1ANTLR 3.5 Release 2January 4, 2012 3 4Terence Parr, parrt at cs usfca edu 5ANTLR project lead and supreme dictator for life 6University of San Francisco 7 8CHANGES 9 10January 4 2012 -- release 3.5 11 12January 3, 2012 13 14* Improve error reporting and recovery for STRAY_BRACKET, fixes antlr/antlr3#42 15* Do not write output files if errors were reported, fixes antlr/antlr3#61 16* Fix AST operator on labeled set of terminals e.g. x=(A|B)^ 17* Fix labeled set of terminals with rewrite e.g. x=(A|B) -> $x 18 19December 1, 2012 20 21* Added error msg for .. in parser 22 23September 17, 2012 24 25* Add Gokulakannan Somasundaram's C++ target based upon C target. 26 Use language=Cpp in options. It's a header-only library, runtime/Cpp/include, 27 so installation is not required. 28 29September 16, 2012 30 31* Python 3.3 target added by Benjamin Wolf based upon Python 2 target 32 https://github.com/antlr/antlr3/pull/23 33 34September 15, 2012 35 36* LookaheadStream bug fixes; 37 https://github.com/antlr/antlr3/pull/21 38 39* Pulled "Fix Python handling of syntactic predicates" 40 https://github.com/antlr/antlr3/pull/33 41 42July 15, 2012 43 44* GUnit improvements 45 https://github.com/antlr/antlr3/pull/27 46 47May 2012: 48 49* ANTLR3 update of ObjC runtime to go with latest ST4-ObjC 50 https://github.com/antlr/antlr3/pull/17 51 52August 9, 2012 53 54* Provide Lexer get end of file method so people can override it. 55 56November 25, 2011 57 58* stderr not test correctly in gunit examineExecResult 59 60October 27, 2011 61 62* Dieter Habelitz reported bug in java code gen with synpreds. labels were 63 being carried from grammar into synpreds but weren't typed properly (they 64 were "void x=null;" for x=ruleref labels) 65 66October 25, 2011 67 68* (Sam) Rule.setOption didn't do memoize option right. 69* (Sam) Fix labels in synpreds 70* (Sam) Fix input index for NoViableAltException during inline prediction 71* (Sam) Fix label aliasing errors in cases like (x=y|x=z) 72 73August 10, 2011 74 75* (Sam) fix stack overflow in semantic context analysis 76 77July 30, 2011 78 79* added range check to BaseTree.insertChild() 80 81July 18, 2011 -- release 3.4 82 83* Added tree method insertChild(int i, Object t). 84 85July 14, 2011 86 87* Added BaesTree.freshenParentAndChildIndexesDeeply() to recursively 88 walk tree and set ptrs. 89 90July 6, 2011 91 92* reset() for token stream didn't skip initial off-channel tokens. 93 94July 5, 2011 95 96* Sam fixes rare infinite loop upon error in nextToken(). 97* rewrites weren't pulled from syntactic predicates. 98 99June 29, 2011 100 101* Sam noticed CommonToken.getText() cached substring pulled from input, which 102 defeated purpose of pointing into input char array. Altered to not cache. 103 Should reduce memory footprint. 104 105June 24, 2011 106 107* empty alts with actions didn't have EPSILON dummy alt node. 108 109June 19, 2011 110 111* Udo noticed that Parser.getMissingSymbol() didn't set invented token's input 112 113June 8, 2011 114 115* Added inContext(String context) back into TreeParser. 116 117April 21, 2011 118 119* Updated for ST v4.0.2 (setting iterateAcrossValues = true as instance var) 120* Needed throws condition for delegatedRules. 121 122April 20, 2011 (Sam Harwell) 123 124* Implement the 'throwsSpec' feature of parser rules for the Java target 125* Improve algorithm for SemanticContext Boolean predicate reduction 126 127April 13, 2011 128 129* Unmangled region names in STViz hiearchy tree display. 130* Removed conversion timeout thing again 131 132April 11, 2011 133 134* Added option -Xconversiontimeout back in. Turns out we hit NFA conversion 135 time landmine occasionally with huge grammars; fails over to backtracking 136 (if turned on) if it can't make DFA. 137 138March 29 - April 10, 2011 139 140* Uses ST v4 now!!! Massive change. Only updated Java target so far. 141 Ripped out ST v3 usage to remove indirect dependency on ANTLR v2. 142 143March 28, 2011 144 145* Sam Harwell ported all v2 grammars to v3! 146 147March 3, 2011 148 149* left-recursion pattern off until all targets catch up 150 151* ANTLRCore.sti no longer used; removed from all targets. 152 153* Adding type, text terminal options 154 155* Replaced hetero arg with terminalOptions arg in all templates that reference hetero 156 (this is the class name / node type used for TOKEN<NODETYPE> references 157 in grammar). Widespread but trivial changes to all targets. hetero is 158 now terminalOptions.node. Can also get terminalOptions.type and 159 terminalOptions.text from dictionary of options. 160 161* Fixed mispelling of license in BSD license headers 162 163March 3, 2011 164 165* Add tree, getTree() to ParserRuleReturnScope to do away with specific ones like: 166 public static class rewrite_template_args_return extends ParserRuleReturnScope { 167 CommonTree tree; 168 public Object getTree() { return tree; } 169 }; 170 Removed these special classes if it's just AST; keep if they have defined "returns" 171 values 172 173February 26, 2011 174 175* All finally {} have comment now to suppress warning. 176 177* removed ; from {;} blank method 178 179* Added @SuppressWarnings({"all"}) to front of each 180 generated class. 181 182* -print wasn't always showing ^ and ! in grammar 183 184* Added java-left-recur/Java.g example. 185 186* left-recursion pattern detection handles backtracking mode now 187 188February 25, 2011 189 190* -Xmaxinlinedfastates went to 60 from 10 for forcing prediction in left- 191 recursive expression rules to stay in rule; preds use a parameter. 192 193* trees know where they came from now start..stop tokens; todo: use for better err handling. 194 195* Got immediate left-recursion working for rules. Added TestLeftRecursion.java 196 197February 21, 2011 198 199* Fixed http://www.antlr.org/jira/browse/ANTLR-437 thanks to Vladislav Kuzkokov. 200 added unit test. 201 202February 17, 2011 203 204* Add -language L option to override language=L option in grammar. Same 205 grammar can yield multiple parsers in different languages now. 206 207February 10, 2011 208 209* Added method to generated Java code to return the array of delegates; e.g., 210 import Sub1, Sub2; 211 yields: 212 213 public Parser[] getDelegates() { 214 return new Parser[] {gSub1, gSub2}; 215 } 216 217January 25, 2011 218 219* Improve error messages for no viable alt parse exceptions 220 221January 20, 2011 222 223* TokenRewriteStream had dead code; also updated insertBefore followed by 224 replace or delete. If input is abc and I did insertBefore(2,"y"), where 225 'c' is index 2, then did delete of 2 previously defined functionality 226 was to ignore the insert. that's weird; fixed to keep insert. Also 227 Delete special case of replace (text==null): 228 D.i-j.u D.x-y.v | boundaries overlap => combine to max(min)..max(right) 229 230December 12, 2010 231 232* Send couldBacktrack now to enterDecision in debug protocol 233 234December 4, 2010 235 236* TreeWizard ctor needed a default tree adapator. 237 238November 29, 2010 -- ANTLR v3.3 239 240November 23, 2010 241 242* CodeGenerator.loadLanguageTarget is now static and available to load 243 targets so we can ask them questions during analysis. 244 245* Fixed and added unit test 246 http://www.antlr.org/jira/browse/ANTLR-370 247 http://www.antlr.org/jira/browse/ANTLR-375 248 249November 23, 2010 250 251* Added source name to syntax error msgs 252 253October 20, 2010 254 255Added boolean couldBacktrack to enterDecision in dbg interface. Breaks AW 256interface and other tools! [BREAKS BACKWARD COMPATIBILITY] 257 258October 17, 2010 259 260* Missing -trace in help msg 261 262November 22, 2010 263 264* Added GrammarAST: public int getCharPositionInLine() { return getColumn()-1; } 265 and Grammar.getHasDelegates() for C# guys 266 267October 16, 2010 268 269* Doesn't write profile data to file anymore; emits decision data to stderr 270 271October 14, 2010 272 273* Make OrderedHashSet have deterministic iteration 274 275July 20, 2010 276 277* greedy=true option shuts off nondeterminism warning. 278 279* code gen for AST and -profile didn't compile. had useless line: 280 281 proxy.setTreeAdaptor(adap); 282 283 284July 17, 2010 285 286* Removed conversion timeout failsafe; no longer needed. 287 288* Stats updated to be correct for -report. 289 290June 10, 2010 291 292* added toArray in OrderedHashSet to make addAll calls get same order for DFA edges and possibly code gen in some areas. 293 294June 5, 2010 295 296* Added -Xsavelexer 297 298May 24, 2010 299 300* lexerStringRef was missing elementIndex attribute. i='import' didn't work 301 in lexer. Altered all target stg files. Set in codegen.g 302 303* output=AST, rewrite=true for tree rewriters broken. nextNode for subtree 304 streams didn't dup node, it gave whole tree back. 305 306March 17, 2010 307 308* Added MachineProbe class to make it easier to highlight ambig paths in 309 grammar. More accurate than DecisionProbe; retrofitted from v4. 310 311February 20, 2010 312 313* added range to TokenStream and implementors: 314 /** How far ahead has the stream been asked to look? The return 315 * value is a valid index from 0..n-1. 