Newer
Older
/*
* This file is part of ADQLLibrary.
*
* ADQLLibrary is free software: you can redistribute it and/or modify
* it under the terms of the GNU Lesser General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* ADQLLibrary 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 Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public License
* along with ADQLLibrary. If not, see <http://www.gnu.org/licenses/>.
*
* Copyright 2012-2019 - UDS/Centre de Données astronomiques de Strasbourg (CDS),
* Astronomisches Rechen Institute (ARI)
* This JavaCC file implements the BNF definition of ADQL v2.0
* (IVOA Recommendation 30 Oct 2008 - http://www.ivoa.net/Documents/cover/ADQL-20081030.html).
*
* To generate the parser with this file use JavaCC. This .jj file has been
* successfully tested with JavaCC 6.0.
*
* The generated parser checks the syntax of the given ADQL query and generates
* an object representation but no coherence with any database is done.
* If the syntax is not conform to the ADQL definition an error message is
* printed else it will be the message "Correct syntax".
* Author: Grégory Mantelet (CDS;ARI)
* Version: 1.5 (03/2019)
*/
/* ########### */
/* # OPTIONS # */
/* ########### */
options {
STATIC = false;
IGNORE_CASE = true;
DEBUG_PARSER = true;
}
/* ########## */
/* # PARSER # */
/* ########## */
package adql.parser;
import java.util.Stack;
import java.util.Vector;
import java.util.ArrayList;
import java.util.Collection;
import java.io.FileReader;
import java.io.IOException;
import adql.db.exception.UnresolvedIdentifiersException;
import adql.parser.IdentifierItems.IdentifierItem;
import adql.parser.ADQLQueryFactory.JoinType;
import adql.query.*;
import adql.query.from.*;
import adql.query.constraint.*;
import adql.query.operand.*;
import adql.query.operand.function.*;
import adql.query.operand.function.geometry.*;
import adql.query.operand.function.geometry.GeometryFunction.GeometryValue;
import adql.translator.PostgreSQLTranslator;
import adql.translator.TranslationException;
/**
* Parses an ADQL query thanks to the {@link ADQLParser#Query()} function.
* <p>
* This parser is able, thanks to a {@link QueryChecker} object, to check each
* {@link ADQLQuery} just after its generation. It could be used to check the
* consistency between the ADQL query to parse and the "database" on which the
* query must be executed. By default, there is no {@link QueryChecker}. Thus
* you must extend {@link QueryChecker} to check semantically all generated
* ADQLQuery objects.
* </p>
* <p>
* To create an object representation of the given ADQL query, this parser uses
* a {@link ADQLQueryFactory} object. So if you want customize some object
* (ie. CONTAINS) of this representation you just have to extend the
* corresponding default object (ie. ContainsFunction) and to extend the
* corresponding function of {@link ADQLQueryFactory}
* (ie. createContains(...)).
* </p>
Grégory Mantelet
committed
*
* <p>Here are the key functions to use:</p>
* <ul>
* <li>{@link #parseQuery(java.lang.String)} (or any of its alternatives)
* to parse an input ADQL query String and get its corresponding ADQL tree
* </li>
* <li>{@link #tryQuickFix(java.lang.String)} to try fixing the most common
* issues with ADQL queries (e.g. Unicode confusable characters,
* unescaped ADQL identifiers, SQL reserved keywords, ...)</li>
* </ul>
* <p><b><u>WARNING:</u>
* To modify this class it's strongly encouraged to modify the .jj file in the
* section between <i>PARSER_BEGIN</i> and <i>PARSER_END</i> and to re-compile
* it with JavaCC.
* </b></p>
*
* @see QueryChecker
* @see ADQLQueryFactory
*
* @author Grégory Mantelet (CDS;ARI)
* @version 1.5 (03/2019)
public class ADQLParser {
/** Tools to build the object representation of the ADQL query. */
private ADQLQueryFactory queryFactory = new ADQLQueryFactory();
/** The stack of queries (because there may be some sub-queries). */
private Stack<ADQLQuery> stackQuery = new Stack<ADQLQuery>();
/** The object representation of the ADQL query to parse.
* (ONLY USED DURING THE PARSING, else it is always <i>null</i>). */
private ADQLQuery query = null;
/** Checks each {@link ADQLQuery} (sub-query or not) just after their
* generation. */
private QueryChecker queryChecker = null;
/** The first token of a table/column name. This token is extracted by
* {@link #Identifier()}. */
private Token currentIdentifierToken = null;
/**
* Builds an ADQL parser without a query to parse.
