Many resources are needed to download a project. Please understand that we have to compensate our server costs. Thank you in advance. Project price only 1 $
You can buy this project and download/modify it how often you want.
/***************************************************************************
* Copyright (C) 2006-2015 by Fabrizio Montesi *
* Copyright (C) 2021-2022 Vicki Mixen *
* *
* This program is free software; you can redistribute it and/or modify *
* it under the terms of the GNU Library General Public License as *
* published by the Free Software Foundation; either version 2 of the *
* License, or (at your option) any later version. *
* *
* This program 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 for more details. *
* *
* You should have received a copy of the GNU Library General Public *
* License along with this program; if not, write to the *
* Free Software Foundation, Inc., *
* 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. *
* *
* For details about the authors of this software, see the AUTHORS file. *
***************************************************************************/
package jolie.lang.parse;
import java.io.EOFException;
import java.io.IOException;
import java.util.ArrayList;
import java.util.Collection;
import java.util.List;
import org.apache.commons.text.similarity.LevenshteinDistance;
import jolie.lang.parse.context.ParsingContext;
import jolie.lang.parse.context.URIParsingContext;
import jolie.lang.CodeCheckMessage;
import jolie.lang.Keywords;
/**
* Skeleton implementation of a parser based on {@link jolie.lang.parse.Scanner}. Note that the
* parsing process is not re-entrant.
*
* @author Fabrizio Montesi
* @see Scanner
*/
public abstract class AbstractParser {
private Scanner scanner; // Input scanner.
protected Scanner.Token token; /// < The current token.
private final List< Scanner.Token > tokens = new ArrayList<>();
private final StringBuilder stringBuilder = new StringBuilder( 256 );
private boolean backup = false;
private final List< Scanner.Token > backupTokens = new ArrayList<>();
private boolean metNewline = false;
protected final String build( String... args ) {
stringBuilder.setLength( 0 );
for( String s : args ) {
stringBuilder.append( s );
}
return stringBuilder.toString();
}
/**
* Constructor
*
* @param scanner The scanner to use during the parsing procedure.
*/
public AbstractParser( Scanner scanner ) {
this.scanner = scanner;
}
protected final void addTokens( Collection< Scanner.Token > tokens ) {
this.tokens.addAll( tokens );
}
protected final void addToken( Scanner.Token token ) {
this.tokens.add( token );
}
protected final void prependToken( Scanner.Token prefixToken ) {
addToken( prefixToken );
addToken( token );
}
private void readToken()
throws IOException {
if( tokens.isEmpty() ) {
token = scanner.getToken();
} else {
token = tokens.remove( 0 );
}
}
/**
* Gets a new token. Whitespace tokens are ignored, but the metNewline flag is set.
*
* @throws IOException If the internal scanner raises one.
*/
protected final void nextToken()
throws IOException {
metNewline = false;
boolean run;
do {
readToken();
run = token.is( Scanner.TokenType.NEWLINE );
metNewline = metNewline || run;
} while( run );
if( backup ) {
backupTokens.add( token );
}
}
/**
* Used to read the rest of the line, after an error has occured.
*/
public final void readLineAfterError() throws IOException {
try {
scanner.readLineAfterError();
} catch( Exception e ) {
// Maybe add text : "Error occured while creating parserException."
throw new IOException( e );
}
}
protected final boolean hasMetNewline() {
return metNewline;
}
/** Recovers the backed up tokens. */
protected final void recoverBackup()
throws IOException {
backup = false;
if( !backupTokens.isEmpty() ) {
addTokens( backupTokens );
backupTokens.clear();
nextToken();
}
}
/** Discards the backed up tokens. */
protected final void discardBackup() {
backup = false;
backupTokens.clear();
}
protected void startBackup() {
if( token != null ) {
backupTokens.add( token );
}
backup = true;
}
/**
* Gets a new token, and throws an {@link EOFException} if such token is of type
* {@code jolie.lang.parse.Scanner.TokenType.EOF}.
