| /* | |
| * Copyright (c) 2015, 2024, Oracle and/or its affiliates. All rights reserved. | |
| * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. | |
| * | |
| * This code is free software; you can redistribute it and/or modify it | |
| * under the terms of the GNU General Public License version 2 only, as | |
| * published by the Free Software Foundation. Oracle designates this | |
| * particular file as subject to the "Classpath" exception as provided | |
| * by Oracle in the LICENSE file that accompanied this code. | |
| * | |
| * This code is distributed in the hope that it will be useful, but WITHOUT | |
| * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or | |
| * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License | |
| * version 2 for more details (a copy is included in the LICENSE file that | |
| * accompanied this code). | |
| * | |
| * You should have received a copy of the GNU General Public License version | |
| * 2 along with this work; if not, write to the Free Software Foundation, | |
| * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. | |
| * | |
| * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA | |
| * or visit www.oracle.com if you need additional information or have any | |
| * questions. | |
| */ | |
| package jdk.jshell; | |
| import com.sun.tools.javac.code.Source; | |
| import com.sun.tools.javac.parser.Scanner; | |
| import com.sun.tools.javac.parser.ScannerFactory; | |
| import com.sun.tools.javac.parser.Tokens.Token; | |
| import com.sun.tools.javac.parser.Tokens.TokenKind; | |
| import com.sun.tools.javac.util.Context; | |
| import com.sun.tools.javac.util.DiagnosticSource; | |
| import com.sun.tools.javac.util.JCDiagnostic; | |
| import com.sun.tools.javac.util.JCDiagnostic.DiagnosticFlag; | |
| import com.sun.tools.javac.util.JCDiagnostic.DiagnosticPosition; | |
| import com.sun.tools.javac.util.JCDiagnostic.Error; | |
| import com.sun.tools.javac.util.Log; | |
| import java.io.PrintWriter; | |
| import java.io.StringWriter; | |
| import java.util.ArrayDeque; | |
| import java.util.Deque; | |
| import java.util.EnumMap; | |
| import java.util.Iterator; | |
| import jdk.jshell.SourceCodeAnalysis.Completeness; | |
| import com.sun.source.tree.Tree; | |
| import static jdk.jshell.CompletenessAnalyzer.TK.*; | |
| import jdk.jshell.TaskFactory.ParseTask; | |
| import jdk.jshell.TaskFactory.Worker; | |
| import java.util.List; | |
| import java.util.function.Function; | |
| import java.util.function.Supplier; | |
| import com.sun.tools.javac.util.Names; | |
| /** | |
| * Low level scanner to determine completeness of input. | |
| * @author Robert Field | |
| */ | |
| class CompletenessAnalyzer { | |
| private final ScannerFactory scannerFactory; | |
| private final JShell proc; | |
| private final Names names; | |
| private static Completeness error() { | |
| return Completeness.UNKNOWN; // For breakpointing | |
| } | |
| static class CaInfo { | |
| CaInfo(Completeness status, int unitEndPos) { | |
| this.status = status; | |
| this.unitEndPos = unitEndPos; | |
| } | |
| final int unitEndPos; | |
| final Completeness status; | |
| } | |
| CompletenessAnalyzer(JShell proc) { | |
| this.proc = proc; | |
| Context context = new Context(); | |
| Log log = CaLog.createLog(context); | |
| context.put(Log.class, log); | |
| context.put(Source.class, Source.JDK9); | |
| names = Names.instance(context); | |
| scannerFactory = ScannerFactory.instance(context); | |
| } | |
| CaInfo scan(String s) { | |
| try { | |
| Parser parser = new Parser( | |
| () -> new Matched(scannerFactory.newScanner(s, false)), | |
| names, | |
| worker -> proc.taskFactory.parse(s, worker)); | |
| Completeness stat = parser.parseUnit(); | |
| int endPos = stat == Completeness.UNKNOWN | |
| ? s.length() | |
| : parser.endPos(); | |
| return new CaInfo(stat, endPos); | |
| } catch (SyntaxException ex) { | |
| return new CaInfo(error(), s.length()); | |
| } | |
| } | |