*/
this(new java.io.ByteArrayInputStream("".getBytes()));
setDebug(false);
* Builds an ADQL parser without a query to parse but with a
* {@link QueryChecker} and a {@link ADQLQueryFactory}.
* @param checker The object to use to check each {@link ADQLQuery}.
* @param factory The object to use to build an object representation of
* the given ADQL query.
public ADQLParser(QueryChecker checker, ADQLQueryFactory factory) {
this();
queryChecker = checker;
if (factory != null)
queryFactory = factory;
}
/**
* Builds an ADQL parser without a query to parse but with a
* {@link QueryChecker}.
* @param checker The object to use to check each {@link ADQLQuery}.
public ADQLParser(QueryChecker checker) {
this(checker, null);
}
/**
* Builds an ADQL parser without a query to parse but with a
* {@link ADQLQueryFactory}.
* @param factory The object to use to build an object representation of
* the given ADQL query.
public ADQLParser(ADQLQueryFactory factory) {
this((QueryChecker)null, factory);
}
/**
* Builds a parser with a stream containing the query to parse.
*
* @param stream The stream in which the ADQL query to parse is given.
* @param checker The object to use to check each {@link ADQLQuery}.
* @param factory The object to use to build an object representation of
* the given ADQL query.
public ADQLParser(java.io.InputStream stream, QueryChecker checker, ADQLQueryFactory factory) {
gmantele
committed
setDebug(false);
setDebug(false);
queryChecker = checker;
if (factory != null)
queryFactory = factory;
}
/**
* Builds a parser with a stream containing the query to parse.
*
* @param stream The stream in which the ADQL query to parse is given.
* @param checker The object to use to check each {@link ADQLQuery}.
public ADQLParser(java.io.InputStream stream, QueryChecker checker) {
this(stream, checker, null);
}
/**
* Builds a parser with a stream containing the query to parse.
*
* @param stream The stream in which the ADQL query to parse is given.
* @param factory The object to use to build an object representation of
* the given ADQL query.
public ADQLParser(java.io.InputStream stream, ADQLQueryFactory factory) {
this(stream, (QueryChecker)null, factory);
}
/**
* Builds a parser with a stream containing the query to parse.
*
* @param stream The stream in which the ADQL query to parse is given.
* @param encoding The supplied encoding.
* @param checker The object to use to check each {@link ADQLQuery}.
* @param factory The object to use to build an object representation
* of the given ADQL query.
public ADQLParser(java.io.InputStream stream, String encoding, QueryChecker checker, ADQLQueryFactory factory) {
this(stream, encoding);
gmantele
committed
setDebug(false);
queryChecker = checker;
if (factory != null)
queryFactory = factory;
}
/**
* Builds a parser with a stream containing the query to parse.
*
* @param stream The stream in which the ADQL query to parse is given.
* @param encoding The supplied encoding.
* @param checker The object to use to check each {@link ADQLQuery}.
public ADQLParser(java.io.InputStream stream, String encoding, QueryChecker checker) {
this(stream, encoding, checker, null);
}
/**
* Builds a parser with a stream containing the query to parse.
*
* @param stream The stream in which the ADQL query to parse is given.
* @param encoding The supplied encoding.
* @param factory The object to use to build an object representation
* of the given ADQL query.
public ADQLParser(java.io.InputStream stream, String encoding, ADQLQueryFactory factory) {
this(stream, encoding, null, factory);
}
/**
* Builds a parser with a reader containing the query to parse.
*
* @param reader The reader in which the ADQL query to parse is given.
* @param checker The object to use to check each {@link ADQLQuery}.
* @param factory The object to use to build an object representation
* of the given ADQL query.
public ADQLParser(java.io.Reader reader, QueryChecker checker, ADQLQueryFactory factory) {
gmantele
committed
setDebug(false);
setDebug(false);
queryChecker = checker;
if (factory != null)
queryFactory = factory;
}
/**
* Builds a parser with a reader containing the query to parse.
*
* @param reader The reader in which the ADQL query to parse is given.
* @param checker The object to use to check each {@link ADQLQuery}.
public ADQLParser(java.io.Reader reader, QueryChecker checker) {
this(reader, checker, null);
}
/**
* Builds a parser with a reader containing the query to parse.