*
* @throws IOException If the internal scanner raises one.
* @throws EOFException If the next token is of type {@code jolie.lang.parse.Scanner.Token.EOF}
*/
protected final void nextTokenNotEOF()
throws IOException, EOFException {
nextToken();
if( token.isEOF() ) {
throw new EOFException();
}
}
/**
* Returns the Scanner object used by this parser.
*
* @return The Scanner used by this parser.
*/
public final Scanner scanner() {
return scanner;
}
protected final void setScanner( Scanner scanner ) {
this.scanner = scanner;
}
/**
* Returns a simple {@link ParsingContext} from the underlying {@link Scanner} and does not take
* startLine and endLine into account
*
* @return the current {@link ParsingContext} from the underlying {@link Scanner}
*/
public final ParsingContext getContext() {
return new URIParsingContext( scanner.source(), scanner.line(), scanner.line(), scanner.errorColumn(),
scanner.errorColumn() + token.content().length(),
scanner.codeLine() );
}
/**
* Returns the current {@link ParsingContext} from the underlying {@link Scanner} taking start and
* end line into account
*
* @return the current {@link ParsingContext} from the underlying {@link Scanner}
*/
public final ParsingContext getContextDuringError() throws IOException {
// read the rest of the line, so we can use the whole line in the error message
readLineAfterError();
int lineState = scanner.lineState();
// If startline, endline and line are 0, the line is 0. If startline and endline are 0, but line is
// not then startline and endline have not been set yet and line is line
if( lineState == 0 ) {
if( scanner.errorColumn() == -1 && scanner.line() > 0 ) {
// If the errorcolumn is -1 then we have read a newline character but nothing on the new line
// meaning we do not need what is on the last line, so we remove that line
// and change the errorcolumn to the last character on the net to last line
scanner.codeLine().remove( scanner.codeLine().size() - 1 );
int column = scanner.codeLine().get( scanner.codeLine().size() - 1 ).length() - 1; // last index of
// (new)last line
URIParsingContext newContext =
new URIParsingContext( scanner.source(), scanner.line() - 1, scanner.line() - 1,
column, column, scanner.codeLine() );
return newContext;
} else if( scanner.errorColumn() == -1 && scanner.line() <= 0 ) { // nothing has been read yet
return new URIParsingContext( scanner.source(), scanner.line(), scanner.line(),
0, 0 + token.content().length(), scanner.codeLine() );
} else { // errorColum >= 0
return new URIParsingContext( scanner.source(), scanner.line(), scanner.line(),
scanner.errorColumn(), scanner.errorColumn() + token.content().length(),
scanner.codeLine() );
}
}
// if the start and end line are larger than 0 and the error column is -1
if( scanner.errorColumn() == -1 && lineState == 1 ) {
int newColumn = scanner.codeLine().get( scanner.codeLine().size() - 1 ).length() - 1;
return new URIParsingContext( scanner.source(), scanner.startLine(), scanner.endLine() - 1,
newColumn, newColumn + token.content().length(),
scanner.codeLine() );
} else if( scanner.errorColumn() == -1 && lineState == 2 ) {
int newColumn = scanner.codeLine().get( scanner.codeLine().size() - 1 ).length() - 1;
return new URIParsingContext( scanner.source(), scanner.startLine() - 1, scanner.endLine() - 1,
newColumn, newColumn + token.content().length(),
scanner.codeLine() );
}
return new URIParsingContext( scanner.source(), scanner.startLine(), scanner.endLine(),
scanner.errorColumn(), scanner.errorColumn() + token.content().length(),
scanner.codeLine() );
}
/**
*
* @return the current line of the parser
*/
public final int line() {
return scanner.line();
}
/**
*
* @return the startLine
*/
public final int startLine() {
return scanner.startLine();
}
/**
* sets the startLine to the current line
*/
public void setStartLine() {
scanner.setStartLine( line() );
}
/**
*
* @return the endLine
*/
public final int endLine() {
return scanner.endLine();
}
/**
* sets the endLine to the current line
*/
public void setEndLine() {
scanner.setEndLine( line() );
}
/**
*
* @return the errorColumn
*/
public final int errorColumn() {
return scanner.errorColumn();
}
/**
*
* @return the codeLine
*/
public final List< String > codeLine() {
return scanner.codeLine();
}
/**
* Eats the current token, asserting its type. Calling eat( type, errorMessage ) is equivalent to
* call subsequently tokenAssert( type, errorMessage ) and getToken().