| @SuppressWarnings("serial") // serialVersionUID intentionally omitted | |
| private static class SyntaxException extends RuntimeException { | |
| } | |
| private static void die() { | |
| throw new SyntaxException(); | |
| } | |
| /** | |
| * Subclass of Log used by compiler API to die on error and ignore | |
| * other messages | |
| */ | |
| private static class CaLog extends Log { | |
| private static CaLog createLog(Context context) { | |
| PrintWriter pw = new PrintWriter(new StringWriter()); | |
| CaLog log = new CaLog(context, pw); | |
| context.put(logKey, log); | |
| return log; | |
| } | |
| private CaLog(Context context, PrintWriter pw) { | |
| super(context, pw); | |
| this.source = DiagnosticSource.NO_SOURCE; | |
| } | |
| @Override | |
| public void error(String key, Object... args) { | |
| die(); | |
| } | |
| @Override | |
| public void error(int pos, Error errorKey) { | |
| die(); | |
| } | |
| @Override | |
| public void error(int pos, String key, Object... args) { | |
| die(); | |
| } | |
| @Override | |
| public void report(JCDiagnostic diagnostic) { | |
| // Ignore | |
| } | |
| } | |
| // Location position kinds -- a token is ... | |
| private static final int XEXPR = 0b1; // OK in expression (not first) | |
| private static final int XDECL = 0b10; // OK in declaration (not first) | |
| private static final int XSTMT = 0b100; // OK in statement framework (not first) | |
| private static final int XEXPR1o = 0b1000; // OK first in expression | |
| private static final int XDECL1o = 0b10000; // OK first in declaration | |
| private static final int XSTMT1o = 0b100000; // OK first or only in statement framework | |
| private static final int XEXPR1 = XEXPR1o | XEXPR; // OK in expression (anywhere) | |
| private static final int XDECL1 = XDECL1o | XDECL; // OK in declaration (anywhere) | |
| private static final int XSTMT1 = XSTMT1o | XSTMT; // OK in statement framework (anywhere) | |
| private static final int XANY1 = XEXPR1o | XDECL1o | XSTMT1o; // Mask: first in statement, declaration, or expression | |
| private static final int XTERM = 0b100000000; // Can terminate (last before EOF) | |
| private static final int XSTART = 0b1000000000; // Boundary, must be XTERM before | |
| private static final int XERRO = 0b10000000000; // Is an error | |
| private static final int XBRACESNEEDED = 0b100000000000; // Expect {ANY} LBRACE | |
| private static final int XMODIFIER = 0b1000000000000; // Modifier | |
| /** | |
| * An extension of the compiler's TokenKind which adds our combined/processed | |
| * kinds. Also associates each TK with a union of acceptable kinds of code | |
| * position it can occupy. For example: IDENTIFER is XEXPR1|XDECL1|XTERM, | |
| * meaning it can occur in expressions or declarations (but not in the | |
| * framework of a statement and that can be the final (terminating) token | |
| * in a snippet. | |
| * <P> | |
| * There must be a TK defined for each compiler TokenKind, an exception | |
| * will | |
| * be thrown if a TokenKind is defined and a corresponding TK is not. Add a | |
| * new TK in the appropriate category. If it is like an existing category | |
| * (e.g. a new modifier or type this may be all that is needed. If it | |
| * is bracketing or modifies the acceptable positions of other tokens, | |
| * please closely examine the needed changes to this scanner. | |
| */ | |
| static enum TK { | |
| // Special | |
| EOF(TokenKind.EOF, 0), // | |
| ERROR(TokenKind.ERROR, XERRO), // | |
| IDENTIFIER(TokenKind.IDENTIFIER, XEXPR1|XDECL1|XTERM), // | |
| UNDERSCORE(TokenKind.UNDERSCORE, XDECL1|XEXPR), // _ | |
| CLASS(TokenKind.CLASS, XEXPR|XDECL1|XBRACESNEEDED), // class decl (MAPPED: DOTCLASS) | |
| MONKEYS_AT(TokenKind.MONKEYS_AT, XEXPR|XDECL1), // @ | |
| IMPORT(TokenKind.IMPORT, XDECL1|XSTART), // import -- consider declaration | |
| SEMI(TokenKind.SEMI, XSTMT1|XTERM|XSTART), // ; | |
| // Shouldn't see -- error | |
| PACKAGE(TokenKind.PACKAGE, XERRO), // package | |