*
* @param reader The reader in which the ADQL query to parse is given.
* @param factory The object to use to build an object representation
* of the given ADQL query.
public ADQLParser(java.io.Reader reader, ADQLQueryFactory factory) {
this(reader, null, factory);
}
/**
* Builds a parser with another token manager.
*
* @param tm The manager which associates a token to a numeric code.
* @param checker The object to use to check each {@link ADQLQuery }.
* @param factory The object to use to build an object representation
* of the given ADQL query.
public ADQLParser(ADQLParserTokenManager tm, QueryChecker checker, ADQLQueryFactory factory) {
gmantele
committed
setDebug(false);
setDebug(false);
queryChecker = checker;
if (factory != null)
queryFactory = factory;
}
/**
* Builds a parser with another token manager.
*
* @param tm The manager which associates a token to a numeric code.
* @param checker The object to use to check each {@link ADQLQuery}.
public ADQLParser(ADQLParserTokenManager tm, QueryChecker checker) {
this(tm, checker, null);
}
/**
* Builds a parser with another token manager.
*
* @param tm The manager which associates a token to a numeric code.
* @param factory The object to use to build an object representation of
* the given ADQL query.
public ADQLParser(ADQLParserTokenManager tm, ADQLQueryFactory factory) {
this(tm, null, factory);
}
Grégory Mantelet
committed
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
/* ADDITIONAL GETTERS & SETTERS */
public final void setDebug(boolean debug){
if (debug) enable_tracing();
else disable_tracing();
}
public final QueryChecker getQueryChecker(){
return queryChecker;
}
public final void setQueryChecker(QueryChecker checker){
queryChecker = checker;
}
public final ADQLQueryFactory getQueryFactory(){
return queryFactory;
}
public final void setQueryFactory(ADQLQueryFactory factory){
queryFactory = (factory!=null)?factory:(new ADQLQueryFactory());
}
/* EXCEPTION HELPER FUNCTION */
private final ParseException generateParseException(Exception ex){
if (!(ex instanceof ParseException)){
ParseException pex = new ParseException("["+ex.getClass().getName()+"] "+ex.getMessage());
pex.setStackTrace(ex.getStackTrace());
return pex;
}else
return (ParseException)ex;
}
/* QUERY PARSING FUNCTIONS */
/**
* Tell whether the given string is a valid ADQL regular identifier.
*
* <p>
* According to the ADQL-2.0's BNF, a regular identifier (i.e. not delimited
* ; not between double quotes) must be a letter followed by a letter, digit
* or underscore. So, the following regular expression:
* </p>
* <pre>[a-zA-Z]+[a-zA-Z0-9_]*</pre>
*
* <p>This is what this function tests on the given string.</p>
*
* @param idCandidate The string to test.
*
* @return <code>true</code> if the given string is a valid regular
* identifier,
* <code>false</code> otherwise.
*
* @see #testRegularIdentifier(adql.parser.Token)
*
* @since 1.5
*/
public final boolean isRegularIdentifier(final String idCandidate) {
return idCandidate.matches("[a-zA-Z]+[a-zA-Z0-9_]*");
}
/**
* Test the given token as an ADQL's regular identifier.
*
* <p>
* This function uses {@link #isRegularIdentifier(java.lang.String)} to
* test the given token's image. If the test fails, a
* {@link adql.parser.ParseException} is thrown.
* </p>
*
* @param token The token to test.
*
* @throws ParseException If the given token is not a valid ADQL regular
* identifier.
*
* @see #isRegularIdentifier(java.lang.String)
*
* @since 1.5
*/
public final void testRegularIdentifier(final Token token) throws ParseException {
if (!isRegularIdentifier(token.image))
throw new ParseException("Invalid ADQL regular identifier: \""+token.image+"\"! If it aims to be a column/table name/alias, you should write it between double quotes.", new TextPosition(token));
}
* Parses the query given at the creation of this parser or in the
* <i>ReInit</i> functions.
* @return The object representation of the given ADQL query.
*
* @throws ParseException If there is at least one syntactic error.
*
* @see ADQLParser#Query()
*/
public final ADQLQuery parseQuery() throws ParseException {
stackQuery.clear();
query = null;
try {
return Query();
}catch(TokenMgrError tme) {
Grégory Mantelet
committed
throw new ParseException(tme);
}
/**
* Parses the query given in parameter.