*
* @param type The type of the token to eat.
* @param errorMessage The error message to display in case of a wrong token type.
* @throws ParserException If the token type is wrong.
* @throws IOException If the internal scanner raises one.
*/
protected final void eat( Scanner.TokenType type, String errorMessage )
throws ParserException, IOException {
assertToken( type, errorMessage );
nextToken();
}
/**
* Eats the current token, asserting its type. Calling eat( type, errorMessage ) is equivalent to
* call subsequently tokenAssert( type, errorMessage ) and getToken().
*
* @param type The type of the token to eat.
* @param errorMessage The error message to display in case of a wrong token type.
* @param scopeName the name of the current scope e.g a service name, if the scope has one
* @param scope the scope e.g inputPort
* @throws ParserException If the token type is wrong.
* @throws IOException If the internal scanner raises one.
*/
protected final void eat( Scanner.TokenType type, String errorMessage, String scopeName, String scope )
throws ParserException, IOException {
assertToken( type, errorMessage, scopeName, scope );
nextToken();
}
protected final void maybeEat( Scanner.TokenType... types )
throws ParserException, IOException {
for( Scanner.TokenType type : types ) {
if( token.is( type ) ) {
nextToken();
break;
}
}
}
protected final void eatKeyword( String keyword, String errorMessage )
throws ParserException, IOException {
assertToken( Scanner.TokenType.ID, errorMessage );
if( !token.content().equals( keyword ) ) {
throwException( errorMessage );
}
nextToken();
}
/**
* Eats the current token, asserting that it is an identifier (or an unreserved keyword). Calling
* eatIdentifier( errorMessage ) is equivalent to call subsequently assertIdentifier( errorMessage )
* and getToken().
*
* @param errorMessage The error message to throw as a {@link ParserException} in case the current
* token is not an identifier.
* @param scopeName the name of the current scope e.g a service name, if the scope has one
* @param scope the scope e.g inputPort
* @throws ParserException If the current token is not an identifier.
* @throws IOException If the internal scanner cannot read the next token.
*/
protected final void eatIdentifier( String errorMessage, String scopeName, String scope )
throws ParserException, IOException {
assertIdentifier( errorMessage, scopeName, scope );
nextToken();
}
protected final void eatIdentifier( String errorMessage )
throws ParserException, IOException {
assertIdentifier( errorMessage );
nextToken();
}
/**
* Asserts that the current token is an identifier (or an unreserved keyword).
*
* @param errorMessage the error message to throw as a {@link ParserException} if the current token
* is not an identifier.
* @throws ParserException if the current token is not an identifier.
*/
protected final void assertIdentifier( String errorMessage )
throws ParserException, IOException {
if( !token.isIdentifier() ) {
throwException( errorMessage );
}
}
/**
* Asserts that the current token is an identifier (or an unreserved keyword).
*
* @param errorMessage the error message to throw as a {@link ParserException} if the current token
* is not an identifier.
* @param scopeName the name of the current scope e.g a service name, if the scope has one
* @param scope the scope e.g inputPort
* @throws ParserException if the current token is not an identifier.
*/
protected final void assertIdentifier( String errorMessage, String scopeName, String scope )
throws ParserException, IOException {
if( !token.isIdentifier() ) {
throwExceptionWithScope( errorMessage, scopeName, scope );
}
}
/**
* Asserts the current token type.