| CONST(TokenKind.CONST, XERRO), // reserved keyword -- const | |
| GOTO(TokenKind.GOTO, XERRO), // reserved keyword -- goto | |
| CUSTOM(TokenKind.CUSTOM, XERRO), // No uses | |
| // Declarations | |
| ENUM(TokenKind.ENUM, XDECL1|XBRACESNEEDED), // enum | |
| IMPLEMENTS(TokenKind.IMPLEMENTS, XDECL), // implements | |
| INTERFACE(TokenKind.INTERFACE, XDECL1|XBRACESNEEDED), // interface | |
| THROWS(TokenKind.THROWS, XDECL|XBRACESNEEDED), // throws | |
| // Primarive type names | |
| BOOLEAN(TokenKind.BOOLEAN, XEXPR1|XDECL1|XTERM), // boolean | |
| BYTE(TokenKind.BYTE, XEXPR1|XDECL1|XTERM), // byte | |
| CHAR(TokenKind.CHAR, XEXPR1|XDECL1|XTERM), // char | |
| DOUBLE(TokenKind.DOUBLE, XEXPR1|XDECL1|XTERM), // double | |
| FLOAT(TokenKind.FLOAT, XEXPR1|XDECL1|XTERM), // float | |
| INT(TokenKind.INT, XEXPR1|XDECL1|XTERM), // int | |
| LONG(TokenKind.LONG, XEXPR1|XDECL1|XTERM), // long | |
| SHORT(TokenKind.SHORT, XEXPR1|XDECL1|XTERM), // short | |
| VOID(TokenKind.VOID, XEXPR1|XDECL1), // void | |
| // Modifiers keywords | |
| ABSTRACT(TokenKind.ABSTRACT, XDECL1 | XMODIFIER), // abstract | |
| FINAL(TokenKind.FINAL, XDECL1 | XMODIFIER), // final | |
| NATIVE(TokenKind.NATIVE, XDECL1 | XMODIFIER), // native | |
| STATIC(TokenKind.STATIC, XDECL1 | XMODIFIER), // static | |
| STRICTFP(TokenKind.STRICTFP, XDECL1 | XMODIFIER), // strictfp | |
| PRIVATE(TokenKind.PRIVATE, XDECL1 | XMODIFIER), // private | |
| PROTECTED(TokenKind.PROTECTED, XDECL1 | XMODIFIER), // protected | |
| PUBLIC(TokenKind.PUBLIC, XDECL1 | XMODIFIER), // public | |
| TRANSIENT(TokenKind.TRANSIENT, XDECL1 | XMODIFIER), // transient | |
| VOLATILE(TokenKind.VOLATILE, XDECL1 | XMODIFIER), // volatile | |
| // Declarations and type parameters (thus expressions) | |
| EXTENDS(TokenKind.EXTENDS, XEXPR|XDECL), // extends | |
| COMMA(TokenKind.COMMA, XEXPR|XDECL), // , | |
| AMP(TokenKind.AMP, XEXPR|XDECL, true), // & | |
| GT(TokenKind.GT, XEXPR|XDECL, true), // > | |
| LT(TokenKind.LT, XEXPR|XDECL1, true), // < | |
| LTLT(TokenKind.LTLT, XEXPR|XDECL1, true), // << | |
| GTGT(TokenKind.GTGT, XEXPR|XDECL, true), // >> | |
| GTGTGT(TokenKind.GTGTGT, XEXPR|XDECL, true), // >>> | |
| QUES(TokenKind.QUES, XEXPR|XDECL, true), // ? | |
| DOT(TokenKind.DOT, XEXPR|XDECL), // . | |
| STAR(TokenKind.STAR, XEXPR, true), // * (MAPPED: DOTSTAR) | |
| // Statement keywords | |
| ASSERT(TokenKind.ASSERT, XSTMT1|XSTART), // assert | |
| BREAK(TokenKind.BREAK, XSTMT1|XTERM|XSTART), // break | |
| CATCH(TokenKind.CATCH, XSTMT1|XSTART), // catch | |
| CONTINUE(TokenKind.CONTINUE, XSTMT1|XTERM|XSTART), // continue | |
| DO(TokenKind.DO, XSTMT1|XSTART), // do | |
| ELSE(TokenKind.ELSE, XSTMT1|XTERM|XSTART), // else | |
| FINALLY(TokenKind.FINALLY, XSTMT1|XSTART), // finally | |
| FOR(TokenKind.FOR, XSTMT1|XSTART), // for | |
| IF(TokenKind.IF, XSTMT1|XSTART), // if | |
| RETURN(TokenKind.RETURN, XSTMT1|XTERM|XSTART), // return | |
| SWITCH(TokenKind.SWITCH, XSTMT1|XEXPR1), // switch | |
| SYNCHRONIZED(TokenKind.SYNCHRONIZED, XSTMT1|XDECL), // synchronized | |
| THROW(TokenKind.THROW, XSTMT1|XSTART), // throw | |
| TRY(TokenKind.TRY, XSTMT1|XSTART), // try | |
| WHILE(TokenKind.WHILE, XSTMT1|XSTART), // while | |
| // Statement keywords that we shouldn't see -- inside braces | |
| CASE(TokenKind.CASE, XSTMT|XSTART), // case | |
| DEFAULT(TokenKind.DEFAULT, XSTMT|XSTART), // default method, default case -- neither we should see | |
| // Expressions (can terminate) | |
| INTLITERAL(TokenKind.INTLITERAL, XEXPR1|XTERM), // | |
| LONGLITERAL(TokenKind.LONGLITERAL, XEXPR1|XTERM), // | |
| FLOATLITERAL(TokenKind.FLOATLITERAL, XEXPR1|XTERM), // | |
| DOUBLELITERAL(TokenKind.DOUBLELITERAL, XEXPR1|XTERM), // | |
| CHARLITERAL(TokenKind.CHARLITERAL, XEXPR1|XTERM), // | |
| STRINGLITERAL(TokenKind.STRINGLITERAL, XEXPR1|XTERM), // | |
| STRINGFRAGMENT(TokenKind.STRINGFRAGMENT, XEXPR1|XTERM), | |
| TRUE(TokenKind.TRUE, XEXPR1|XTERM), // true | |
| FALSE(TokenKind.FALSE, XEXPR1|XTERM), // false | |