316 */ 317 int range(); 318 319* added new method to get subset of tokens to buffered token streams: 320 public List get(int start, int stop); 321 322February 15, 2010 323 324* Refs to other tokens in a lexer rule didn't get its line/charpos right. 325 altered Java.stg. 326 327January 31, 2010 328 329* Creating token from another token didn't copy input stream in CommonToken. 330 makes sense to copy too; i don't think anybody relies on it being null after 331 a copy. We might want to know where token came from. 332 333January 26, 2009 334 335* TreeParser.getMissingSymbol() use CommonTree instead of using 336 adaptor.create() 337 338December 8, 2009 339 340* Instead of sharing Token.EOF_TOKEN, I'm now creating EOF tokens so I can set the char position for better error messages. 341 342December 5, 2009 343 344* Fixed bug in TreeVisitor when rewrites altered number of children. Thanks to Chris DiGiano. 345 346* added new buffered on-demand streams: BufferedTokenStream. Renamed CommonTokenStream to LegacyCommonTokenStream and made new one as subclass of BufferedTokenStream. 347 348November 3, 2009 349 350* Added org.antlr.runtime.UnbufferedTokenStream. Was trivial and works! 351 352November 1, 2009 353 354* Couldn't properly reuse parser state; ctor reset the state; fixed. 355 Parser(TokenStream input, RecognizerSharedState state) 356 357* LookaheadStream<T> used some hardcoded Object return types for LT, etc... 358 uses T now. 359 360September 23, 2009 -- ANTLR v3.2 361 362September 21, 2009 [Jim Idle] 363 364* Added new options for tool invocation to control the points at which the code 365 generator tells the target code to use its equivalent of switch() instead of 366 inline ifs. 367 -Xmaxswitchcaselabels m don't generate switch() statements for dfas 368 bigger than m [300] 369 -Xminswitchalts m don't generate switch() statements for dfas smaller 370 than m [3] 371* Upgraded -X help output to include new optins and provide the default 372 settings, as well as provide units for those settings that need them. 373 374* Change the C Target to overide the deafults for the new settings to 375 generate the most optimizable C code from the modern C compiler point of 376 view. This is essentially to always use swtich statements unless there 377 is absolutely no other option. C defaults are to use 1 for minimum and 378 3000 for maximum number of alts that trigger switch(). This results in 379 object code that is 30% smaller and up to 20% faster. 380 381April 23, 2009 382 383* Added reset to TreeNodeStream interface. 384 385April 22, 2009 386 387* Fixed ANTLR-374. Was caused by moved of grammars. %foo() stuff didn't work 388 389April 9, 2009 390 391* .g and .g3 file extensions work again. 392* introduced bug in 3.1.3: gives exception not error msg upon 393 missing .g file 394 395March 26, 2009 396 397* Made ctor in TreeRewriter and TreeFilter call this not super. 398 399March 21, 2009 400 401* Added ctor to RecognizerSharedState to allow cloning it. 402 403March 17, 2009 -- ANTLR v3.1.3 404 405* improved ANTLRv3.g to handle <...> element options 406 407March 15, 2009 408 409* Fixed ANTLR-389. Strip didn't ignore options in subrules; also seemed 410 to demand stdin. 411 412March 15, 2009 413 414* ANTLR always sorts incoming grammar list by dependency. For example, 415 If W.g depends on tokens from P.g then P.g is done first even if 416 W.g mentioned first on command line. It does not ignore any files you 417 specify the commandline. If you do *.g and that includes some 418 imported grammars, it will run antlr on them. 419 420* -make option prevents ANTLR from running on P.g if P older than 421 generated files. 422 423* Added org.antlr.tool.GrammarSpelunker to build a faster dependency 424 checker (what grammars depend on etc...). Totally independent of any 425 ANTLR code; easy to pull into other tools. 426 427* Added org.antlr.misc.Graph, a general graph with nodes 428 containing an Object payload. It knows how to do a topological sort 429 on the nodes. 430 431March 10, 2009 432 433* Added associativity token option to support upcoming special expression 434 parsing. Added rule option strategy=precedence also 435 436March 1, 2009 437 438* Changed ANTLRWorks debug port from 49153 to 49100. Apparently we change the port in 439 ANTLRWorks to 49100 in 1.2 but forgot to do so in the ANTLR targets. 440 441START CHANGES FOR TREE FILTER MODE (pulled from dev branch) 442 443This feature will be announced in 3.2, but I am integrating from my development branch now into the mainline so target developers have a chance to implement. We might release 3.1.3 bug fix release before 3.2. 444 445* CommonTreeNodeStream -> BufferedTreeNodeStream. Now, 446 CommonTreeNodeStream is completely unbuffered unless you are 447 backtracking. No longer making a list of all nodes before tree parsing. 448 449* Added tree grammar filter=true mode. 450 451 Altered templates: 452 Java.stg: added filterMode to genericParser and treeParser. 453 This required a change to ANTLRCore.sti 454 Defined a default parameter in treeParser to set the superclass 455 to TreeFilter for tree grammar with filter=true. It sets 456 superclass to TreeRewriter if filter=true and output=AST. 457 Other them that, I only had to change ANTLR itself a little bit. 458 Made filter mode valid for tree grammars and have it automatically set 459 the necessary elements: @synpredgate, backtrack=true, rewrite=true 460 (if output=AST). Added error message for detecting conflicting 461 options. 462 463* Added misc.FastQueue and TestFastQueue: 464 A queue that can dequeue and get(i) in O(1) and grow arbitrarily large. 465 A linked list is fast at dequeue but slow at get(i). An array is 466 the reverse. This is O(1) for both operations. 467 468* Added tree.TreeIterator, a generator that walks a doubly linked tree. 469 The nodes must know what index they are. It's an Iterator but 470 remove() is not supported. Returns navigation nodes always: 471 UP, DOWN, EOF. 472 473* Added misc.LookaheadStream: A lookahead queue that knows how 474 to mark/release locations in the buffer for backtracking purposes. 475 I hope to use for both tree nodes and tokens. Just implement 476 nextElement() to say how to get next node or token. 477 478END CHANGES FOR TREE FILTER MODE 479 480February 23, 2009 -- ANTLR v3.1.2 481 482February 18, 2009 483 484* Added org.antlr.tool.Strip (reads from file arg or stdin, emits to stdout) 485 to strip actions from a grammar. 486 487February 4, 2009 488 489* Added CommonTree.setUnknownTokenBoundaries(). Sometimes we build trees 490 in a grammar and some of the token boundaries are not set properly. 491 This only matters if you want to print out the original text associated 492 with a subtree. Check this out rule: 493 494 postfixExpression 495 : primary ('.'^ ID)* 496 ; 497 498 For a.b.c, we get a '.' that does not have the token boundaries set. 499 ANTLR only sets token boundaries for subtrees returned from a rule. 500 SO, the overall '.' operator has the token boundaries set from 'a' 501 to 'c' tokens, but the lower '.' subtree does not get the boundaries 502 set (they are -1,-1). Calling setUnknownTokenBoundaries() on 503 the returned tree sets the boundaries appropriately according to the 504 children's token boundaries. 505 506January 22, 2009 507 508* fixed to be listeners.add(listener); in addListener() of DebugEventHub.java 509 510January 20, 2009 511 512* Removed runtime method: mismatch in BaseRecognizer and TreeParser. Seems 513 to be unused. Had to override method recoverFromMismatchedToken() in 514 TreeParser to get rid of single token insertion and deletion for 515 tree parsing because it makes no sense with all of the up-and-down nodes. 516 517* Changed JIRA port number from 8888 to no port spec (aka port 80) and all 518 refs to it in this file. 519 520* Changed BaseTree to Tree typecase in getChild and toStringTree() and 521 deleteChild() to make more generic. 522 523December 16, 2008 524 525* Added -verbose cmd-line option and turned off standard header 526 and list of read files. Silent now without -verbose. 527 528November 24, 2008 529 530* null-ptr protected getParent and a few others. 531 532* Added new ctor to CommonTreeNodeStream for walking subtrees. Avoids 533 having to make new serialized stream as it can reuse overall node stream 534 buffer. 535 536November 20, 2008 537 538* Updated BaseTest to isolate tests better. 539 540November 17, 2008 541 542* BaseTreeAdaptor.getType() was hosed; always gave 0. Thanks to Sam Harwell. 543 544November 8, 2008 545 546* Added methods to BaseRecognizer: 547 public void setBacktrackingLevel(int n) { state.backtracking = n; } 548 /** Return whether or not a backtracking attempt failed. */ 549 public boolean failed() { return state.failed; } 550 551November 5, 2008 552 553* Tweaked traceIn/Out to say "fail/succeeded" 554 555* Bug in code gen for tree grammar wildcard list label x+=. 556 557* Use of backtrack=true anywhere in grammar causes backtracking sensitive 558 code to be generated. Actions are gated etc... Previously, that only 559 happened when a syntactic predicate appeared in a DFA. But, we need 560 to gate actions when backtracking option is set even if no decision 561 is generated to support filtering of trees. 