*
* @param type The token type to assert.
* @param errorMessage The error message to display in case of a wrong token type.
* @throws ParserException If the token type is wrong.
*/
protected final void assertToken( Scanner.TokenType type, String errorMessage )
throws ParserException, IOException {
if( token.isNot( type ) ) {
throwException( errorMessage );
}
}
protected final void assertToken( Scanner.TokenType type, String errorMessage, String scopeName, String scope )
throws ParserException, IOException {
if( token.isNot( type ) ) {
throwExceptionWithScope( errorMessage, scopeName, scope );
}
}
/**
* Creates help message from the context, the token/term which was incorrect during parsing and a
* list of possible tokens/terms which can replace the current token. If the tokenContent is not
* empty, the possibleTokens is checked whether it contains keywords close in levenshteins distance
* to the tokenContent, and these words are added to the help message, such that the user can see
* correct keywords matching closely to their own token.
*
* @param context context of the code, when parsing error occured
* @param tokenContent content of the token, which was wrong or missing(if empty)
* @param possibleTokens list of possible keywords, which the user could have meant or are missing
*/
public static String createHelpMessage( ParsingContext context, String tokenContent,
List< String > possibleTokens ) {
if( possibleTokens.isEmpty() ) {
// We cannot suggest any possible terms
return "";
}
if( tokenContent.isEmpty() ) { // if the token is missing, the user is given all
// possible tokens that can be written
StringBuilder help = new StringBuilder( "You are missing a keyword. Possible inputs are:\n" )
.append( String.join( ", ", possibleTokens ) );
return help.toString();
}
// if tokenContent not empty, look for closely related correct keywords
LevenshteinDistance dist = new LevenshteinDistance();
ArrayList< String > proposedWord = new ArrayList<>();
for( String correctToken : possibleTokens ) {
if( dist.apply( tokenContent, correctToken ) <= 2 ) {
proposedWord.add( correctToken );
}
}
if( proposedWord.isEmpty() && !tokenContent.isEmpty() ) { // If none of the correct terms match the written
// token
StringBuilder help = new StringBuilder( "The term did not match possible terms. Possible inputs are:\n" )
.append( String.join( ", ", possibleTokens ) );
return help.toString();
} else { // If one or more words are really close to the token, the suggested correct token will be presented
// with the original code line, to show how to write it correctly
StringBuilder help = new StringBuilder( "Your term is similar to what would be valid input: " )
.append( String.join( ", ", proposedWord ) )
.append( ". Perhaps you meant:\n" );
// Since we are not looking at a whole scope, but a single line, we just assume the error is at the
// startLine,
// which is the first line in the enclosingCode
int columnSpace = context.startColumn() + (context.startLine() + 1 + ":").length();
help.append( context.enclosingCode().get( 0 ).substring( 0, columnSpace ) )
.append( proposedWord.get( 0 ) )
.append( context.enclosingCode().get( 0 ).substring( tokenContent.length() + columnSpace ) );
if( !context.enclosingCode().get( 0 ).endsWith( "\n" ) ) {
help.append( "\n" );
}
for( int j = 0; j < columnSpace; j++ ) {
help.append( " " );
}
help.append( '^' );
return help.toString();
}
}
/**
* Creates the help message just like createHelpMessage, taking the code lines in the scope of the
* error into consideration
*
* @param context
* @param tokenContent
* @param scope
* @return the help message
*/
public static String createHelpMessageWithScope( ParsingContext context, String tokenContent,
String scope ) {
StringBuilder help = new StringBuilder();
// Get keywords that the programmer might have errors with for the scope
List< String > possibleTerms = Keywords.getKeywordsForScope( scope );