| NULL(TokenKind.NULL, XEXPR1|XTERM), // null | |
| THIS(TokenKind.THIS, XEXPR1|XTERM), // this -- shouldn't see | |
| // Expressions maybe terminate //TODO handle these case separately | |
| PLUSPLUS(TokenKind.PLUSPLUS, XEXPR1|XTERM), // ++ | |
| SUBSUB(TokenKind.SUBSUB, XEXPR1|XTERM), // -- | |
| // Expressions cannot terminate | |
| INSTANCEOF(TokenKind.INSTANCEOF, XEXPR, true), // instanceof | |
| NEW(TokenKind.NEW, XEXPR1), // new (MAPPED: COLCOLNEW) | |
| SUPER(TokenKind.SUPER, XEXPR1|XDECL), // super -- shouldn't see as rec. But in type parameters | |
| ARROW(TokenKind.ARROW, XEXPR), // -> | |
| COLCOL(TokenKind.COLCOL, XEXPR), // :: | |
| LPAREN(TokenKind.LPAREN, XEXPR), // ( | |
| RPAREN(TokenKind.RPAREN, XEXPR), // ) | |
| LBRACE(TokenKind.LBRACE, XEXPR), // { | |
| RBRACE(TokenKind.RBRACE, XEXPR), // } | |
| LBRACKET(TokenKind.LBRACKET, XEXPR), // [ | |
| RBRACKET(TokenKind.RBRACKET, XEXPR), // ] | |
| ELLIPSIS(TokenKind.ELLIPSIS, XEXPR), // ... | |
| EQ(TokenKind.EQ, XEXPR), // = | |
| BANG(TokenKind.BANG, XEXPR1), // ! | |
| TILDE(TokenKind.TILDE, XEXPR1), // ~ | |
| COLON(TokenKind.COLON, XEXPR|XTERM), // : | |
| EQEQ(TokenKind.EQEQ, XEXPR, true), // == | |
| LTEQ(TokenKind.LTEQ, XEXPR, true), // <= | |
| GTEQ(TokenKind.GTEQ, XEXPR, true), // >= | |
| BANGEQ(TokenKind.BANGEQ, XEXPR, true), // != | |
| AMPAMP(TokenKind.AMPAMP, XEXPR, true), // && | |
| BARBAR(TokenKind.BARBAR, XEXPR, true), // || | |
| PLUS(TokenKind.PLUS, XEXPR1, true), // + | |
| SUB(TokenKind.SUB, XEXPR1 | XDECL, true), // - | |
| SLASH(TokenKind.SLASH, XEXPR, true), // / | |
| BAR(TokenKind.BAR, XEXPR, true), // | | |
| CARET(TokenKind.CARET, XEXPR, true), // ^ | |
| PERCENT(TokenKind.PERCENT, XEXPR, true), // % | |
| PLUSEQ(TokenKind.PLUSEQ, XEXPR), // += | |
| SUBEQ(TokenKind.SUBEQ, XEXPR), // -= | |
| STAREQ(TokenKind.STAREQ, XEXPR), // *= | |
| SLASHEQ(TokenKind.SLASHEQ, XEXPR), // /= | |
| AMPEQ(TokenKind.AMPEQ, XEXPR), // &= | |
| BAREQ(TokenKind.BAREQ, XEXPR), // |= | |
| CARETEQ(TokenKind.CARETEQ, XEXPR), // ^= | |
| PERCENTEQ(TokenKind.PERCENTEQ, XEXPR), // %= | |
| LTLTEQ(TokenKind.LTLTEQ, XEXPR), // <<= | |
| GTGTEQ(TokenKind.GTGTEQ, XEXPR), // >>= | |
| GTGTGTEQ(TokenKind.GTGTGTEQ, XEXPR), // >>>= | |
| // combined/processed kinds | |
| UNMATCHED(XERRO), | |
| PARENS(XEXPR1|XDECL|XSTMT|XTERM), | |
| BRACKETS(XEXPR|XDECL|XTERM), | |
| BRACES(XSTMT1|XEXPR|XTERM), | |
| DOTSTAR(XDECL|XTERM), // import foo.* | |
| COLCOLNEW(XEXPR|XTERM), // :: new | |
| DOTCLASS(XEXPR|XTERM), // class decl and .class | |
| ; | |
| static final EnumMap<TokenKind,TK> tokenKindToTKMap = new EnumMap<>(TokenKind.class); | |
| final TokenKind tokenKind; | |
| final int belongs; | |
| final boolean valueOp; | |
| Function<TK,TK> mapping; | |
| TK(int b) { | |
| this(null, b); | |
| } | |
| TK(TokenKind tokenKind, int b) { | |
| this(tokenKind, b, false); | |
| } | |
| TK(TokenKind tokenKind, int b, boolean valueOp) { | |
| this.tokenKind = tokenKind; | |
| this.belongs = b; | |
| this.valueOp = valueOp; | |
| this.mapping = null; | |
| } | |
| private static TK tokenKindToTK(TK prev, TokenKind kind) { | |
| TK tk = tokenKindToTKMap.get(kind); | |
| if (tk == null) { | |
| System.err.printf("No corresponding %s for %s: %s\n", | |
| TK.class.getCanonicalName(), | |
| TokenKind.class.getCanonicalName(), | |
| kind); | |
| throw new InternalError("No corresponding TK for TokenKind: " + kind); | |
| } | |
| return tk.mapping != null | |
| ? tk.mapping.apply(prev) | |
| : tk; | |
| } | |
| boolean isOkToTerminate() { | |
| return (belongs & XTERM) != 0; | |
| } | |
| boolean isExpression() { | |
| return (belongs & XEXPR) != 0; | |
| } | |
| boolean isDeclaration() { | |
| return (belongs & XDECL) != 0; | |
| } | |
| boolean isError() { | |
| return (belongs & XERRO) != 0; | |
| } | |
| boolean isStart() { | |
| return (belongs & XSTART) != 0; | |
| } | |
| boolean isBracesNeeded() { | |
| return (belongs & XBRACESNEEDED) != 0; | |
| } | |
| boolean isModifier() { | |