*
* @param q The ADQL query to parse.
*
* @return The object representation of the given ADQL query.
*
* @throws ParseException If there is at least one syntactic error.
*
* @see ADQLParser#ReInit(java.io.InputStream)
* @see ADQLParser#setDebug(boolean)
* @see ADQLParser#Query()
*/
public final ADQLQuery parseQuery(String q) throws ParseException {
stackQuery.clear();
query = null;
ReInit(new java.io.ByteArrayInputStream(q.getBytes()));
try {
return Query();
}catch(TokenMgrError tme) {
Grégory Mantelet
committed
throw new ParseException(tme);
}
/**
* Parses the query contained in the stream given in parameter.
*
* @param stream The stream which contains the ADQL query to parse.
*
* @return The object representation of the given ADQL query.
*
* @throws ParseException If there is at least one syntactic error.
*
* @see ADQLParser#ReInit(java.io.InputStream)
* @see ADQLParser#setDebug(boolean)
* @see ADQLParser#Query()
*/
public final ADQLQuery parseQuery(java.io.InputStream stream) throws ParseException {
stackQuery.clear();
query = null;
ReInit(stream);
try {
return Query();
}catch(TokenMgrError tme) {
Grégory Mantelet
committed
throw new ParseException(tme);
Grégory Mantelet
committed
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
/* CORRECTION SUGGESTION */
/**
* Try fixing tokens/terms of the input ADQL query.
*
* <p>
* <b>This function does not try to fix syntactical or semantical errors.</b>
* It just try to fix the most common issues in ADQL queries, such as:
* </p>
* <ul>
* <li>some Unicode characters confusable with ASCII characters (like a
* space, a dash, ...) ; this function replace them by their ASCII
* alternative,</li>
* <li>any of the following are double quoted:
* <ul>
* <li>non regular ADQL identifiers
* (e.g. <code>_RAJ2000</code>),</li>
* <li>ADQL function names used as identifiers
* (e.g. <code>distance</code>)</li>
* <li>and SQL reserved keywords
* (e.g. <code>public</code>).</li>
* </ul>
* </li>
* </ul>
*
* <p><i><b>Note 1:</b>
* The given stream is NOT closed by this function even if the EOF is
* reached. It is the responsibility of the caller to close it.
* </i></p>
*
* <p><i><b>Note 2:</b>
* This function does not use any instance variable of this parser
* (especially the InputStream or Reader provided at initialisation or
* ReInit).
* </i></p>
*
* @param input Stream containing the input ADQL query to fix.
*
* @return The suggested correction of the input ADQL query.
*
* @throws java.io.IOException If there is any error while reading from the
* given input stream.
* @throws ParseException If any unrecognised character is encountered,
* or if anything else prevented the tokenization
* of some characters/words/terms.
*
* @see #tryQuickFix(java.lang.String)
*
* @since 1.5
*/
public final String tryQuickFix(final java.io.InputStream input) throws java.io.IOException, ParseException {
// Fetch everything into a single string:
StringBuffer buf = new StringBuffer();
byte[] cBuf = new byte[1024];
int nbChar;
while((nbChar = input.read(cBuf)) > -1){
buf.append(new String(cBuf, 0, nbChar));
}
// Convert the buffer into a String and now try to fix it:
return tryQuickFix(buf.toString());
Grégory Mantelet
committed
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
/**
* Try fixing tokens/terms of the given ADQL query.
*
* <p>
* <b>This function does not try to fix syntactical or semantical errors.</b>
* It just try to fix the most common issues in ADQL queries, such as:
* </p>
* <ul>
* <li>some Unicode characters confusable with ASCII characters (like a
* space, a dash, ...) ; this function replace them by their ASCII
* alternative,</li>
* <li>any of the following are double quoted:
* <ul>
* <li>non regular ADQL identifiers
* (e.g. <code>_RAJ2000</code>),</li>
* <li>ADQL function names used as identifiers
* (e.g. <code>distance</code>)</li>
* <li>and SQL reserved keywords
* (e.g. <code>public</code>).</li>
* </ul>
* </li>
* </ul>
*
* <p><i><b>Note:</b>
* This function does not use any instance variable of this parser
* (especially the InputStream or Reader provided at initialisation or
* ReInit).
* </i></p>
*
* @param adqlQuery The input ADQL query to fix.
*
* @return The suggested correction of the given ADQL query.