if( tokenContent == null || tokenContent.isEmpty() ) {
// No token when the parsing error occured, so all possible keywords for scope are suggested
help.append( "A term is missing. Possible inputs are:\n" );
for( String term : possibleTerms ) {
help.append( term ).append( "\n" );
}
} else {
// Compare token to possible terms, to suggest closest possible terms as help
LevenshteinDistance dist = new LevenshteinDistance();
ArrayList< String > proposedWord = new ArrayList<>();
for( String correctTerm : possibleTerms ) {
if( dist.apply( tokenContent, correctTerm ) <= 2 ) {
proposedWord.add( correctTerm );
}
}
if( !proposedWord.isEmpty() ) {
// Possible terms were found, create help message with these terms
help.append( "\nYour term is similar to what would be valid input: " )
.append( String.join( ", ", proposedWord ) )
.append( ". Perhaps you meant:\n" );
int numberSpaces = context.startColumn() + (":" + (context.endLine() + 1)).length();
for( String line : context.enclosingCodeWithLineNumbers() ) {
if( line.contains( context.endLine() + 1 + ":" ) ) {
// For the line with error, substitute the token with the first proposed term
help.append( line.substring( 0, numberSpaces ) )
.append( proposedWord.get( 0 ) )
.append( line.substring( numberSpaces + tokenContent.length() ) );
} else {
// All other lines should just be added as they are
help.append( line );
}
if( !line.endsWith( "\n" ) ) {
// Make sure all lines have endline
help.append( "\n" );
}
}
for( int j = 0; j < numberSpaces; j++ ) {
help.append( " " );
}
help.append( '^' );
} else { // If none of the correct terms match the written token, suggest all possible terms
help = new StringBuilder( "The term did not match possible terms. Possible inputs are:\n" )
.append( String.join( ", ", possibleTerms ) );
}
}
return help.toString();
}
/**
* Shortcut to throw a correctly formed ParserException.
*
* @param mesg The message to insert in the ParserException.
* @throws ParserException Every time, as its the purpose of this method.
*/
protected final void throwException( String mesg )
throws ParserException, IOException {
CodeCheckMessage exceptionMessage;
ParsingContext context = getContext();
if( !token.content().equals( "" ) ) {
if( !mesg.equals( "" ) ) {
if( mesg.contains( "term" ) ) {
mesg += ": " + token.content() + "\n";
} else {
mesg += ". Found term: " + token.content() + "\n";
}
} else {
mesg += ". Found term: " + token.content() + "\n";
}
exceptionMessage = CodeCheckMessage.withoutHelp( context, mesg );
} else {
mesg += "\n";
// I remove 1 from the columns, because the message otherwise looks as if the error is within the
// curly bracket and not at/before the curly bracket
// example, if service does not have a name
context =
new URIParsingContext( context.source(), context.startLine(), context.endLine(),
context.startColumn() - 1, context.endColumn() - 1,
context.enclosingCode() );
exceptionMessage = CodeCheckMessage.withoutHelp( context, mesg );
}
throw new ParserException( exceptionMessage );
}
/**
* Method to find all lines of code of the scope with the name we are looking at during a parsing
* error
*
* @param name The name of the scope, e.g. the inputPort name
* @param scope The scope we are in, e.g. inputPort
*/
protected final List< String > getWholeScope( String name, String scope ) {
List< String > allLines = scanner.getAllCodeLines();
ArrayList< String > lines = new ArrayList<>();
for( int i = startLine(); i <= endLine(); i++ ) {
try {
String currentLine = allLines.get( i );
lines.add( currentLine );
} catch( IndexOutOfBoundsException e ) {
// do nothing
}
}
return lines;
}
/**
* Shortcut to throw a correctly formed ParserException.
*
* @param mesg The message to insert in the ParserException.
* @param scopeName The name of the scope, e.g. the inputPort name
* @param scope The scope we are in, e.g. an inputPort
* @throws ParserException Every time, as its the purpose of this method.