| return (belongs & XMODIFIER) != 0; | |
| } | |
| /** | |
| * After construction, check that all compiler TokenKind values have | |
| * corresponding TK values. | |
| */ | |
| static { | |
| for (TK tk : TK.values()) { | |
| if (tk.tokenKind != null) { | |
| tokenKindToTKMap.put(tk.tokenKind, tk); | |
| } | |
| } | |
| for (TokenKind kind : TokenKind.values()) { | |
| tokenKindToTK(null, kind); // assure they can be retrieved without error | |
| } | |
| // Mappings of disambiguated contexts | |
| STAR.mapping = prev -> prev == DOT ? DOTSTAR : STAR; | |
| NEW.mapping = prev -> prev == COLCOL ? COLCOLNEW : NEW; | |
| CLASS.mapping = prev -> prev == DOT ? DOTCLASS : CLASS; | |
| } | |
| } | |
| /** | |
| * A completeness scanner token. | |
| */ | |
| private static class CT { | |
| /** The token kind */ | |
| public final TK kind; | |
| /** The end position of this token */ | |
| public final int endPos; | |
| /** The error message **/ | |
| public final String message; | |
| public final Token tok; | |
| private CT(TK tk, Token tok, String msg) { | |
| this.kind = tk; | |
| this.endPos = tok.endPos; | |
| this.message = msg; | |
| this.tok = tok; | |
| //throw new InternalError(msg); /* for debugging */ | |
| } | |
| private CT(TK tk, Token tok) { | |
| this.kind = tk; | |
| this.endPos = tok.endPos; | |
| this.message = null; | |
| this.tok = tok; | |
| } | |
| private CT(TK tk, int endPos) { | |
| this.kind = tk; | |
| this.endPos = endPos; | |
| this.message = null; | |
| this.tok = null; | |
| } | |
| } | |
| /** | |
| * Look for matching tokens (like parens) and other special cases, like "new" | |
| */ | |
| private static class Matched implements Iterator<CT> { | |
| private final Scanner scanner; | |
| private Token current; | |
| private CT prevCT; | |
| private CT currentCT; | |
| private final Deque<Token> stack = new ArrayDeque<>(); | |
| Matched(Scanner scanner) { | |
| this.scanner = scanner; | |
| advance(); | |
| prevCT = currentCT = new CT(SEMI, 0); // So is valid for testing | |
| } | |
| @Override | |
| public boolean hasNext() { | |
| return currentCT.kind != EOF; | |
| } | |
| private Token advance() { | |
| Token prev = current; | |
| if (current != null && current.kind == TokenKind.STRINGFRAGMENT) { | |
| int endPos = current.endPos; | |
| do { | |
| scanner.nextToken(); | |
| current = scanner.token(); | |
| } while (current != null && current.endPos <= endPos && current.kind != TokenKind.EOF); | |
| } else { | |
| scanner.nextToken(); | |
| current = scanner.token(); | |
| } | |
| return prev; | |
| } | |
| @Override | |
| public CT next() { | |
| prevCT = currentCT; | |
| currentCT = nextCT(); | |
| return currentCT; | |
| } | |
| private CT match(TK tk, TokenKind open) { | |
| Token tok = advance(); | |
| db("match desired-tk=%s, open=%s, seen-tok=%s", tk, open, tok.kind); | |
| if (stack.isEmpty()) { | |
| return new CT(ERROR, tok, "Encountered '" + tok + "' with no opening '" + open + "'"); | |
| } | |
| Token p = stack.pop(); | |
| if (p.kind != open) { | |
| return new CT(ERROR, tok, "No match for '" + p + "' instead encountered '" + tok + "'"); | |
| } | |
| return new CT(tk, tok); | |
| } | |
| private void db(String format, Object ... args) { | |
| // System.err.printf(format, args); | |
| // System.err.printf(" -- stack("); | |
| // if (stack.isEmpty()) { | |
| // | |
| // } else { | |
| // for (Token tok : stack) { | |
| // System.err.printf("%s ", tok.kind); | |
| // } | |
| // } | |
| // System.err.printf(") current=%s / currentCT=%s\n", current.kind, currentCT.kind); | |
| } | |
| /** | |
| * @return the next scanner token | |
| */ | |
| private CT nextCT() { | |
| // TODO Annotations? | |
| TK prevTK = currentCT.kind; | |
| while (true) { | |
| db("nextCT"); | |
| CT ct; | |
| switch (current.kind) { | |
| case EOF: | |
| db("eof"); | |
| if (stack.isEmpty()) { | |
| ct = new CT(EOF, current); | |
| } else { | |
| TokenKind unmatched = stack.pop().kind; | |