*
* @throws ParseException If any unrecognised character is encountered,
* or if anything else prevented the tokenization
* of some characters/words/terms.
*
* @since 1.5
*/
public String tryQuickFix(String adqlQuery) throws ParseException {
StringBuffer suggestedQuery = new StringBuffer();
// 1. Replace all Unicode confusable characters:
adqlQuery = replaceUnicodeConfusables(adqlQuery);
/* 1.bis. Normalise new lines and tabulations
* (to simplify the column counting): */
adqlQuery = adqlQuery.replaceAll("(\r\n|\r|\n)", System.getProperty("line.separator")).replaceAll("\t", " ");
// 2. Analyse the query token by token:
ADQLParserTokenManager parser = new ADQLParserTokenManager(new SimpleCharStream(new java.io.ByteArrayInputStream(adqlQuery.getBytes())));
Grégory Mantelet
committed
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
final String[] lines = adqlQuery.split(System.getProperty("line.separator"));
try{
String suggestedToken;
int lastLine = 1, lastCol = 1;
Token token = null, nextToken = parser.getNextToken();
// for all tokens until the EOF or EOQ:
do{
// get the next token:
token = nextToken;
nextToken = (isEnd(token) ? null : parser.getNextToken());
// 3. Double quote any suspect token:
if (mustEscape(token, nextToken)){
suggestedToken = "\"" + token.image + "\"";
}else
suggestedToken = token.image;
/* 4. Append all space characters (and comments) before the
* token: */
/* same line, just get the space characters between the last
* token and the one to append: */
if (lastLine == token.beginLine){
suggestedQuery.append(lines[lastLine - 1].substring(lastCol - 1, token.beginColumn - (isEnd(token) ? 0 : 1)));
lastCol = token.endColumn + 1;
}
// not the same line...
else{
/* append all remaining space characters until the position
* of the token to append: */
do{
suggestedQuery.append(lines[lastLine - 1].substring(lastCol - 1)).append('\n');
lastLine++;
lastCol = 1;
}while(lastLine < token.beginLine);
/* if there are still space characters before the token,
* append them as well: */
if (lastCol < token.beginColumn)
suggestedQuery.append(lines[lastLine - 1].substring(lastCol - 1, token.beginColumn - 1));
// finally, set the correct column position:
lastCol = token.endColumn + 1;
}
// 5. Append the suggested token:
suggestedQuery.append(suggestedToken);
}while(!isEnd(token));
}catch(TokenMgrError err){
// wrap such errors and propagate them:
throw new ParseException(err);
}
return suggestedQuery.toString();
Grégory Mantelet
committed
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
/**
* All of the most common Unicode confusable characters and their
* ASCII/UTF-8 alternative.
*
* <p>
* Keys of this map represent the ASCII character while the values are the
* regular expression for all possible Unicode alternatives.
* </p>
*
* <p><i><b>Note:</b>
* All of them have been listed using
* <a href="https://unicode.org/cldr/utility/confusables.jsp">Unicode Utilities: Confusables</a>.
* </i></p>
*
* @since 1.5
*/
protected final static java.util.Map<String, String> REGEX_UNICODE_CONFUSABLES = new java.util.HashMap<String, String>(10);
/** Regular expression matching all Unicode alternatives for <code>-</code>.
* @since 1.5 */
protected final static String REGEX_DASH = "[\u002D\u02D7\u06D4\u2010\u2011\u2012\u2013\u2043\u2212\u2796\u2CBA\uFE58\u2014\u2015\u207B\u208B\u0096\u058A\uFE63\uFF0D]";
/** Regular expression matching all Unicode alternatives for <code>_</code>.
* @since 1.5 */
protected final static String REGEX_UNDERSCORE = "[\u005F\u07FA\uFE4D\uFE4E\uFE4F]";
/** Regular expression matching all Unicode alternatives for <code>'</code>.
* @since 1.5 */
protected final static String REGEX_QUOTE = "[\u0027\u0060\u00B4\u02B9\u02BB\u02BC\u02BD\u02BE\u02C8\u02CA\u02CB\u02F4\u0374\u0384\u055A\u055D\u05D9\u05F3\u07F4\u07F5\u144A\u16CC\u1FBD\u1FBF\u1FEF\u1FFD\u1FFE\u2018\u2019\u201B\u2032\u2035\uA78C\uFF07\uFF40]";
/** Regular expression matching all Unicode alternatives for <code>"</code>.