*/
protected final void throwExceptionWithScope( String mesg, String scopeName, String scope )
throws ParserException, IOException {
CodeCheckMessage exceptionMessage;
// Get a context with more accurate informatin with respect to the error
ParsingContext context = getContextDuringError();
if( !token.content().equals( "" ) ) {
// Add the token we found to the message
mesg += ". Found term: " + token.content() + "\n";
} else {
mesg += "\n";
}
String help;
List< String > extralines;
// Create the help and exception message according to the scope
switch( scope ) {
case Keywords.INPUT_PORT:
extralines = getWholeScope( scopeName, scope );
if( mesg.contains( "expected {" ) ) {
// if the starting curly bracket is missing,
// set the correct line and column for the context, so it points to the start of the port
int column = context.enclosingCode().get( 0 ).length() - 1;
context = new URIParsingContext( context.source(), context.startLine(), context.startLine(), column,
column, context.enclosingCode() );
exceptionMessage = CodeCheckMessage.withoutHelp( context, mesg );
} else if( extralines.get( extralines.size() - 1 ).contains( "}" ) ) {
// If a term is missing, it needs the column of the last curly bracket instead of where it
// originally threw the error
int columnNumber = extralines.get( extralines.size() - 1 ).lastIndexOf( "}" );
context =
new URIParsingContext( context.source(), context.startLine(), context.endLine(), columnNumber,
columnNumber + token.content().length(),
extralines );
help = createHelpMessageWithScope( context, token.content(), scope );
exceptionMessage = CodeCheckMessage.withHelp( context, mesg, help );
} else if( extralines.get( 0 ).contains( "{" ) && mesg.contains( "expected inputPort name" ) ) {
// inputport is missing name, set column to curly starting curly bracket
int columnNumber = extralines.get( 0 ).lastIndexOf( "{" );
context = new URIParsingContext( context.source(), context.startLine(), context.endLine(), columnNumber,
columnNumber, extralines );
exceptionMessage = CodeCheckMessage.withoutHelp( context, mesg );
} else if( mesg.contains( "expected :" ) ) {
// location or others is missing ':'
// point to the word after the colon
int column;
if( mesg.contains( "after Location" ) ) {
column = extralines.get( extralines.size() - 1 ).toLowerCase().indexOf( "location" ) + 8;
} else if( mesg.contains( "after Protocol" ) ) {
column = extralines.get( extralines.size() - 1 ).toLowerCase().indexOf( "protocol" ) + 8;
} else if( mesg.contains( "after Aggregates" ) ) {
column = extralines.get( extralines.size() - 1 ).toLowerCase().indexOf( "aggregates" ) + 10;
} else if( mesg.contains( "after Interfaces" ) ) {
column = extralines.get( extralines.size() - 1 ).toLowerCase().indexOf( "interfaces" ) + 10;
} else if( mesg.contains( "after Redirects" ) ) {
column = extralines.get( extralines.size() - 1 ).toLowerCase().indexOf( "redirects" ) + 9;
} else {
column = context.startColumn();
}
context = new URIParsingContext( context.source(), context.startLine(), context.endLine(),
column, column, extralines );
exceptionMessage = CodeCheckMessage.withoutHelp( context, mesg );
} else {
// in all other cases, set the endColumn to the startColumn+token
context = new URIParsingContext( context.source(), context.startLine(), context.endLine(),
context.startColumn(), context.startColumn() + token.content().length(), extralines );
help = createHelpMessageWithScope( context, token.content(), scope );
exceptionMessage = CodeCheckMessage.withHelp( context, mesg, help );
}
break;
case Keywords.EXECUTION:
extralines = getWholeScope( "", scope ); // look for line containing execution
int startColumn;
int endColumn;
if( extralines.get( 0 ).contains( "execution:" ) && extralines.get( 0 ).contains( token.content() ) ) {
// if execution: 'token' is written at the line, then set the column to the token, as this word is
// the error
startColumn = extralines.get( 0 ).indexOf( token.content() );