562 563October 25, 2008 564 565* Fixed debug event socket protocol to allow spaces in filenames. 566 567* Added TreeVisitor and TreeVisitorAction to org.antlr.runtime.tree. 568 569October 22, 2008 570 571* Added inContext() to TreeParser. Very useful for predicating 572 tree grammar productions according to context (their parent list). 573 Added new TestTreeContext unit tests (15). 574 575 /** Check if current node in input has a context. Context means sequence 576 * of nodes towards root of tree. For example, you might say context 577 * is "MULT" which means my parent must be MULT. "CLASS VARDEF" says 578 * current node must be child of a VARDEF and whose parent is a CLASS node. 579 * You can use "..." to mean zero-or-more nodes. "METHOD ... VARDEF" 580 * means my parent is VARDEF and somewhere above that is a METHOD node. 581 * The first node in the context is not necessarily the root. The context 582 * matcher stops matching and returns true when it runs out of context. 583 * There is no way to force the first node to be the root. 584 */ 585 public boolean inContext(String context) {...} 586 587* Added 3 methods to Tree interface [BREAKS BACKWARD COMPATIBILITY] 588 589 /** Is there is a node above with token type ttype? */ 590 public boolean hasAncestor(int ttype); 591 592 /** Walk upwards and get first ancestor with this token type. */ 593 public Tree getAncestor(int ttype); 594 595 /** Return a list of all ancestors of this node. The first node of 596 * list is the root and the last is the parent of this node. 597 */ 598 public List getAncestors(); 599 600October 21, 2008 601 602* Updated unit tests to be correct for \uFFFE->\uFFFF change 603 604* Made . in tree grammar look like ^(. .*) to analysis, though ^(. foo) 605 is illegal (can't have . at root). Wildcard is subtree or node. 606 Fixed bugs: 607 http://www.antlr.org/browse/ANTLR-248 608 http://www.antlr.org/browse/ANTLR-344 609 610October 1, 2008 -- ANTLR v3.1.1 611 612September 8, 2008 613 614* Labels on tokens, rules carry into synpreds now so semantic predicates work. 615 This didn't work since labels were stripped in the synpred and they weren't 616 defined in the generated method. 617 618 a : x=A z=a {$x.text.equals($z.text)}? A 619 | y=A a A A 620 ; 621 622September 3, 2008 623 624* Made a REV static variable in Tool so that we can change the rev for 625 daily builds. 626 627* Made \uFFFF a valid character. Token types are 32-bit clean using -1 628 not 0x0000FFFF as -1 so it should be okay. Label.java: 629 public static final int MIN_CHAR_VALUE = '\u0000'; 630 public static final int MAX_CHAR_VALUE = '\uFFFF'; 631 632August 30, 2008 633 634* Changed messages in en.stg so that TOKEN_NONDETERMINISM correctly 635 indicates when actions hid semantic predicates. 636 637August 15, 2008 638 639* Tweaked build properties and build.xml 640 641August 13, 2008 642 643* Fixed ANTLR-314; 3.1 introduced a problem with list labels += 644 645August 12, 2008 -- ANTLR v3.1 646 647* Added JavaScript target 648 649August 7, 2008 650 651* an NFA target of EOF predicate transition in DFA cause an exception in 652 getPredicatesPerNonDeterministicAlt(). 653 654* Kay Roepke found a nasty bug when debugging AST-constructing 655 composite recognizers. If the input state was null to the constructor, 656 super class constructor created a new parser state object. 657 Later, though we passed the argument state not this.state 658 to the delegate constructors, forcing them to share a different 659 state objects! Changed state to this.state in Dbg.stg constructors. 660 661* Ack. messed up debug/AST. Have to set proxy's tree adaptor; it's 662 a circular ref. Just an ASTDbg.stg change. 663 664August 4, 2008 665 666* superClass works now for lexers 667 668* Made Grammar.defineNamedAction propogate header actions down to all 669 delegates if root grammar; regardless of lexer/parser scope. 670 671* Rejiggered AST templates to propogate changes to tree adaptor 672 for delegate grammars. Fixes ANTLR-302 673 674August 4, 2008 675 676* FOLLOW set computations altered constant FOLLOW bit sets. 677 678* Added (...) are all predicate evaluations. 679 680* Extra init code for tree parser nonrewrite mode removed. 681 682* Added empty child list check in becomeRoot 683 684August 3, 2008 685 686* Was using RuleReturnScope not Rulename_return for list labels in tree 687 parser. 688 689* Didn't set _last in tree parser for rule ref track stuff (rewrite=true) 690 691August 2, 2008 692 693* Benjamin found another rewrite engine bug. 694 695July 30, 2008 696 697* CommonTreeNodeStream / CommonTokenStream did not reset properly. 698 699July 29, 2008 700 701* Fixed another bug in TokenRewriteStream; didn't like inserts after end. 702 703July 28, 2008 704 705* Fixed bug in TokenRewriteStream.toString(start,stop); it ignored 706 parameters. ;) 707 708July 17, 2008 709 710* allow qualified type names in hetero <...> options like T<a.b.c.Node> 711 712July 5, 2008 713 714* treeLevel not set for setBlock alts; added unit test 715 716July 3, 2008 717 718* Fixed ANTLR-267. parse tree added nodes during backtracking and 719 cyclic DFAs. tracks hidden tokens too now. Added toInputString() to 720 get text back including hidden tokens. Shows <epsilon> for rules 721 that match nothing. 722 723June 26, 2008 724 725* Added gParent ptr that points to immediate parent grammar. E.g., 726 // delegators 727 public MParser gM; 728 public M_S gS; 729 public M_S gParent = gS; // NEW 730 731* Grammar imports didn't set all the delegate pointers...lots of imported 732 grammars would cause a null ptr exception. Fixes ANTLR-292. 733 734June 25, 2008 735 736* List labels in tree construction didn't always track the tree; sometimes 737 had a rule result structure. 738 739June 4, 2008 740 741* Improved unit testing so that each test suite executes and builds grammars 742 in a separate temporary directory. This means they can execute concurrently. 743 Also seem to be a problem with my class path during execution. Moved 744 tmpdir for ahead of standard CLASSPATH. 745 746* By virtue of an improvement to StringTemplate, output newlines 747 in generated files should be normalized to whatever your host uses. 748 749June 3, 2008 750 751* Restrict legality of grammar options; for example you cannot use output option 752 in lexer anymore. 753 754June 2, 2008 755 756* Throw illegal arg exception upon invalid TokenRewriteStream ops. Rewrote 757 core of engine. Slightly different operation. Added many more unit tests. 758 7593.1b1 - May 20, 2008 760 761May 11, 2008 762 763* rewrite=true, output=AST for tree grammar was not working. Altered trees were not 764 propagated back up the rule reference chain. Required a number of mods to 765 ASTTreeParser.stg. Added unit tests. 766 767May 10, 2008 768 769* [BACKWARD INCOMPATIBLE if you override match()] 770 I had turned off single token insertion and deletion because I could not figure 771 out how to work with trees and actions. Figure that out and so I turned it back on. 772 match() returns Object matched now (parser, tree parser) so we can set labels 773 on token refs properly after single token ins/del error recovery. Allows actions 774 and tree construction to proceed normally even though we recover in the middle of 775 an alternative. Added methods for conjuring up missing symbols: getMissingSymbol(). 776 777* refactored BaseRecognizer error handling routines 778 779* Single token error recovery was not properly taking into consideration EOF. 780 781* ANTLR no longer tries to recover in tree parsers inline using single node deletion or insertion; throw exception. Trees should be well formed as they are not created by users. 782 783* Added empty constructors to the exception classes that did not have them so that ANTLRWorks can create the exceptions. 784 785* Made debug tree adaptor deal with tokens conjured up during error recovery. 786 787* Removed extra location() debug element that was emitted. 788 789May 8, 2008 790 791* ANTLR didn't update line/col to DFA map for AW. 792 793May 6-7, 2008 794 795* Insufficiently covered (with semantic predicates) alt warnings are now emitted before 796 nondeterminisms so it's clear the nondeterminism is a result of insufficient preds. 797 798* Improved insufficiently covered alt warnings from: 799 warning(203): T.g:2:3: The following alternatives are insufficiently covered with predicates: 1 800 to: 801 warning(203): T.g:2:3: Input B is insufficiently covered with predicates at loca 802tions: alt 1: line 3:15, alt 2: line 2:9 803 804* Improved nondeterminism warning to have: 805 Semantic predicates were present but were hidden by actions. 