| stack.clear(); // So we will get EOF next time | |
| ct = new CT(UNMATCHED, current, "Unmatched " + unmatched); | |
| } | |
| break; | |
| case LPAREN: | |
| case LBRACE: | |
| case LBRACKET: | |
| stack.push(advance()); | |
| prevTK = SEMI; // new start | |
| continue; | |
| case RPAREN: | |
| ct = match(PARENS, TokenKind.LPAREN); | |
| break; | |
| case RBRACE: | |
| ct = match(BRACES, TokenKind.LBRACE); | |
| break; | |
| case RBRACKET: | |
| ct = match(BRACKETS, TokenKind.LBRACKET); | |
| break; | |
| default: | |
| ct = new CT(TK.tokenKindToTK(prevTK, current.kind), advance()); | |
| break; | |
| } | |
| // Detect an error if we are at starting position and the last | |
| // token wasn't a terminating one. Special cases: | |
| // -within braces, comma can procede semicolon, e.g. the values list in enum | |
| // -arrow can be followed by a throw, e.g. in a switch/switch expression | |
| if (ct.kind.isStart() && !prevTK.isOkToTerminate() && prevTK != COMMA && | |
| !(prevTK == ARROW && ct.kind == THROW)) { | |
| return new CT(ERROR, current, "No '" + prevTK + "' before '" + ct.kind + "'"); | |
| } | |
| if (stack.isEmpty() || ct.kind.isError()) { | |
| return ct; | |
| } | |
| prevTK = ct.kind; | |
| } | |
| } | |
| } | |
| /** | |
| * Fuzzy parser based on token kinds | |
| */ | |
| private static class Parser { | |
| private final Supplier<Matched> matchedFactory; | |
| private final Function<Worker<ParseTask, Completeness>, Completeness> parseFactory; | |
| private Matched in; | |
| private CT token; | |
| private Completeness checkResult; | |
| private final Names names; | |
| Parser(Supplier<Matched> matchedFactory, | |
| Names names, | |
| Function<Worker<ParseTask, Completeness>, Completeness> parseFactory) { | |
| this.matchedFactory = matchedFactory; | |
| this.parseFactory = parseFactory; | |
| this.names = names; | |
| resetInput(); | |
| } | |
| final void resetInput() { | |
| this.in = matchedFactory.get(); | |
| nextToken(); | |
| } | |
| final void nextToken() { | |
| in.next(); | |
| token = in.currentCT; | |
| } | |
| boolean shouldAbort(TK tk) { | |
| if (token.kind == tk) { | |
| nextToken(); | |
| return false; | |
| } | |
| switch (token.kind) { | |
| case EOF: | |
| checkResult = ((tk == SEMI) && in.prevCT.kind.isOkToTerminate()) | |
| ? Completeness.COMPLETE_WITH_SEMI | |
| : Completeness.DEFINITELY_INCOMPLETE; | |
| return true; | |
| case UNMATCHED: | |
| checkResult = Completeness.DEFINITELY_INCOMPLETE; | |
| return true; | |
| default: | |
| checkResult = error(); | |
| return true; | |
| } | |
| } | |
| Completeness lastly(TK tk) { | |
| if (shouldAbort(tk)) return checkResult; | |
| return Completeness.COMPLETE; | |
| } | |
| Completeness optionalFinalSemi() { | |
| if (!shouldAbort(SEMI)) return Completeness.COMPLETE; | |
| if (checkResult == Completeness.COMPLETE_WITH_SEMI) return Completeness.COMPLETE; | |
| return checkResult; | |
| } | |
| boolean shouldAbort(Completeness flags) { | |
| checkResult = flags; | |
| return flags != Completeness.COMPLETE; | |
| } | |
| public int endPos() { | |
| return in.prevCT.endPos; | |
| } | |
| public Completeness parseUnit() { | |
| //System.err.printf("%s: belongs %o XANY1 %o\n", token.kind, token.kind.belongs, token.kind.belongs & XANY1); | |
| switch (token.kind.belongs & XANY1) { | |
| case XEXPR1o: | |
| return parseExpressionOptionalSemi(); | |
| case XSTMT1o: { | |
| Completeness stat = parseSimpleStatement(); | |
| return stat==null? error() : stat; | |
| } | |
| case XDECL1o: | |
| return parseDeclaration(); | |
| case XSTMT1o | XDECL1o: | |
| case XEXPR1o | XDECL1o: | |
| return disambiguateDeclarationVsExpression(); | |
| case 0: | |
| if ((token.kind.belongs & XERRO) != 0) { | |
| return parseExpressionStatement(); // Let this gen the status | |
| } | |
| return error(); | |
| case XSTMT1o | XEXPR1o: | |
| return disambiguateStatementVsExpression(); | |
| default: | |
| throw new InternalError("Case not covered " + token.kind.belongs + " in " + token.kind); | |