endColumn = startColumn + token.content().length();
} else {
// we are probably not at the correct line, so just set the columns to the original context's
// columns
startColumn = context.startColumn();
endColumn = context.endColumn();
}
context = new URIParsingContext( context.source(), context.startLine(), context.endLine(),
startColumn, endColumn, extralines );
help = createHelpMessageWithScope( context, token.content(), scope );
exceptionMessage = CodeCheckMessage.withHelp( context, mesg, help );
break;
case Keywords.SERVICE:
if( mesg.contains( "unexpected term found inside service" ) && token.content().isEmpty() ) {
// if the service does not have an ending curlybracket
extralines = getWholeScope( scopeName, scope );
int columnNumber = (!extralines.isEmpty()) ? extralines.get( extralines.size() - 1 ).length() - 1 : 0;
context =
new URIParsingContext( context.source(), context.startLine(), context.endLine(), columnNumber,
columnNumber, extralines );
help = createHelpMessageWithScope( context, token.content(), scope );
exceptionMessage = CodeCheckMessage.withoutHelp( context,
"Service " + scopeName + " does not have an ending '}'\n" );
} else if( mesg.contains( "expected {" ) && token.content().isEmpty() ) {
// service is missing starting curly bracket
extralines = getWholeScope( scopeName, scope );
// set column to last index on line with service
int column = extralines.get( 0 ).length() - 1;
context = new URIParsingContext( context.source(), context.startLine(), context.startLine(),
column, column, extralines );
exceptionMessage = CodeCheckMessage.withoutHelp( context, mesg );
} else if( mesg.contains( "expected {" ) ) {
// no need to getWholeScope when token.conten() is not empty
int column = context.enclosingCode().get( 0 ).length() - 1;
context = new URIParsingContext( context.source(), context.startLine(), context.startLine(),
column, column, context.enclosingCode() );
exceptionMessage = CodeCheckMessage.withoutHelp( context, mesg );
} else if( token.content().isEmpty() ) {
// if we have no token, then we cannot make a help message
exceptionMessage = CodeCheckMessage.withoutHelp( context, mesg );
} else {
// all other cases we make help
help = createHelpMessageWithScope( context, token.content(), scope );
exceptionMessage = CodeCheckMessage.withHelp( context, mesg, help );
}
break;
case Keywords.IMPORT:
if( mesg.contains( "expected identifier for importing target after from" ) ) {
extralines = getWholeScope( null, scope ); // find line where import is spelled wrong
String[] tempSplit = extralines.get( 0 ).split( " ", 0 );
if( tempSplit.length > 1 ) {
int columnNumber = (tempSplit[ 0 ] + " " + tempSplit[ 1 ] + " ").length();
// set the columns to match the wrongly spelled import
context =
new URIParsingContext( context.source(), context.startLine(), context.endLine(), columnNumber,
columnNumber + tempSplit[ 2 ].length(), extralines );
String importHelp = createHelpMessage( context, tempSplit[ 2 ], List.of( "from" ) );
exceptionMessage = CodeCheckMessage.withHelp( context, mesg, importHelp );
break;
}
}
exceptionMessage = CodeCheckMessage.withoutHelp( context, mesg );
break;
case Keywords.OUTER:
help = createHelpMessageWithScope( context, token.content(), scope );
exceptionMessage = CodeCheckMessage.withHelp( context, mesg, help );
break;
case Keywords.INTERFACE:
if( token.content().isEmpty() ) {
if( mesg.contains( "expected {" ) ) { // interface is missing starting curly bracket
extralines = getWholeScope( scopeName, scope );
int column = extralines.get( 0 ).length() - 1;
context = new URIParsingContext( context.source(), context.startLine(), context.startLine(),
column, column, List.of( extralines.get( 0 ) ) );
exceptionMessage = CodeCheckMessage.withoutHelp( context, mesg );
} else if( mesg.contains( "expected }" ) ) { // interface is missing ending curly bracket
extralines = getWholeScope( scopeName, scope );
int column = extralines.get( extralines.size() - 1 ).length() - 1;