* @since 1.5 */
protected final static String REGEX_DOUBLE_QUOTE = "[\u02BA\u02DD\u02EE\u02F6\u05F2\u05F4\u1CD3\u201C\u201D\u201F\u2033\u2036\u3003\uFF02]";
/** Regular expression matching all Unicode alternatives for <code>.</code>.
* @since 1.5 */
protected final static String REGEX_STOP = "[\u002E\u0660\u06F0\u0701\u0702\u2024\uA4F8\uA60E]";
/** Regular expression matching all Unicode alternatives for <code>+</code>.
* @since 1.5 */
protected final static String REGEX_PLUS = "[\u002B\u16ED\u2795]";
/** Regular expression matching all Unicode alternatives for <code> </code>.
* @since 1.5 */
protected final static String REGEX_SPACE = "[\u0020\u00A0\u1680\u2000\u2001\u2002\u2003\u2004\u2005\u2006\u2007\u2008\u2009\u200A\u202F\u205F]";
/** Regular expression matching all Unicode alternatives for <code><</code>.
* @since 1.5 */
protected final static String REGEX_LESS_THAN = "[\u003C\u02C2\u1438\u16B2\u2039\u276E]";
/** Regular expression matching all Unicode alternatives for <code>></code>.
* @since 1.5 */
protected final static String REGEX_GREATER_THAN = "[\u003E\u02C3\u1433\u203A\u276F]";
/** Regular expression matching all Unicode alternatives for <code>=</code>.
* @since 1.5 */
protected final static String REGEX_EQUAL = "[\u003D\u1400\u2E40\u30A0\uA4FF]";
static {
REGEX_UNICODE_CONFUSABLES.put("-", REGEX_DASH);
REGEX_UNICODE_CONFUSABLES.put("_", REGEX_UNDERSCORE);
REGEX_UNICODE_CONFUSABLES.put("'", REGEX_QUOTE);
REGEX_UNICODE_CONFUSABLES.put("\"", REGEX_DOUBLE_QUOTE);
REGEX_UNICODE_CONFUSABLES.put(".", REGEX_STOP);
REGEX_UNICODE_CONFUSABLES.put("+", REGEX_PLUS);
REGEX_UNICODE_CONFUSABLES.put(" ", REGEX_SPACE);
REGEX_UNICODE_CONFUSABLES.put("<", REGEX_LESS_THAN);
REGEX_UNICODE_CONFUSABLES.put(">", REGEX_GREATER_THAN);
REGEX_UNICODE_CONFUSABLES.put("=", REGEX_EQUAL);
Grégory Mantelet
committed
/**
* Replace all Unicode characters that can be confused with other ASCI/UTF-8
* characters (e.g. different spaces, dashes, ...) in their ASCII version.
*
* @param adqlQuery The ADQL query string in which Unicode confusable
* characters must be replaced.
*
* @return The same query without the most common Unicode confusable
* characters.
*
* @since 1.5
*/
protected String replaceUnicodeConfusables(final String adqlQuery){
String newAdqlQuery = adqlQuery;
for(java.util.Map.Entry<String, String> confusable : REGEX_UNICODE_CONFUSABLES.entrySet())
newAdqlQuery = newAdqlQuery.replaceAll(confusable.getValue(), confusable.getKey());
return newAdqlQuery;
Grégory Mantelet
committed
/**
* Tell whether the given token represents the end of an ADQL query.
*
* @param token Token to analyze.
*
* @return <code>true</code> if the given token represents a query end,
* <code>false</code> otherwise.
*
* @since 1.5
*/
protected boolean isEnd(final Token token){
return token.kind == ADQLParserConstants.EOF || token.kind == ADQLParserConstants.EOQ;
Grégory Mantelet
committed
/**
* Tell whether the given token must be double quoted.
*
* <p>
* This function considers all the following as terms to double quote:
* </p>
* <ul>
* <li>SQL reserved keywords</li>,
* <li>unrecognised regular identifiers (e.g. neither a delimited nor a
* valid ADQL regular identifier)</li>
* <li>and ADQL function name without a parameters list.</li>
* </ul>
*
* @param token The token to analyze.
* @param nextToken The following token. (useful to detect the start of a
* function's parameters list)
*
* @return <code>true</code> if the given token must be double quoted,
* <code>false</code> to keep it as provided.