806parser grammar U; 807a : (A B)? ; 808b : X a {p1}? A B | Y a {a1} {p2}? A B | Z a ; 809 810To create the prediction DFA for the optional sub rule in 'a', ANTLR must find all references to 'a' to determine what can follow. A B can follow 'a' in the first two alts rule 'b'. To resolve the conflict between matching A B immediately in the sub rule and exiting rule 'a' to match it in 'b', ANTLR looks for predicates. In this case, there are two predicates that indicate the semantic context in which the surrounding alternatives are valid. The problem is that one of the predicates is hidden by an action. It took me 1.5 days, but I've finally have gotten ANTLR to properly track the insufficiently covered alternatives. Further, I have gotten it to tell you precisely where the uncovered predicates are even if they are simply hidden by actions. I have also updated all of the nondeterminism warnings so that it tells you if there was a predicate but one hidden by an action (this could be a separate condition from insufficiently covered predicates). here are your messages from ANTLR: 811 812ANTLR Parser Generator Version 3.1b1 (??) 1989-2007 813warning(203): U.g:2:5: Input such as "A B" is insufficiently covered with predicates at locations: alt 2: line 3:38 at B 814Semantic predicates were present but were hidden by actions. 815warning(200): U.g:2:5: Decision can match input such as "A B" using multiple alternatives: 1, 2 816As a result, alternative(s) 2 were disabled for that input 817Semantic predicates were present but were hidden by actions. 818 819* Fixed issue where 820r41 821 : (INT -> INT) ( ('+' i=INT) -> ^($i $r41) )* ';' 822 ; 823still warned about $r41 being ambig. 824 825* actions are now added to the NFA. 826 827* Fixed ANTLR-222. ANTLR now ignores preds after actions. 828 829May 5, 2008 830 831* Fixed ANTLR-235 by backing out a change from 12-31-07. 832 833* Fixed ANTLR-249; I include semantic context again in closure busy signal. 834 835May 3, 2008 836 837* Fixed ANTLR-208. Looks in library or in -o output path. antlr -o foo T.g U.g where U needs T.tokens won't work unless we look in foo too. fixed. 838 839* Refactored assign.types.g to move methods to a class called AssignTokenTypesBehavior. 840 841* Fixed ANTLR-207. Lexers importing vocabs didn't see ';'=4 type aliases in .tokens. 842 843* Fixed ANTLR-228. Couldn't use wildcard in alts with AST rewrites. 844 845May 2, 2008 846 847* Fixed ANTLR-230; can use \' now in action. 848 849* Scope attributes no longer have a stack depth check on front. If you ref $r::a when r has not invoked you, then you get an exception not a default value. Back to the way 3.0.1 worked. 850 851* $channel was a global variable in 3.0.1 unlike $type which did not affect an invoking lexer rule. Now it's local too. Only $type and $channel are ever set with regularity. Setting those should not affect an invoking lexer rule as in the following should work: 852 853 X : ID WS? '=' ID ; // result is X on normal channel 854 WS : ' '+ {$channel = HIDDEN; } ; 855 856 STRING : '"' (ESC|.)* '"' ; // result is STRING not ESC 857 858 FLOAT : INT '.' INT? ; // should be FLOAT 859 INT : Digit+ ; 860 fragment 861 Digit : '0'..'9' ; 862 863* Fixed bug in interpreter regarding (...)* loops 864 865May 1, 2008 866 867* Fixed ANTLR-202. These now give warnings about ambig ref to $a. 868 a : ID a -> $a | INT ; 869 and 870 a : A a {$a.text} | B ; 871 872April 30, 2008 873 874* Fixed ANTLR-237. updated -depend to know about imported grammars. 875$ java org.antlr.Tool -depend -lib foo T.g 876 ANTLR Parser Generator Version 3.1b1 (??) 1989-2007 877 T.g: foo/Java.g 878 TParser.java : T.g 879 T.tokens : T.g 880 TLexer.java : T.g 881 T_Java : T.g 882 883April 29, 2008 884 885* Fixed ANTLR-217; scope A,B,C; didn't work 886 887* Fixed ANTLR-224; ! or ^ on item in alt with rewrite gave exception 888 889* Added token options to terminals: ID<node=V; foo="Big bob"> etc... 890 node is default so you can do ID<V> for hetero tree types. most common. 891 892April 17, 2008 893 894* Use default msg if unknown recog type in getErrorMessage(): 895 String msg = e.getMessage(); 896 897April 14, 2008 898 899* %x.y = foo; was not working in @members section 900 901March 29, 2008 902 903* Import couldn't handle A imports B imports C. 904 905March 27, 2008 906 907* Added get/setInputStream to Token interface and affected classes. 908 909February 26, 2008 910 911* made fillBuffer public in CommonTreeNodeStream so we can add trees 912 to stream for interactive interpreters. 913 914February 14, 2008 915 916* Fixed a bug in the code generation where tree level 0 was used 917 no matter what to rewrite trees in tree grammars. added unit test 918 919* Fixed ANTLR-221. exceptions were generated when using 920 AST construction operators and no output=AST option. 921 922February 13, 2008 923 924* Improved error msgs for unreachable alts and tokens. 925 926February 11-12, 2008 927 928* Fixed ANTLR-219. 929 It looks like the AST construction code for sets was totally messed up. 930 This was for not only the new tree parser AST construction, but also 931 the regular tree construction for parsers. I had to introduce templates 932 in the ASTTreeParser.stg file to deal with this. added unit tests: 933 TestTreeGrammarRewriteAST.testSetMatchNoRewrite(), 934 testSetMatchNoRewriteLevel2(), testSetMatchNoRewriteLevel2Root(). 935 Had to add template matchRuleBlockSet() 936 to differentiate between a regular set in one that is an entire rule. 937 If it is an entire rule, it has to set the return value, retval.tree. 938 939* Fixed ANTLR-220. 940 Made TreeAdaptor dupNode and dupTree events emit debugging events 941 so AW could see tree node duplications. 942 943February 4, 2008 944 945* BACKWARD INCOMPATIBILITY 946 Added getSourceName to IntStream and TokenSource interfaces and also the 947 BaseRecognizer. Have to know where char come from for error messages. 948 Widespread change, but a trivial one. 949 950January 17, 2008 951 952* Interpreter throws FailedPredicateException now when it sees a predicate; 953 before it was silently failing. I'll make it work one of these days. ;) 954 955January 12, 2008 956 957* Copy ctor not copying start and stop for common token. Fixes ANTLR-212 958 959* Removed single token insertion and deletion for tokens, sets. 960 Required a change to the code generation for matchSet() template 961 and a tweak inside the BaseRecognizer. To engage this again is easy, 962 just override mismatch() to call mismatchRecover(). I changed it to simply 963 throw an exception. 964 965* Added syntaxError recognizer state var so you can easily tell if 966 a recognizer failed. Added getNumberOfSyntaxErrors() to recognizers. 967 968* Added doc for the error node stuff: 969 http://www.antlr.org/wiki/display/ANTLR3/Tree+construction 970 971* Fixed ANTLR-193 972 973* Added recognizer methods to answer questions about current mismatched 974 token error. Useful now since i don't automatically recover inline 975 to such errors (I throw exception): 976 mismatchIsUnwantedToken(IntStream input, int ttype) 977 mismatchIsMissingToken(IntStream input, BitSet follow) 978 979* Added UnwantedTokenException and MissingTokenException to make 980 match() problems more precise in case you want to catch differently. 981 Updated getErrorMessage() to be more precise. Says: 982 983 line 2:9 missing EQ at '0' 984 985 now instead of 986 987 line 2:9 mismatched input '0' expecting EQ 988 989 Input "x=9 9;" gives 990 991 line 3:8 extraneous input '9' expecting ';' 992 993 When very confused, "x=9 for;", you still get old mismatched message: 994 995 line 3:8 extraneous input 'for' expecting ';' 996 line 3:11 mismatched input ';' expecting '(' 997 998* Added unit tests to TestAutoAST and copied to TestRewriteAST with 999 suitable rewrites to ensure basic error node insertion works. 1000 1001January 11, 2008 1002 1003* Adding errorNode to TreeAdaptor and various debug 1004 events/listeners. Had to add new class runtime.tree.CommonErrorNode 1005 to hold all the goodies: input stream, start/stop objects. 1006 1007* Tweaked CommonTree.getType() to return INVALID_TOKEN_TYPE 1008 instead of literal 0 (same thing at moment though). 1009 1010* Updated ANTLRWorks to show error nodes in tree as much as I could; Jean 1011 will get to rest of it. 1012 1013January 9-10, 2008 1014 1015* Continued work on debugging/profiling composite grammars. 1016 1017* Updated debug protocol for debugging composite grammars. enter/exit 1018 rule needs grammar to know when to flip display in AW. 1019 1020* Fixed ANTLR-209. ANTLR consumed 2 not 1 char to recover in lexer. 1021 1022* Added two faqs instead of making changes to antlr runtime about 1023 lexer error handling: 1024 http://www.antlr.org/wiki/pages/viewpage.action?pageId=5341230 1025 http://www.antlr.org/wiki/pages/viewpage.action?pageId=5341217 1026 1027January 1-8, 2008 1028 1029* Making debugging/profiling work with composite grammars. 1030 1031* Updated ANTLRWorks so it works still for noncomposite grammars. 1032 1033* two new examples: import and composite-java (the java example grammar 1034 broken up into multiple pieces using import). 