|
    return style;
 | 
						|
  }
 | 
						|
 | 
						|
  return {
 | 
						|
    name: "python",
 | 
						|
 | 
						|
    startState: function() {
 | 
						|
      return {
 | 
						|
        tokenize: tokenBase,
 | 
						|
        scopes: [{offset: 0, type: "py", align: null}],
 | 
						|
        indent: 0,
 | 
						|
        lastToken: null,
 | 
						|
        lambda: false,
 | 
						|
        dedent: 0
 | 
						|
      };
 | 
						|
    },
 | 
						|
 | 
						|
    token: function(stream, state) {
 | 
						|
      var addErr = state.errorToken;
 | 
						|
      if (addErr) state.errorToken = false;
 | 
						|
      var style = tokenLexer(stream, state);
 | 
						|
 | 
						|
      if (style && style != "comment")
 | 
						|
        state.lastToken = (style == "keyword" || style == "punctuation") ? stream.current() : style;
 | 
						|
      if (style == "punctuation") style = null;
 | 
						|
 | 
						|
      if (stream.eol() && state.lambda)
 | 
						|
        state.lambda = false;
 | 
						|
      return addErr ? ERRORCLASS : style;
 | 
						|
    },
 | 
						|
 | 
						|
    indent: function(state, textAfter, cx) {
 | 
						|
      if (state.tokenize != tokenBase)
 | 
						|
        return state.tokenize.isString ? null : 0;
 | 
						|
 | 
						|
      var scope = top(state)
 | 
						|
      var closing = scope.type == textAfter.charAt(0) ||
 | 
						|
          scope.type == "py" && !state.dedent && /^(else:|elif |except |finally:)/.test(textAfter)
 | 
						|
      if (scope.align != null)
 | 
						|
        return scope.align - (closing ? 1 : 0)
 | 
						|
      else
 | 
						|
        return scope.offset - (closing ? hangingIndent || cx.unit : 0)
 | 
						|
    },
 | 
						|
 | 
						|
    languageData: {
 | 
						|
      autocomplete: commonKeywords.concat(commonBuiltins).concat(["exec", "print"]),
 | 
						|
      indentOnInput: /^\s*([\}\]\)]|else:|elif |except |finally:)$/,
 | 
						|
      commentTokens: {line: "#"},
 | 
						|
      closeBrackets: {brackets: ["(", "[", "{", "'", '"', "'''", '"""']}
 | 
						|
    }
 | 
						|
  };
 | 
						|
};
 | 
						|
 | 
						|
var words = function(str) { return str.split(" "); };
 | 
						|
 | 
						|
const python = mkPython({})
 | 
						|
 | 
						|
const cython = mkPython({
 | 
						|
  extra_keywords: words("by cdef cimport cpdef ctypedef enum except "+
 | 
						|
                        "extern gil include nogil property public "+
 | 
						|
                        "readonly struct union DEF IF ELIF ELSE")
 | 
						|
})
 | 
						|
 | 
						|
 | 
						|
/***/ })
 | 
						|
 | 
						|
}]);
 | 
						|
//# sourceMappingURL=2871.46ec88c6997ef947f39f.js.map?v=46ec88c6997ef947f39f
 |