| } | |
| } | |
| public Completeness parseDeclaration() { | |
| boolean isImport = token.kind == IMPORT; | |
| boolean isRecord = false; | |
| boolean afterModifiers = false; | |
| boolean isBracesNeeded = false; | |
| while (token.kind.isDeclaration()) { | |
| isBracesNeeded |= token.kind.isBracesNeeded(); | |
| isRecord |= !afterModifiers && token.kind == TK.IDENTIFIER && token.tok.name() == names.record; | |
| afterModifiers |= !token.kind.isModifier(); | |
| nextToken(); | |
| } | |
| switch (token.kind) { | |
| case EQ: | |
| nextToken(); | |
| return parseExpressionStatement(); | |
| case BRACES: | |
| case SEMI: | |
| nextToken(); | |
| return Completeness.COMPLETE; | |
| case UNMATCHED: | |
| nextToken(); | |
| return Completeness.DEFINITELY_INCOMPLETE; | |
| case EOF: | |
| switch (in.prevCT.kind) { | |
| case BRACES: | |
| case SEMI: | |
| return Completeness.COMPLETE; | |
| case IDENTIFIER: | |
| return isBracesNeeded || isRecord | |
| ? Completeness.DEFINITELY_INCOMPLETE | |
| : Completeness.COMPLETE_WITH_SEMI; | |
| case BRACKETS: | |
| return Completeness.COMPLETE_WITH_SEMI; | |
| case DOTSTAR: | |
| if (isImport) { | |
| return Completeness.COMPLETE_WITH_SEMI; | |
| } else { | |
| return Completeness.UNKNOWN; | |
| } | |
| default: | |
| return Completeness.DEFINITELY_INCOMPLETE; | |
| } | |
| default: | |
| return error(); | |
| } | |
| } | |
| public Completeness disambiguateStatementVsExpression() { | |
| if (token.kind == SWITCH) { | |
| nextToken(); | |
| switch (token.kind) { | |
| case PARENS: | |
| nextToken(); | |
| break; | |
| case UNMATCHED: | |
| nextToken(); | |
| return Completeness.DEFINITELY_INCOMPLETE; | |
| case EOF: | |
| return Completeness.DEFINITELY_INCOMPLETE; | |
| default: | |
| return error(); | |
| } | |
| switch (token.kind) { | |
| case BRACES: | |
| nextToken(); | |
| break; | |
| case UNMATCHED: | |
| nextToken(); | |
| return Completeness.DEFINITELY_INCOMPLETE; | |
| case EOF: | |
| return Completeness.DEFINITELY_INCOMPLETE; | |
| default: | |
| return error(); | |
| } | |
| if (token.kind.valueOp) { | |
| return parseExpressionOptionalSemi(); | |
| } else { | |
| return Completeness.COMPLETE; | |
| } | |
| } else { | |
| throw new InternalError("Unexpected statement/expression not covered " + token.kind.belongs + " in " + token.kind); | |
| } | |
| } | |
| public Completeness disambiguateDeclarationVsExpression() { | |
| // String folding messes up position information. | |
| return parseFactory.apply(pt -> { | |
| List<? extends Tree> units = pt.units(); | |
| if (units.isEmpty()) { | |
| return error(); | |
| } | |
| Tree unitTree = units.get(0); | |
| switch (unitTree.getKind()) { | |
| case EXPRESSION_STATEMENT: | |
| return parseExpressionOptionalSemi(); | |
| case LABELED_STATEMENT: | |
| if (shouldAbort(IDENTIFIER)) return checkResult; | |
| if (shouldAbort(COLON)) return checkResult; | |
| return parseStatement(); | |
| case VARIABLE: | |
| case IMPORT: | |
| case CLASS: | |
| case ENUM: | |
| case ANNOTATION_TYPE: | |
| case INTERFACE: | |
| case RECORD: | |
| case METHOD: | |
| return parseDeclaration(); | |
| default: | |
| return error(); | |
| } | |
| }); | |
| } | |
| public Completeness parseExpressionStatement() { | |
| if (shouldAbort(parseExpression())) return checkResult; | |
| return lastly(SEMI); | |
| } | |
| public Completeness parseExpressionOptionalSemi() { | |
| if (shouldAbort(parseExpression())) return checkResult; | |
| return optionalFinalSemi(); | |
| } | |
| public Completeness parseExpression() { | |
| while (token.kind.isExpression()) { | |
| CT prevToken = in.prevCT; | |
| nextToken(); | |
| // primitive types can only appear in the end of an `instanceof` expression | |
| switch (token.kind) { | |
| case EOF: | |
| switch (in.prevCT.kind) { | |
| case BYTE, SHORT, CHAR, INT, LONG, FLOAT, DOUBLE, BOOLEAN: | |
| switch (prevToken.kind) { | |
| case INSTANCEOF: | |
| return Completeness.COMPLETE; | |
| default: | |