1035 1036* Worked on composite grammars. Had to refactor a lot of code to make 1037 ANTLR deal with one meta grammar made up of multiple grammars. I 1038 thought I had it sort of working back in August. Yes, but barely. Lots 1039 of work to do it seemed. Lots of clean up work. Many new unit tests 1040 in TestCompositeGrammars. Had to add new error messages warning about 1041 conflicting tokens inherited from multiple grammars etc... 1042 1043 TOKEN_ALIAS_CONFLICT(arg,arg2) ::= 1044 "cannot alias <arg>; string already assigned to <arg2>" 1045 TOKEN_ALIAS_REASSIGNMENT(arg,arg2) ::= 1046 "cannot alias <arg>; token name already assigned to <arg2>" 1047 TOKEN_VOCAB_IN_DELEGATE(arg,arg2) ::= 1048 "tokenVocab option ignored in imported grammar <arg>" 1049 INVALID_IMPORT(arg,arg2) ::= 1050 "<arg.grammarTypeString> grammar <arg.name> cannot import <arg2.grammarTypeString> grammar <arg2.name>" 1051 IMPORTED_TOKENS_RULE_EMPTY(arg,arg2) ::= 1052 "no lexer rules contributed to <arg> from imported grammar <arg2>" 1053 IMPORT_NAME_CLASH(arg,arg2) ::= 1054 "combined grammar <arg.name> and imported <arg2.grammarTypeString> grammar <arg2.name> both generate <arg2.recognizerName>; import ignored" 1055 1056 This stuff got really really complicated. Syntactic predicate names even 1057 had to be scoped per grammar so they don't conflict. 1058 1059* When using subrules like (atom->atom) to set result tree, it was not 1060 properly setting result (early enough). Future code got null for 1061 $rule.tree. 1062 1063December 31, 2007 1064 1065* Added the start of a semantic predicate computation for LL(1) to 1066 solve a problem with slow grammar analysis even with k=1 due to 1067 predicates. Then I realized the problem with that grammar was 1068 elsewhere. Semantic context really shouldn't be used when 1069 preventing closure recomputation (May 2008 I discovered I was 1070 wrong--you do need it). The predicates became huge even though the 1071 reduced value would be no different. The analyzer seems faster now 1072 that I am not testing predicate values all the time. Further it may 1073 terminate sooner just due to reduced closure recursion. 1074 1075* Moved FIRST/FOLLOW computations to a separate class LL1Analyzer to 1076 tidy up. 1077 1078* ANTLR lexer allowed octal escapes, but they didn't work. ;) Rather than 1079 fix, I'm removing. Use '\uxxxx' to get even 8 bit char values: \u00xx. 1080 1081December 29, 2007 1082 1083* Fixed ANTLR-206. I wasn't avoiding analyzing decisions in 1084 left-recursive rules. 1085 1086* Had to add hetero arg to all tokenRef*() templates. Added _last 1087 local var to track last child so we can do replaceChildren() during 1088 AST rewrite mode for tree grammars. Should be useful later for .text 1089 property. Ack, hetero arg is on lots of templates. :( Moved 1090 ruleCleanUp() template into ASTTreeParser and ASTParser groups. 1091 1092* added noRewrite() template (to Java.stg) so we can insert code during 1093 rewrite mode to return original tree if no rewrite. Might be useful 1094 for token rewrites later. For templates too? 1095 1096* Had to add if !rewriteMode around tree construction in tree parser 1097 templates. 1098 1099* Harald Muller pointed out that we need to use <initValue(attr.type)> 1100 in our tests for null token/rule property references. For int types 1101 we need 0 not null. (p!=null?p.line:0). Changed scopeAttributeRef, 1102 ruleLabelRef. Also changed the known typed attributes like 1103 lexerRuleLabelPropertyRef_line to yield 0 upon null rule ref to 1104 be consistent with case when we don't know the type. Fixes ANTLR-195. 1105 Added testTypeOfGuardedAttributeRefIsCorrect test and reset expected 1106 output for 13 tests that now "fail". 1107 1108December 28, 2007 1109 1110* added polydiff example (Java target) 1111 1112* added "int" property for token and lexer rule refs. super convenient. E.g., 1113 a : b=INT {int x = $b.int;} ; 1114 1115December 27, 2007 1116 1117* Changed -Xnoinlinedfa to -Xmaxinlinedfastates m where m is 1118 maximum number of states a DFA can have before ANTLR avoids 1119 inlining it. Instead, you get a table-based DFA. This 1120 affectively avoids some acyclic DFA that still have many states 1121 with multiple incident edges. The combinatorial explosion smacks 1122 of infinite loop. Fixes ANTLR-130. 1123 1124* [...] are allowed in args now but ] must be escaped as \]. E.g., 1125 a[String[\] ick, int i] : ... ; 1126 And calling a rule: foo[x[i\], 34] 1127 Fixes ANTLR-140. 1128 1129* Fixed ANTLR-105. Target.getTargetStringLiteralFromANTLRStringLiteral() 1130 escaped " that were already escaped. 1131 1132* target's can now specify how to encode int as char escape. Moved 1133 DFA.encodeIntAsCharEscape to Target. 1134 1135* Bug in runtime.DFA. If a special state (one with predicate) failed, it 1136 tried to continue (causing out of range exception due to state = -1) 1137 instead of reporting error. 1138 1139* If -dfa with combined grammar T.g, builds T.dec-*.dot and TLexer.dec-*.dot 1140 1141* Fix ANTLR-165. 1142 Generate TParser.java and TLexer.java from T.g if combined, else 1143 use T.java as output regardless of type. 1144 BACKWARD INCOMPATIBILITY since file names change. 1145 I changed the examples-v3/java to be consistent. Required XML.g -> 1146 XMLLexer.java and fuzzy/Main.java change. 1147 1148* Fix ANTLR-169. Deletes tmp lexer grammar file. 1149 1150December 25, 2007 1151 1152* Fixed ANTLR-111. More unit tests in TestAttributes. 1153 1154December 25, 2007 1155 1156* Dangling states ("decision cannot distinguish between alternatives 1157 for at least one input sequence") is now an error not a warning. 1158 1159* Added sample input sequence that leads to dangling DFA state, one 1160 that cannot reach an accept state. ANTLR ran into a case where 1161 the same input sequence reaches multiple locations in the NFA 1162 (and so not nondeterministic), but analysis ran out of further 1163 NFA states to look for more input. Commonly at EOF target states. 1164 Now says: 1165 1166 error(202): CS.g:248:95: the decision cannot distinguish between alternative(s) 1,2 for input such as "DOT IDENTIFIER EOF" 1167 1168 Also fixed bug where dangling states did not resolve to stop states. 1169 1170* Fixed ANTLR-123 1171 1172December 17-21, 2007 1173 1174* k=1 doesn't prevent backtracking anymore as in 1175 (options {k=1;}:'else' statement)? 1176 if backtrack=true for overall grammar. Set to false in subrule. 1177 1178* Optimized the analysis engine for LL(1). Doesn't attempt LL(*) unless 1179 LL(1) fails. If not LL(1) but autobacktracking but no other kind of 1180 predicate, it also avoids LL(*). This is only important for really 1181 big 4000 line grammars etc... 1182 1183* Lots of code clean up 1184 1185December 16, 2007 1186 1187* Yet more Kay pair programming. Saved yet more RAM; 15% by 1188 wacking NFA configurations etc in each DFA state after DFA construction. 1189 1190* Overall we drop from 2m49s to 1m11s for a huge 4000 line TSQL grammar 1191 with k=*. Only needs -Xconversiontimeout 2000 now not 1192 -Xconversiontimeout 5000 too. With k=1, it's 1m30s down to 40s. 1193 1194December 15, 2007 1195 1196* Working with Kay Roepke, we got about 15% speed improvement in 1197 overall ANTLR exec time. Memory footprint seems to be about 50% 1198 smaller. 1199 1200December 13-14, 2007 1201 1202* I abort entire DFA construction now when I see recursion in > 1 alt. 1203 Decision is non-LL(*) even if some pieces are LL(*). Safer to bail 1204 out and try with fixed k. If user set fixed k then it continues because 1205 analysis will eventually terminate for sure. If a pred is encountered 1206 and k=* and it's non-LL(*), it aborts and retries at k=1 but does NOT 1207 emit an error. 1208 1209* Decided that recursion overflow while computing a lookahead DFA is 1210 serious enough that I should bail out of entire DFA computation. 1211 Previously analysis tried to keep going and made the rules about 1212 how analysis worked more complicated. Better to simply abort when 1213 decision can't be computed with current max stack (-Xm option). 1214 User can adjust or add predicate etc... This is now an error 1215 not a warning. 1216 1217* Recursion overflow and unreachable alt is now a fatal error; no code gen. 1218 The decision will literally not work. 1219 1220* Cleaned up how DFA construction/analysis aborts due to non-LL(*) and 1221 overflow etc... Throws exceptions now, which cleans up a bunch of IF 1222 checks etc... Very nice now. Exceptions: 1223 analysis/AnalysisRecursionOverflowException.java 1224 analysis/AnalysisTimeoutException.java 1225 analysis/NonLLStarDecisionException.java 1226 1227* ErrorManager.grammarWarning() counted them as errors not warnings. 1228 1229* Unreachable alt warnings are now errors. 1230 1231* The upshot of these changes is that I fixed ANTLR-178 and did 1232 lots of refactoring of code handling analysis failure. 1233 1234December 11, 2007 1235 1236* Could not deal with spaces, oddly enough in arg lists: 1237 grammar Bad; 1238 a : A b["foo", $A.text] ; 1239 b[String x, String y] : C ; 1240 1241October 28, 2007 1242 1243* Made ANTLR emit a better error message when it cannot write the 1244 implicit lexer file from a combined grammar. Used to say "cannot open 1245 file", now says "cannot write file" and gives backtrace. 1246 1247September 15, 2007 1248 1249add getCharStream to Lexer. 1250 1251September 10, 2007 1252 1253* Added {{...}} forced action executed even during backtracking. 1254 1255September 9, 2007 1256 1257* r='string' in lexer got a duplicate label definition. 1258 1259August 21, 2007 1260 1261* $scope::variable refs now check for empty stack so that expr == null if 1262 $scope has an empty stack. Works for $scope[...]::variable too. Nice! 1263 1264August 20, 2007 1265 1266* Added reset() to CommonTreeNodeStream, token stream too 1267 1268* Made refs to rule/token properties use ?: to avoid null ptr exception. 