context = new URIParsingContext( context.source(), context.startLine(), context.endLine(),
column, column, extralines );
exceptionMessage = CodeCheckMessage.withoutHelp( context, mesg );
} else {
exceptionMessage = CodeCheckMessage.withoutHelp( context, mesg );
}
} else if( mesg.contains( "expected {" ) ) { // missing starting curly bracket and token.content() is not
// empty
extralines = getWholeScope( scopeName, scope );
int column = extralines.get( 0 ).length() - 1;
context = new URIParsingContext( context.source(), context.startLine(), context.startLine(),
column, column, List.of( extralines.get( 0 ) ) );
exceptionMessage = CodeCheckMessage.withoutHelp( context, mesg );
} else if( mesg.contains( "expected }" ) ) { // missing ending curly bracket and token.content() is not
// empty
// only have the last line of enclosingCode in the context
context = new URIParsingContext( context.source(), context.endLine(), context.endLine(),
context.startColumn(), context.endColumn(),
List.of( context.enclosingCode().get( context.enclosingCode().size() - 1 ) ) );
help = createHelpMessageWithScope( context, token.content(), scope );
exceptionMessage = CodeCheckMessage.withHelp( context, mesg, help );
} else {
// otherwise
help = createHelpMessageWithScope( context, token.content(), scope );
exceptionMessage = CodeCheckMessage.withHelp( context, mesg, help );
}
break;
case Keywords.MAIN:
if( mesg.contains( "expected basic statement" ) ) { // if main is empty
// set column to ending curly bracket
int column = context.enclosingCode().get( context.enclosingCode().size() - 1 ).lastIndexOf( "}" );
context = new URIParsingContext( context.source(), context.startLine(), context.endLine(),
column, column, context.enclosingCode() );
exceptionMessage = CodeCheckMessage.withoutHelp( context, mesg );
} else {
context = new URIParsingContext( context.source(), context.startLine(), context.endLine(),
context.endColumn(), context.endColumn(), context.enclosingCode() );
exceptionMessage = CodeCheckMessage.withoutHelp( context, mesg );
}
break;
case Keywords.TYPE:
if( mesg.contains( "expected type for node" ) ) {
// if nothing is written after the colon, when defining a type
extralines = getWholeScope( scopeName, scope );
String correctLine = "";
int i = context.startLine();
for( String string : extralines ) {
if( string.matches( "^.+:\\s*$" ) ) { // finds the line with nothing after the colon
// TODO: should match even when a comment is written after
// "^.+:(\\s*|\\s+//.*|\\s+/\\*.*)$" (does not work for some reason)
correctLine = string;
int column = correctLine.indexOf( ':' );
context =
new URIParsingContext( context.source(), i, i, column, column, List.of( correctLine ) );
break;
}
i += 1;
}
if( correctLine.equals( "" ) ) {
context = new URIParsingContext( context.source(), context.startLine(), context.endLine(),
context.startColumn(), context.endColumn(), extralines );
}
exceptionMessage = CodeCheckMessage.withoutHelp( context, mesg );
} else {
// any other error for type scope
extralines = getWholeScope( scopeName, scope );
context = new URIParsingContext( context.source(), context.startLine(), context.startLine(),
context.startColumn(), context.endColumn(), extralines );
exceptionMessage = CodeCheckMessage.withoutHelp( context, mesg );
}
break;
default:
exceptionMessage = CodeCheckMessage.withoutHelp( context, mesg );
}
throw new ParserException( exceptionMessage );
}
/**
* Shortcut to throw a correctly formed ParserException, getting the message from an existing
* exception.
*
* @param exception The exception to get the message from.
* @throws ParserException Every time, as its the purpose of this method.
*/
protected final void throwException( Exception exception )
throws ParserException {
ParsingContext context = getContext();
CodeCheckMessage exceptionMessage = CodeCheckMessage.withoutHelp( context, exception.getMessage() );
throw new ParserException( exceptionMessage );
}
}