| return Completeness.DEFINITELY_INCOMPLETE; | |
| } | |
| } | |
| } | |
| } | |
| return Completeness.COMPLETE; | |
| } | |
| public Completeness parseStatement() { | |
| Completeness stat = parseSimpleStatement(); | |
| if (stat == null) { | |
| return parseExpressionStatement(); | |
| } | |
| return stat; | |
| } | |
| /** | |
| * Statement = Block | IF ParExpression Statement [ELSE Statement] | FOR | |
| * "(" ForInitOpt ";" [Expression] ";" ForUpdateOpt ")" Statement | FOR | |
| * "(" FormalParameter : Expression ")" Statement | WHILE ParExpression | |
| * Statement | DO Statement WHILE ParExpression ";" | TRY Block ( | |
| * Catches | [Catches] FinallyPart ) | TRY "(" ResourceSpecification | |
| * ";"opt ")" Block [Catches] [FinallyPart] | SWITCH ParExpression "{" | |
| * SwitchBlockStatementGroups "}" | SYNCHRONIZED ParExpression Block | | |
| * RETURN [Expression] ";" | THROW Expression ";" | BREAK [Ident] ";" | | |
| * CONTINUE [Ident] ";" | ASSERT Expression [ ":" Expression ] ";" | ";" | |
| */ | |
| public Completeness parseSimpleStatement() { | |
| switch (token.kind) { | |
| case BRACES: | |
| return lastly(BRACES); | |
| case IF: { | |
| nextToken(); | |
| if (shouldAbort(PARENS)) return checkResult; | |
| Completeness thenpart = parseStatement(); | |
| if (shouldAbort(thenpart)) return thenpart; | |
| if (token.kind == ELSE) { | |
| nextToken(); | |
| return parseStatement(); | |
| } | |
| return thenpart; | |
| } | |
| case FOR: { | |
| nextToken(); | |
| if (shouldAbort(PARENS)) return checkResult; | |
| if (shouldAbort(parseStatement())) return checkResult; | |
| return Completeness.COMPLETE; | |
| } | |
| case WHILE: { | |
| nextToken(); | |
| if (shouldAbort(PARENS)) return error(); | |
| return parseStatement(); | |
| } | |
| case DO: { | |
| nextToken(); | |
| switch (parseStatement()) { | |
| case DEFINITELY_INCOMPLETE: | |
| case CONSIDERED_INCOMPLETE: | |
| case COMPLETE_WITH_SEMI: | |
| return Completeness.DEFINITELY_INCOMPLETE; | |
| case UNKNOWN: | |
| return error(); | |
| case COMPLETE: | |
| break; | |
| } | |
| if (shouldAbort(WHILE)) return checkResult; | |
| if (shouldAbort(PARENS)) return checkResult; | |
| return lastly(SEMI); | |
| } | |
| case TRY: { | |
| boolean hasResources = false; | |
| nextToken(); | |
| if (token.kind == PARENS) { | |
| nextToken(); | |
| hasResources = true; | |
| } | |
| if (shouldAbort(BRACES)) return checkResult; | |
| if (token.kind == CATCH || token.kind == FINALLY) { | |
| while (token.kind == CATCH) { | |
| if (shouldAbort(CATCH)) return checkResult; | |
| if (shouldAbort(PARENS)) return checkResult; | |
| if (shouldAbort(BRACES)) return checkResult; | |
| } | |
| if (token.kind == FINALLY) { | |
| if (shouldAbort(FINALLY)) return checkResult; | |
| if (shouldAbort(BRACES)) return checkResult; | |
| } | |
| } else if (!hasResources) { | |
| if (token.kind == EOF) { | |
| return Completeness.DEFINITELY_INCOMPLETE; | |
| } else { | |
| return error(); | |
| } | |
| } | |
| return Completeness.COMPLETE; | |
| } | |
| case SWITCH: { | |
| nextToken(); | |
| if (shouldAbort(PARENS)) return checkResult; | |
| return lastly(BRACES); | |
| } | |
| case SYNCHRONIZED: { | |
| nextToken(); | |
| if (shouldAbort(PARENS)) return checkResult; | |
| return lastly(BRACES); | |
| } | |
| case THROW: { | |
| nextToken(); | |
| if (shouldAbort(parseExpression())) return checkResult; | |
| return lastly(SEMI); | |
| } | |
| case SEMI: | |
| return lastly(SEMI); | |
| case ASSERT: | |
| nextToken(); | |
| // Crude expression parsing just happily eats the optional colon | |
| return parseExpressionStatement(); | |
| case RETURN: | |
| case BREAK: | |
| case CONTINUE: | |
| nextToken(); | |
| return parseExpressionStatement(); | |
| // What are these doing here? | |
| case ELSE: | |
| case FINALLY: | |
| case CATCH: | |
| return error(); | |
| case EOF: | |
| return Completeness.CONSIDERED_INCOMPLETE; | |
| default: | |
| return null; | |
| } | |
| } | |
| } | |
| } |