1269 $label.st now is label!=null?label.st:null. Updated TestAttributes. 1270 This is useful not only for optional rule/token refs, but also during 1271 error recovery. If ID is not matched, $ID.text won't cause a null ptr. 1272 1273August 20, 2007 1274* Fixed ANTLR-177: hashCode/equals not consistent for label 1275 Fixed bug where Rule was compared to string; introduced from dev branch 1276 1277August 15, 2007 -- Got rough draft of the grammar import working. 1278 Submit to dev and then integrate into mainline. 1279 1280 All file changes/additions: 1281 1282 README.txt # edit 1283 CHANGES.txt # add 1284 Factored out the changes from the readme. 1285 1286 runtime/Java/src/org/antlr/runtime/BaseRecognizer.java # edit 1287 runtime/Java/src/org/antlr/runtime/DFA.java # edit 1288 runtime/Java/src/org/antlr/runtime/Lexer.java # edit 1289 runtime/Java/src/org/antlr/runtime/Parser.java # edit 1290 runtime/Java/src/org/antlr/runtime/debug/DebugParser.java # edit 1291 runtime/Java/src/org/antlr/runtime/tree/TreeParser.java # edit 1292 Factored state fields into RecognizerSharedState 1293 object. You will see a lot of things like 1294 state.errorRecovery = false; 1295 runtime/Java/src/org/antlr/runtime/RecognizerSharedState.java # add 1296 Shares all recognizer state variables including lexer even though 1297 these are superfluous to parsers and tree parsers. There 1298 was a casting issue that I could not resolve. 1299 1300 src/org/antlr/Tool.java # edit 1301 Broke a part Grammar.setGrammarContent() into 1302 parseAndBuildAST() and analyzeGrammar() to make the grammar 1303 import work. I needed to be able to look at the trees for 1304 imported grammars before analyzing them and building DFA. Added 1305 use of the CompositeGrammar object and handling of multiple 1306 delegate grammars. Changed decision DFA DOT file names to 1307 include the grammar name. 1308 1309 src/org/antlr/analysis/DFA.java # edit 1310 Just tweaked to use generics, updated a comment. 1311 1312 src/org/antlr/analysis/DecisionProbe.java # edit 1313 Just tweaked to use generics. 1314 1315 src/org/antlr/analysis/NFA.java # edit 1316 NFA now span multiple grammars and so I moved the NFAs state 1317 tracking to the composite grammar object. 1318 1319 src/org/antlr/analysis/NFAState.java # edit 1320 Added some null checking and made a field public. 1321 1322 src/org/antlr/analysis/NFAToDFAConverter.java # edit 1323 Changed a method call to directly access a field. 1324 1325 src/org/antlr/analysis/RuleClosureTransition.java # edit 1326 Instead of using a rule index, which does not span multiple 1327 grammars, the transition object now attracts a pointer to 1328 the actual Rule definition object. 1329 1330 src/org/antlr/analysis/SemanticContext.java # edit 1331 Tweaked to use a field instead of a method 1332 1333 src/org/antlr/codegen/ActionTranslator.g # edit 1334 src/org/antlr/codegen/ActionTranslatorLexer.java # edit 1335 Tweaked to use new runtime and they changed method name. 1336 1337 src/org/antlr/codegen/CodeGenerator.java # edit 1338 Tweaked comments. 1339 1340 src/org/antlr/codegen/codegen.g # edit 1341 Added import grammar syntax and altered rule atom to pass a 1342 scope around so that grammar.rule works. Caution this 1343 feature is used internally by ANTLR and is not meant to be 1344 used by users at this point. 1345 1346 src/org/antlr/codegen/templates/ANTLRCore.sti # edit 1347 Added scope to all ruleref template interfaces. 1348 1349 src/org/antlr/codegen/templates/Java/Java.stg # edit 1350 Grammars can now import other grammars, which I implemented 1351 using a delegation pointer to the other grammar(s). So if 1352 grammar A imports grammars B and C, then the generated 1353 recognizer for A must have delegation pointers to BParser 1354 and CParser objects. These are now fields: 1355 1356 // delegates 1357 <grammar.delegates: 1358 {g|public <g.name>Lexer <g:delegateName()>;}; separator="\n"> 1359 1360 Also, B and C must have back pointers to the delegator so 1361 that they can refer to rules that have been overridden. 1362 This is a mechanism akin to static inheritance: 1363 1364 // delegators 1365 <grammar.delegators: 1366 {g|public <g.name>Lexer <g:delegateName()>;}; separator="\n"> 1367 1368 This file also has a lot of changes so that state variables 1369 now are state.backtracking instead of the implied 1370 this.backtracking. 1371 1372 The file also refers to grammar.delegatedRules attribute 1373 which is the list of Rule objects for which you must 1374 generate manual delegation. This amounts to a stub whereby 1375 rule foo's method foo() simply calls X.foo() if foo is not 1376 defined inside the delegator. 1377 1378 You will notice that the ruleref templates now take a scope 1379 so that I can have implicit rule Tokens referred to 1380 delegate.Tokens rule in a delegate grammar. This is the way 1381 I do lexer grammar imports. 1382 1383 I added a template called delegateName which uses the 1384 grammar name to compute a delegate name if the user does not 1385 specify a label in the import statement such as: 1386 1387 import x=X; 1388 1389 Oh, note that rule reference templates all receive a Rule 1390 object now instead of the simple rule name as the 'rule' 1391 attribute. You will see me doing <rule.name> instead of 1392 <name> now. 1393 1394 src/org/antlr/codegen/templates/Java/Dbg.stg # edit 1395 Changes mirroring the constructor and field stuff from 1396 Java.stg. Part of this is a cut and paste because of a bug 1397 in ST. 1398 1399 src/org/antlr/codegen/templates/Java/AST.stg # edit 1400 src/org/antlr/codegen/templates/Java/ASTParser.stg # edit 1401 src/org/antlr/codegen/templates/Java/ASTTreeParser.stg # edit 1402 Just added the scope attribute. 1403 1404 src/org/antlr/test/BaseTest.java # edit 1405 Added functionality to support testing composite grammars. 1406 execLexer() 1407 1408 src/org/antlr/test/TestAttributes.java # edit 1409 Tweak to deal with shared recognizer state. 1410 1411 src/org/antlr/test/TestCompositeGrammars.java # add 1412 Start of my unit tests. 1413 1414 src/org/antlr/tool/CompositeGrammar.java # add 1415 src/org/antlr/tool/CompositeGrammarTree.java # add 1416 Tracks main grammar and all delegate grammars. Tracks unique 1417 NFA state numbers and unique token types. This keeps a tree 1418 of grammars computed from the import/delegation chain. When 1419 you want to look up a rule, it starts at the root of the 1420 tree and does a pre-order search to find the rule. 1421 1422 src/org/antlr/tool/ActionAnalysis.g # edit 1423 src/org/antlr/tool/ActionAnalysisLexer.java # edit 1424 1425 src/org/antlr/tool/AttributeScope.java # edit 1426 Updated to use generics in one place. 1427 1428 src/org/antlr/tool/DOTGenerator.java # edit 1429 Updated to indicate when nonlocal rules are referenced. 1430 1431 src/org/antlr/tool/ErrorManager.java # edit 1432 Added some error messages for import grammars; I need more. 1433 1434 src/org/antlr/tool/FASerializer.java # edit 1435 Tweaked to use a field not method. 1436 1437 src/org/antlr/tool/Grammar.java # edit 1438 This is where most of the meat is for the grammar import 1439 stuff as you can imagine. I factored out the token type 1440 tracking into the CompositeGrammar object. I added code to 1441 the addArtificialMatchTokensRule method so that it includes 1442 references to all delegate lexer Tokens rules. Altered the 1443 rule lookup stuff so that it knows about delegate grammars. 1444 1445 src/org/antlr/tool/GrammarAST.java # edit 1446 src/org/antlr/tool/GrammarAnalysisAbortedMessage.java # edit 1447 src/org/antlr/tool/GrammarReport.java # edit 1448 src/org/antlr/tool/NonRegularDecisionMessage.java # edit 1449 Made enclosing rule visible as field. 1450 1451 src/org/antlr/tool/GrammarSanity.java # edit 1452 General cleanup and addition of generics. 1453 1454 src/org/antlr/tool/Interpreter.java # edit 1455 Reference fields instead of methods. 1456 1457 src/org/antlr/tool/NFAFactory.java # edit 1458 General cleanup and use of Rule object instead of rule 1459 index. 1460 1461 src/org/antlr/tool/NameSpaceChecker.java # edit 1462 A little bit of cleanup and changes to use either the local 1463 or globally visible rule. Added code to check that scopes 1464 are valid on scoped rule references. again this is an 1465 internal feature, not to be used by users. 1466 1467 src/org/antlr/tool/RandomPhrase.java # edit 1468 Tweaked. 1469 1470 src/org/antlr/tool/Rule.java # edit 1471 Added field imported. Removed some unused methods by 1472 commenting them out. Made toString() more expressive. 1473 1474 src/org/antlr/tool/antlr.g # edit 1475 src/org/antlr/tool/antlr.print.g # edit 1476 src/org/antlr/tool/assign.types.g # edit 1477 src/org/antlr/tool/buildnfa.g # edit 1478 src/org/antlr/tool/define.g # edit 1479 Added syntax for import statement. assign.types.g is the 1480 grammar that invokes Grammar.importGrammar(). 1481 1482 src/org/antlr/tool/templates/messages/languages/en.stg # edit 1483 Added error messages. 