*
* @since 1.5
*/
protected boolean mustEscape(final Token token, final Token nextToken){
switch(token.kind){
case ADQLParserConstants.SQL_RESERVED_WORD:
Grégory Mantelet
committed
return true;
case ADQLParserConstants.REGULAR_IDENTIFIER_CANDIDATE:
Grégory Mantelet
committed
return !isRegularIdentifier(token.image);
default:
return isFunctionName(token) && (nextToken == null || nextToken.kind != ADQLParserConstants.LEFT_PAR);
Grégory Mantelet
committed
}
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
/**
* Tell whether the given token matches to an ADQL function name.
*
* @param token The token to analyze.
*
* @return <code>true</code> if the given token is an ADQL function name,
* <code>false</code> otherwise.
*
* @since 1.5
*/
protected boolean isFunctionName(final Token token){
switch(token.kind){
case ADQLParserConstants.COUNT:
case ADQLParserConstants.EXISTS:
case ADQLParserConstants.AVG:
case ADQLParserConstants.MAX:
case ADQLParserConstants.MIN:
case ADQLParserConstants.SUM:
case ADQLParserConstants.BOX:
case ADQLParserConstants.CENTROID:
case ADQLParserConstants.CIRCLE:
case ADQLParserConstants.POINT:
case ADQLParserConstants.POLYGON:
case ADQLParserConstants.REGION:
case ADQLParserConstants.CONTAINS:
case ADQLParserConstants.INTERSECTS:
case ADQLParserConstants.AREA:
case ADQLParserConstants.COORD1:
case ADQLParserConstants.COORD2:
case ADQLParserConstants.COORDSYS:
case ADQLParserConstants.DISTANCE:
case ADQLParserConstants.ABS:
case ADQLParserConstants.CEILING:
case ADQLParserConstants.DEGREES:
case ADQLParserConstants.EXP:
case ADQLParserConstants.FLOOR:
case ADQLParserConstants.LOG:
case ADQLParserConstants.LOG10:
case ADQLParserConstants.MOD:
case ADQLParserConstants.PI:
case ADQLParserConstants.POWER:
case ADQLParserConstants.RADIANS:
case ADQLParserConstants.RAND:
case ADQLParserConstants.ROUND:
case ADQLParserConstants.SQRT:
case ADQLParserConstants.TRUNCATE:
case ADQLParserConstants.ACOS:
case ADQLParserConstants.ASIN:
case ADQLParserConstants.ATAN:
case ADQLParserConstants.ATAN2:
case ADQLParserConstants.COS:
case ADQLParserConstants.COT:
case ADQLParserConstants.SIN:
case ADQLParserConstants.TAN:
case ADQLParserConstants.USING:
return true;
default:
return false;
}
}
/* MAIN PROGRAM */
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
/**
* Gets the specified ADQL query and parses the given ADQL query. The SQL
* translation is then printed if the syntax is correct.
*
* <p>
* <b>ONLY the syntax is checked: the query is NOT EXECUTED !</b>
* </p>
*
* @param args
* @throws Exception
*/
public static final void main(String[] args) throws Exception {
final String USAGE = "Usage:\n adqlParser.jar [-d] [-v] [-e] [-a|-s] [-f] [<FILE>|<URL>]\n\nNOTE: If no file or URL is given, the ADQL query is expected in the standard\n input. This query must end with a ';' or <Ctrl+D>!\n\nParameters:\n -v or --verbose : Print the main steps of the parsing\n -d or --debug : Print stack traces when a grave error occurs\n -e or --explain : Explain the ADQL parsing (or Expand the parsing tree)\n -a or --adql : Display the understood ADQL query\n -s or --sql : Ask the SQL translation of the given ADQL query\n (SQL compatible with PostgreSQL)\n -f or --try-fix : Try fixing the most common ADQL query issues before\n attempting to parse the query.\n\nReturn:\n By default: nothing if the query is correct. Otherwise a message explaining\n why the query is not correct is displayed.\n With the -s option, the SQL translation of the given ADQL query will be\n returned.\n With the -a option, the ADQL query is returned as it has been understood.\n\nExit status:\n 0 OK !\n 1 Parameter error (missing or incorrect parameter)\n 2 File error (incorrect file/url, reading error, ...)\n 3 Parsing error (syntactic or semantic error)\n 4 Translation error (a problem has occurred during the translation of the\n given ADQL query in SQL).";
