X-Git-Url: https://git.kianting.info/?a=blobdiff_plain;f=src%2Findex.js;h=15e18628b9bd679b10e00ed1215069f029b1dfe2;hb=d4dfd2e99f564ca880d40172687986ea3ce757f0;hp=e18ebd5a1c4ff82a83542e067c0c39c7c233d41c;hpb=6f2e788329da7702ea96dc28ae04499917ec8152;p=clo diff --git a/src/index.js b/src/index.js index e18ebd5..15e1862 100644 --- a/src/index.js +++ b/src/index.js @@ -23,97 +23,184 @@ var __importStar = (this && this.__importStar) || function (mod) { return result; }; Object.defineProperty(exports, "__esModule", { value: true }); -exports.m1TType = void 0; +exports.tkTreeToSExp = void 0; var fs = require('fs'); -const node_process_1 = require("node:process"); -const tk = __importStar(require("./tokenize.js")); const util = __importStar(require("util")); +const p = __importStar(require("typescript-parsec")); /** - * @description - * match one token type. * - * it returns a function which test if the type of first token of the `remained` part of - * the argument of the function is `typ` , if it's true, update the `TokenMatcheePair` wrapped - * in `Some`. Otherwise, it returns `None`. - * * @param typ : the type to be test. - * @returns the updated `TokenMatcheePair` wrapped in `Some(x)` or `None`. + * # REPRESENTATION */ -function m1TType(typ) { - return (m) => { - if (m.remained.length == 0) { - return { _tag: "None" }; - } - /** - * token to be matched - * */ - const ttbm = m.remained[0]; - if (ttbm.type == typ) { - m.matched.push(ttbm); - return { - _tag: "Some", value: { - matched: m.matched, - remained: m.remained.slice(1) - } - }; +/** + * convert a `tkTree` AST to S-expr string + * @param t the `tkTree` + * @returns S-expr String + */ +function tkTreeToSExp(t) { + var str = ""; + if (Array.isArray(t)) { + let strArray = t.map((x) => tkTreeToSExp(x)); + str = "(" + strArray.join(" ") + ")"; + } + else { + if (t === undefined) { + str = "%undefined"; } else { - return { _tag: "None" }; + str = t; } - }; + } + return str; } -exports.m1TType = m1TType; -; -let toSome = tk.toSome; -let thenDo = tk.thenDo; -let orDo = tk.orDo; -node_process_1.argv.forEach((val, index) => { - console.log(`${index}=${val}`); -}); -let commandInput = node_process_1.argv[2]; -let commandInputTokenized = tk.tokenize(commandInput); -console.log(commandInputTokenized); +exports.tkTreeToSExp = tkTreeToSExp; +/**inspect the inner of the representation. */ +let repr = (x) => { return util.inspect(x, { depth: null }); }; +var TokenKind; +(function (TokenKind) { + TokenKind[TokenKind["Seperator"] = 0] = "Seperator"; + TokenKind[TokenKind["Semicolon"] = 1] = "Semicolon"; + TokenKind[TokenKind["Number"] = 2] = "Number"; + TokenKind[TokenKind["Op"] = 3] = "Op"; + TokenKind[TokenKind["ExprMark"] = 4] = "ExprMark"; + TokenKind[TokenKind["ExcapeAt"] = 5] = "ExcapeAt"; + TokenKind[TokenKind["Paren"] = 6] = "Paren"; + TokenKind[TokenKind["SpaceNL"] = 7] = "SpaceNL"; + TokenKind[TokenKind["Id"] = 8] = "Id"; + TokenKind[TokenKind["Str"] = 9] = "Str"; + TokenKind[TokenKind["Comment"] = 10] = "Comment"; +})(TokenKind || (TokenKind = {})); /** - * matchee pair of commandInputTokenized + * Parsing + */ +const lexer = p.buildLexer([ + [true, /^\d+(\.\d+)?/g, TokenKind.Number], + [true, /^\\\@/g, TokenKind.ExcapeAt], + [true, /^\/\*([^/]|\/[^*])*\*\//g, TokenKind.Comment], + [true, /^\;/g, TokenKind.Semicolon], + [true, /^[-][-][-]/g, TokenKind.Seperator], + [true, /^[\+\-\*\/\&\|\!\^\<\>\~\=\?]+/g, TokenKind.Op], + [true, /^\@/g, TokenKind.ExprMark], + [true, /^[()\[\]{}]/g, TokenKind.Paren], + [true, /^["]([\"]|[\\].)*["]/g, TokenKind.Str], + [true, /^[']([\']|[\\].)