1484 1485 Added 1486 1487 CHANGES.txt 1488 runtime/Java/src/org/antlr/runtime/RecognizerSharedState.java 1489 src/org/antlr/test/TestCompositeGrammars.java 1490 src/org/antlr/tool/CompositeGrammar.java 1491 src/org/antlr/tool/CompositeGrammarTree.java 1492 14933.0.1 - August 13, 2007 1494 1495[See target pages on the wiki for more information on the non-Java targets] 1496 1497August 7, 2007 1498 1499* added escaping of double quotes in DOTTreeGenerator 1500 1501July 22, 2007 1502 1503* fixed dynamic scope implementation in lexers. They were not creating new scope 1504 entries on the stack. Unsupported feature! 1505 1506July 30, 2007 1507 1508* float return values were initalized to 0.0 not 0.0f in java. 1509 1510July 28, 2007 1511 1512* Sam Ellis points out an init var bug in ANTLRReaderStream. 1513 1514July 27, 2007 (done in dev branch) 1515 1516* Moved token type index stuff from CommonTreeNodeStream to TreeWizard 1517 1518* Added getChildren to BaseTree. 1519 1520* Added heterogeneous tree functionality; rewrite for parser/tree parser 1521 and auto AST constr. for parser. 1522 1523 org/antlr/runtime/tree/RewriteRuleElementStream.java 1524 org/antlr/runtime/tree/RewriteRuleNodeStream.java 1525 org/antlr/runtime/tree/RewriteRuleTokenStream.java 1526 Renamed method next() and re-factor things to have more 1527 specific methods: nextToken, nextNode, nextTree. 1528 1529 codegen/codegen.g 1530 Updated to include new <NodeType> AST structure for 1531 token references. Pushed hereto attribute into 1532 all tokenRef* templates. 1533 codegen/templates/Java/AST.stg 1534 Factored out a few templates: 1535 createImaginaryNode(tokenType,hetero,args) 1536 createRewriteNodeFromElement(token,hetero,args) 1537 Converted a lot of stream next() calls to more specific 1538 nextToken, nextNode, nextTree per above. 1539 codegen/templates/Java/ASTParser.stg 1540 Added createNodeFromToken template and re-factored creation 1541 sites to use that template. Added hetero attribute. 1542 codegen/templates/Java/ASTTreeParser.stg 1543 Added createRewriteNodeFromElement template and re-factored. 1544 1545 test/TestHeteroAST.java 1546 New file. Unit tests to test new hetero tree construction. 1547 test/TestRewriteAST.java 1548 Fixed test. Nil single-node trees no longer return nil; 1549 They return null. 1550 1551 tool/ErrorManager.java 1552 tool/templates/messages/languages/en.stg 1553 Added error message: 1554 HETERO_ILLEGAL_IN_REWRITE_ALT(arg) ::= 1555 "alts with rewrites can't use heterogeneous types left of ->" 1556 1557 tool/antlr.g 1558 tool/antlr.print.g 1559 tool/assign.types.g 1560 tool/buildnfa.g 1561 tool/define.g 1562 Added syntax for <NodeType> to token references. 1563 Altered AST structure rippled through different phases. 1564 1565July 24, 2007 1566 1567* Deleted DoubleLinkTree.java; CommonTree does that now. 1568 1569July 23, 2007 1570 1571* template group outputFile; changed rewrite arg to rewriteMode. 1572 1573* added rewrite mode for tree parser build AST. 1574 1575July 22, 2007 1576 1577* Kay fixed dynamic scope implementation in lexers. They were not 1578 creating new scope entries on the stack. This is an UNSUPPORTED feature. 1579 1580* added getParent and getChildIndex to TreeAdaptor. Added 1581 implementation to CommonTree. It's just too useful having those 1582 parent and child indexes available for rewriting etc... I tried 2x 1583 to make an implementation of tree rewriting w/o this and the 1584 constraints just made it too expensive and complicated. Have to 1585 update adaptors to set parent, child index values. Updated Tree 1586 interface and BaseTree also. Should only affect target developers 1587 not users. Well, unless they impl Tree. 1588 1589* dupNode (via ctor) of CommonTree didn't copy start/stop token indexes. 1590 1591TARGET DEVELOPERS WARNING -- AST.stg split with some functionality 1592 going into ASTParser.stg then I added 1593 ASTTreeParser.stg. CodeGenerator 1594 assumes new subgroups exist. 1595 1596July 20, 2007 1597 1598* Added AST construction for tree parsers including -> rewrite rules. 1599 Rewrite mode (rewrite=true) alters the tree in place rather than 1600 constructing a whole new tree. Implementation notes: 1601 1602 org/antlr/runtime/tree/Tree.java 1603 Add methods for parent and child index functionality. 1604 Also added freshenParentAndChildIndexes() which you can use 1605 to ensure that all double linking is set up right after you 1606 manipulate the tree manually. The setChild preteens etc. do 1607 the proper thing so you shouldn't need this. 1608 Added replaceChildren() to support tree rewrite mode in tree parsers 1609 org/antlr/runtime/tree/BaseTree.java 1610 Updated to set parent and child index stuff. Added replaceChildren 1611 method etc... It still only has a list of children as sole field 1612 but calls methods that subclasses can choose to implement such as 1613 CommonTree. 1614 org/antlr/runtime/tree/CommonTree.java 1615 Added parent and childIndex fields to doubly link. 1616 org/antlr/runtime/tree/TreeAdaptor.java 1617 Added methods for new parent and child index functionality. 1618 Also added method for rewrite mode in tree parsers: 1619 replaceChildren(Object parent, int startChildIndex, 1620 int stopChildIndex, Object t); 1621 Added setChild and deleteChild methods 1622 org/antlr/runtime/tree/BaseTreeAdaptor.java 1623 Moved dupTree here from BaseTree. 1624 Updated rulePostProcessing to deal with parent and child index. 1625 Added setChild and deleteChild implementations 1626 org/antlr/runtime/tree/CommonTreeAdaptor.java 1627 Added methods to deal with the parent and child index for a node. 1628 1629 org/antlr/runtime/tree/CommonTreeNodeStream.java 1630 Removed token type index and method fillReverseIndex etc... 1631 Probably will move into the tree wizard in the future. 1632 Changed call/seek stack implementation to use IntArray 1633 Added replaceChildren interface. 1634 org/antlr/runtime/tree/TreeNodeStream.java 1635 Added replaceChildren. 1636 org/antlr/runtime/tree/UnBufferedTreeNodeStream.java 1637 Added replaceChildren method but no implementation 1638 1639 codegen/templates/ANTLRCore.sti 1640 Changed rewrite to a better name: rewriteMode 1641 Added tree level argument to alt, tree so that auto AST 1642 construction can occur while recognizing in tree parsers. 1643 1644 codegen/templates/Java/AST.stg 1645 Split template group: added two subclasses to handle different 1646 functionality for normal parsing and tree parsing + AST 1647 construction. Tree parsers default behavior is to dup tree 1648 not construct another. Added ASTParser.stg and 1649 ASTTreeParser.stg to handle auto AST construction during 1650 recognition for the two different parser types. I just copied 1651 the token, rule, set, wildcard templates to the subclasses. 1652 The rewrite templates are still in AST.stg. I factored out the 1653 node creation so that the same rewrite templates can be used 1654 for both parsing and tree parsing. 1655 1656 codegen/templates/Java/ASTParser.stg 1657 The templates needed to build trees with auto construction 1658 during parsing. 1659 codegen/templates/Java/ASTTreeParser.stg 1660 The templates needed to build trees with auto construction 1661 during tree parsing. 1662 codegen/templates/Java/Java.stg 1663 genericParser now has rewriteElementType (Note or Token) so 1664 that the rewrite streams know what kind of elements are inside 1665 during rewrite rule tree construction. 1666 codegen/templates/Java/ST.stg 1667 rewrite attribute name change to rewriteMode 1668 1669 org/antlr/runtime/debug/DebugTreeAdaptor.java 1670 org/antlr/runtime/debug/DebugTreeNodeStream.java 1671 Updated to handle new interfaces 1672 1673 test/BaseTest.java 1674 Added test rig update to handle AST construction by tree parsers. 1675 All tree construction runs automatically test sanity of parent 1676 and child indexes. 1677 test/TestTreeGrammarRewriteAST.java 1678 test/TestTreeNodeStream.java 1679 test/TestTrees.java 1680 new file; tests the new parent and child index stuff in trees. 1681 1682July 19, 2007 1683 1684* implemented new unique ID; GC was causing non unique hash codes. Debugging 1685 tree grammars was messing up. 1686 1687* got tree rewrites working in tree grammars. It builds a completely new 1688 tree from old tree; i.e., you get two trees in memory. W/o a rewrite 1689 rule, the input for that rule is duplicated and returned. -> w/o elements 1690 to the right means don't return anything; i.e., delete. Ooops...way 1691 harder than I thought. Real implementation notes above. 1692 1693INCOMPATIBILITY WARNING -- templates have changed; must regen output from 1694 grammars. Runtime libraries have also changed. 1695 Debug event listener interface has changed also. 1696 1697July 17, 2007 1698 1699* Added line/charposition to node socket events and event dump so 1700 we have more info during tree parsing. Only works if your 1701 tree adaptor returns a value Token object from getToken(treenode) 1702 with line/col set. Refactored consumeNode/LN to use deserializeNode(). 1703 1704* Fixed mismatched tree node exceptions; for imaginary nodes, it said 1705 "missing null". Now prints the token type we found. 