ADQLParser parser;
final String urlRegex = "^(https?|ftp|file)://[-a-zA-Z0-9+&@#/%?=~_|!:,.;]*[-a-zA-Z0-9+&@#/%=~_|]";
String file = null, metaFile = null;
short mode = -1;
boolean verbose=false, debug=false, explain=false, tryFix=false;
// Parameters reading:
for(int i=0; i<args.length; i++){
if (args[i].equalsIgnoreCase("-d") || args[i].equalsIgnoreCase("--debug"))
debug = true;
else if (args[i].equalsIgnoreCase("-v") || args[i].equalsIgnoreCase("--verbose"))
verbose = true;
else if (args[i].equalsIgnoreCase("-e") || args[i].equalsIgnoreCase("--explain"))
explain = true;
else if (args[i].equalsIgnoreCase("-a") || args[i].equalsIgnoreCase("--adql")){
if (mode != -1){
System.err.println("((!)) Too much parameter: you must choose between -s, -c, -a or nothing ((!))\n"+USAGE);
System.exit(1);
}else
mode = 1;
}else if (args[i].equalsIgnoreCase("-s") || args[i].equalsIgnoreCase("--sql")){
if (mode != -1){
System.err.println("((!)) Too much parameter: you must choose between -s, -c, -a or nothing ((!))\n"+USAGE);
System.exit(1);
}else
mode = 2;
}else if (args[i].equalsIgnoreCase("-f") || args[i].equalsIgnoreCase("--try-fix"))
tryFix = true;
else if (args[i].equalsIgnoreCase("-h") || args[i].equalsIgnoreCase("--help")){
System.out.println(USAGE);
System.exit(0);
}else if (args[i].startsWith("-")){
System.err.println("((!)) Unknown parameter: \""+args[i]+"\" ((!))\u005cn"+USAGE);
System.exit(1);
}else
file = args[i].trim();
}
try{
// Try fixing the query, if asked:
if (tryFix) {
if (verbose)
System.out.println("((i)) Trying to automatically fix the query...");
String query;
java.io.InputStream in = null;
try {
// get the input stream...
if (file == null || file.length() == 0)
in = System.in;
else if (file.matches(urlRegex))
in = (new java.net.URL(file)).openStream();
else
in = new java.io.FileInputStream(file);
// ...and try fixing the query:
query = (new ADQLParser()).tryQuickFix(in);
} finally {
// close the stream (if opened):
if (in != null)
in.close();
in = null;
}
if (verbose)
System.out.println("((i)) SUGGESTED QUERY:\n" + query);
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
// Initialise the parser with this fixed query:
parser = new ADQLParser(new java.io.ByteArrayInputStream(query.getBytes()));
}
// Otherwise, take the query as provided:
else {
// Initialise the parser with the specified input:
if (file == null || file.length() == 0)
parser = new ADQLParser(System.in);
else if (file.matches(urlRegex))
parser = new ADQLParser((new java.net.URL(file)).openStream());
else
parser = new ADQLParser(new java.io.FileInputStream(file));
}
// Enable/Disable the debugging in function of the parameters:
parser.setDebug(explain);
// Query parsing:
try{
if (verbose) System.out.print("((i)) Parsing ADQL query...");
ADQLQuery q = parser.parseQuery();
if (verbose) System.out.println("((i)) CORRECT ADQL QUERY ((i))");
if (mode==2){
PostgreSQLTranslator translator = new PostgreSQLTranslator();
if (verbose) System.out.print("((i)) Translating in SQL...");
String sql = translator.translate(q);
if (verbose) System.out.println("ok");
System.out.println(sql);
}else if (mode==1){
System.out.println(q.toADQL());
}
}catch(UnresolvedIdentifiersException uie){
System.err.println("((X)) "+uie.getNbErrors()+" unresolved identifiers:");
for(ParseException pe : uie)
System.err.println("\t - at "+pe.getPosition()+": "+uie.getMessage());
if (debug) uie.printStackTrace(System.err);
System.exit(3);
}catch(ParseException pe){
System.err.println("((X)) Syntax error: "+pe.getMessage()+" ((X))");
if (debug) pe.printStackTrace(System.err);
System.exit(3);
}catch(TranslationException te){
if (verbose) System.out.println("error");
System.err.println("((X)) Translation error: "+te.getMessage()+" ((X))");
if (debug) te.printStackTrace(System.err);
System.exit(4);
}