*[']/g, TokenKind.Str], + [true, /^[()\[\]{}]/g, TokenKind.Paren], + [true, /^[^\/\\\@\s\n\t\r;]+/g, TokenKind.Id], + [true, /^(\s|\n|\r|\t)+/g, TokenKind.SpaceNL], +]); +/** + * + * # TEST */ -let commandTPair = { matched: [], - remained: commandInputTokenized }; -let tInt = m1TType(tk.TokenType.INT); -let tFlo = m1TType(tk.TokenType.FLO); -let tStr = m1TType(tk.TokenType.STR); -function tBool(x) { - let text = x.remained[0].text; - if (text == "true" || text == "false") { - return thenDo(toSome(x), m1TType(tk.TokenType.ID)); +const inputTxt = `import a as b; /*bacourt*/ +/* ba choir +ipsum lorem*/ + +import you as john; +--- + +臺中市\\\@ + +公園 +@1+2==3; + +console.log("122");@ + +山頂 +`; +const PROG = p.rule(); +const SEGMENT = p.rule(); +const IMPORT = p.rule(); +const IMPORTS = p.rule(); +const SEMICOLON = p.rule(); +const EXCAPE_AT = p.rule(); +const NOT_AT_TEXT = p.rule(); +const CONTENT = p.rule(); +let doubleMinus = { type: 'Punctuator', value: '--' }; +let doubleMinus2 = p.str('--'); +const TERM = p.rule(); +function applySegment(input) { + let unpackedInnerExprs = input[1].map((x) => { return x.text; }); + return ["%exprs", unpackedInnerExprs]; +} +function applySemiColon(value) { + return value.text; +} +function applyParts(first, second) { + return ["%clo", first, second[1]]; +} +function applyComment(value) { + return [value.text]; +} +function applyImport(input) { + let importTail = input[1].map(x => x.text); + return ["import"].concat(importTail); +} +; +/* +function applyImportComment(input: [Token,Token[], + tkTree, Token]) : tkTree{ + let importTail = input[1].map(x=>x.text); + let comment = [input[3].text]; + return ["import"].concat(importTail).concat(comment); +};*/ +function applyImports(input) { + let resultBody = [input[0]].concat(input[1]); + let resultWrapper = ["%import", resultBody]; + return resultWrapper; +} +; +function applyNotAtText(value) { + if (value.text == "\\\@") { + return '@'; } else { - return { _tag: "None" }; + return value.text; } } +; +function applyText(input) { + return ["%text", input]; +} +; +function applyContent(input) { + return ["%content", input]; +} +; +function applySpaceNL(value) { + return value.text; +} /** - * define the right hand side of a grammar - * eg. `LHS ::= a + b` - * @param process the right hand side processing : eg. `a + b` in `LHS` - * @param arrange define the order (0 starting) of the elements of the result. - * ast. : eg. `a + c` is `1 0 2` `(+ a c)` - * @returns the processed ast. + * IMPORTEE: Number, Op, Paren, Id, Str, Comment, */ -function gramRHS(process, arrange) { - return (m) => { - let result = process(m); - console.log(`result ${result}`); - if (result._tag == "None") { - return result; - } - else { - let matched = result.value.matched; - let return_array = Array(arrange.length); - arrange.forEach((val, index) => { - return_array[arrange[index]] = matched[index]; - }); - return return_array; - } - }; -} -var constParser = gramRHS((x) => { return thenDo(toSome(x), orDo(orDo(orDo(tInt, tFlo), tStr), tBool)); }, [0]); -let tree = constParser(commandTPair); -console.log(util.inspect(tree, { showHidden: true, depth: null })); +let IMPORTEE = p.alt(p.tok(TokenKind.Number), p.tok(TokenKind.Op), p.tok(TokenKind.Paren), p.tok(TokenKind.Id), p.tok(TokenKind.Str), p.tok(TokenKind.SpaceNL), p.tok(TokenKind.Comment)); +let NOT_AT = p.alt(p.tok(TokenKind.Seperator), p.tok(TokenKind.Semicolon), p.tok(TokenKind.Number), p.tok(TokenKind.ExcapeAt), p.tok(TokenKind.Op), p.tok(TokenKind.Paren), p.tok(TokenKind.SpaceNL), p.tok(TokenKind.Id), p.tok(TokenKind.Str), p.tok(TokenKind.Comment)); +/** + * PROG : IMPORTS '---' CONTENT; + */ +PROG.setPattern(p.lrec_sc(IMPORTS, p.seq(p.str('---'), CONTENT), applyParts)); +/** + * NOT_AT_TEXT : NOT_AT + */ +NOT_AT_TEXT.setPattern(p.apply(NOT_AT, applyNotAtText)); +IMPORTS.setPattern(p.apply(p.seq(IMPORT, p.rep(IMPORT)), applyImports)); +/** + * IMPORT : + * 'import' IMPORTEE* SEMICOLON | + * COMMENT | + */ +IMPORT.setPattern(p.alt(p.apply(p.seq(p.str('import'), p.rep_sc(IMPORTEE), SEMICOLON), applyImport), p.apply(p.tok(TokenKind.Comment), applyComment), p.apply(p.tok(TokenKind.SpaceNL), applySpaceNL))); +/** + * SEMICOLON : ';'; + */ +SEMICOLON.setPattern(p.apply(p.tok(TokenKind.Semicolon), applySemiColon)); +/** + * SEGMENT : '@' NOT_AT* '@' | + * (NOT_AT_TEXT | EXCAPE_AT)* + */ +SEGMENT.setPattern(p.alt(p.apply(p.rep_sc(NOT_AT_TEXT), applyText), p.apply(p.seq(p.str('@'), p.rep(NOT_AT), p.str('@')), applySegment))); +/** + * CONTENT : SEGMENT* + */ +CONTENT.setPattern(p.apply(p.rep(SEGMENT), applyContent)); +let tree = p.expectSingleResult(p.expectEOF(PROG.parse(lexer.parse(inputTxt)))); +console.log("RESULT=" + tkTreeToSExp(tree));