1706 1707* Cleaned up exception stuff. MismatchedTreeNodeException was setting 1708 line/col, but only RecognitionException should do that. 1709 1710* If imaginary token gets a mismatch, there is no line info. Search 1711 backwards in stream if input node stream supports to find last 1712 node with good line/col info. E.g., 1713 1714ANTLRv3Tree.g: node from after line 156:72 mismatched tree node: EOA expecting <UP> 1715 1716 which used to be: 1717 1718ANTLRv3Tree.g: node from line 0:0 mismatched tree node: null expecting <UP> 1719 1720* mismatched tree node exceptions were not sent to the debug event stream. 1721 Due to a type being slightly different on recoverFromMismatchedToken() 1722 in DebugTreeParser. Was calling BaseRecognizer version not subclass. 1723 Now we get: 1724 1725 9459: Recognition exception MismatchedTreeNodeException(0!=0) 1726 1727* List labels were not allowed as root nodes in tree rewrites like 1728 ^($listlabel ...). Had to add a template to AST.stg: 1729 1730 /** Gen ^($label ...) where label+=... */ 1731 rewriteTokenListLabelRefRoot ::= rewriteTokenLabelRefRoot 1732 1733 1734July 16, 2007 1735 1736* fixed nextNode in RewriteRuleSubtreeStream was dup'ing too much, 1737 screwing up debug event stream. Also there was a bug in how 1738 the rewrite tree stream stuff decided to dup nodes. 1739 1740* fixed bug in LT for tree parsing; text was not transmitted properly; 1741 only single words worked. 1742 1743* made decision for rule put line/col on colon not first token of first alt. 1744 1745* remote ProxyToken now emits token index for easier debugging when looking 1746 at AW's event stream. For example, the @5 here is the token index: 1747 1748 31 Consume hidden [ /<64>,channel=99,30:7, @5] 1749 1750* same is true for consume nodes now: 1751 1752 25586 Consume node [')'/, <44>, 4712040,@1749] 25 1753 1754 When debugging tree parsers, it helps to track errors when you know 1755 what corresponding input symbol created this tree node. 1756 1757* Changed debug events associated with trees quite a bit. Passes nodes around 1758 now rather than text, type, unique IDs etc... Mostly affects internal stuff. 1759 Target developers will have some work in their runtime to do to match 1760 this change. :( BUT, there is only a slight tweak in the Dbg.stg 1761 and ASTDbg.stg templates. 1762 Interface just didn't make sense as is. If you turn on debugging, and 1763 want to track a node creation, you want the node pointer not its ID, 1764 text, etc... 1765 Added ProxyTree for passing across socket. Has line/charpos and tokenIndex 1766 1767July 15, 2007 1768 1769* added null ptr protection in CommonTreeAdaptor. 1770 1771July 14, 2007 1772 1773* null child in TreeAdaptor does nothing now. Changed interface and 1774 implementation. Changed DebugTreeAdaptor to not fire events on null add 1775 as well. 1776 1777July 12, 2007 1778 1779* added get method for the line/col to DFA map in Grammar.java 1780 1781July 7, 2007 1782 1783* fixed wrong order of test for exceptions in Lexer.getErrorMessage() 1784 1785June 28, 2007 1786 1787* Added ability to set the port number in the constructor for the debug parser. 1788 1789June 5, 2007 1790 1791* Changed (hidden) option -verbose to -Xnfastates; this just prints out the NFA states along each nondeterministic path for nondeterminism warnings. 1792 1793May 18, 2007 1794 1795* there were some dependencies with org.antlr.* that I removed from 1796 org.antlr.runtime.* 1797 17983.0 final - May 17, 2007 1799 1800May 14, 2007 1801 1802* Auto backtracking didn't work with ! and ^ suffixes on first element 1803 of an alt. 1804 1805* Auto backtracking didn't work with an action as first element. 1806 1807May 10, 2007 1808 1809* turn off the warning about no local messages: 1810 no such locale file org/antlr/tool/templates/messages/languages/ru.stg retrying with English locale 1811 1812May 5, 2007 1813 1814* moving org.antlr.runtime to runtime/Java/src/org/... Other target 1815 source / libs are under runtime/targetname. 1816 1817May 4, 2007 1818 1819* You could not use arguments on a token reference that was a route in a 1820 tree rewrite rule like -> ^(ID[args] ...). 1821 1822May 3, 2007 1823 1824* Fixed ANTLR-82. Actions after the root were considered part of 1825 an optional child. They were not always executed. Required a change 1826 to the ANTLRCore.sti interface for tree() template. 1827 1828May 2, 2007 1829 1830* Fixed ANTLR-117. Wasn't building decisions properly for subrules in 1831 syntactic predicates. 1832 1833April 22, 2007 1834 1835* Made build.xml ref all jars in antlr lib. Thanks to Miguel Ping. 1836 1837* Fixed ANTLR-11 1838 1839* Now labels on ranges and such in lexer work properly. 1840 1841* ActionAnalysisLexer was in wrong package. 1842 1843April 21, 2007 1844 1845* Pushing a huge update that fixes: 1846 http://www.antlr.org/browse/ANTLR-112 1847 http://www.antlr.org/browse/ANTLR-110 1848 http://www.antlr.org/browse/ANTLR-109 1849 http://www.antlr.org/browse/ANTLR-103 1850 http://www.antlr.org/browse/ANTLR-97 1851 http://www.antlr.org/browse/ANTLR-113 1852 http://www.antlr.org/browse/ANTLR-66 1853 http://www.antlr.org/browse/ANTLR-98 1854 http://www.antlr.org/browse/ANTLR-24 1855 http://www.antlr.org/browse/ANTLR-114 1856 http://www.antlr.org/browse/ANTLR-5 1857 http://www.antlr.org/browse/ANTLR-6 1858 1859 Basically, I gutted the way AST rewrites work. MUCH better. 1860 1861* Fixed lots of little label issues in the lexer. Couldn't do x+=ID 1862 in lexer, for example. Fixed ANTLR-114, ANTLR-112 1863 1864* Isolated EOT transition in lexer generated dangling else clause. 1865 Fixed ANTLR-113. 1866 1867April 17, 2007 1868 1869* Fixed a major problem with gated semantic predicates. Added more 1870 unit tests. 1871 1872* Fixed bug in cyclic DFA with syntactic predicates. Wasn't rewinding 1873 properly. Further, mark() in token stream did not fill buffer so 1874 when you rewound back to last marker index was -1 not 0. At same time 1875 I fixed ANTLR-103. Syn preds evaluated only once now. 1876 1877* Altered code gen file writing so it writes directly to a file 1878 instead of building a big string and then writing that out. Should 1879 be faster and much less memory intensive. 1880 1881* Fixed so antlr writes files to correct location again. See: 1882 1883http://www.antlr.org/wiki/pages/viewpage.action?pageId=1862 1884 18853.0b7 - April 12, 2007 1886 1887April 10, 2007 1888 1889* Allows -> {...} actions now when building ASTs. Fixed ANTLR-14. 1890 1891* Allows ! on sets and wildcard now during output=AST option. Fixed ANTLR-17. 1892 1893* Fixed ANTLR-92 bug. Couldn't use sets with -> tree construction. 1894 1895* No lexer rule for a token type is now a warning. 1896 1897* Fixed set labels in lexer; ANTLR-60 bug 1898 1899* Fixed problem with duplicate state variable definitions in switch-case 1900 1901April 9, 2007 1902 1903* Gated predicates didn't work properly in cyclic DFA. 1904 1905April 7, 2007 1906 1907* Couldn't have more than one set per rule it seems. Fixed. 1908 1909April 3, 2007 1910 1911* Fix a problem in my unused label optimization. Added new 1912 pass over actions to examine them. 1913 1914* RuleReturnScope has method back: 1915 /** Has a value potentially if output=template; Don't use StringTemplate 1916 * type as it then causes a dependency with ST lib. 1917 */ 1918 public Object getTemplate() { return null; } 1919 1920March 30, 2007 1921 1922* Fixed ANTLR-8. Labels to rules w/o return values caused compile errors. 1923 1924* Fixed ANTLR-89; semantic predicates in lexer sometimes 1925 caused exception in code gen. 1926 1927* Fixed ANTLR-36; remove runtime dependency with ST 1928 1929March 29, 2007 1930 1931* Over last few days, I've gutted how ANTLR handles sets of chars or 1932 tokens. I cleaned up a lot of stuff in the grammars and added lots 1933 of unit tests. 1934 1935March 26, 2007 1936 1937* CommonTreeNodeStream didn't push correctly; couldn't handle very 1938 deeply nested trees. 1939 1940* Fixed bug that E : 'a' 'b' ; made E be seen as an alias of 'a'. 1941 1942March 22, 2007 1943 1944* Working with Egor Ushakov from Sun Optimization / NetBeans team I 1945 made all the Java lexer transition tables static w/o screwing up 1946 ability to reference semantic predicates etc... Only changed Java.stg 1947 1948* cached text string in CommonToken.getText(); saves on repeated calls; 1949 Java mode. 1950 1951* made all generated methods final; saves a few percent speed according to 1952 Egor Ushakov (Java only). 1953 1954* removed most assignments from each lexer rule and even the Lexer.emit() 1955 call! All done in nextToken now. Saves on code gen size and a wee bit of 1956 execution speed probably. Variables became fields: type, channel, line, 1957 etc... Now emit() needs no args even. Again, Egor helped on this. 1958 1959March 17, 2007 1960 1961* Jonathan DeKlotz updated C# templates to be 3.0b6 current 1962 1963March 14, 2007 1964 1965* Manually-specified (...)=> force backtracking eval of that predicate. 1966 backtracking=true mode does not however. Added unit test. 1967 1968March 14, 2007 1969 1970* Fixed bug in lexer where ~T didn't compute the set from rule T. 1971 1972* Added -Xnoinlinedfa make all DFA with tables; no inline prediction with IFs 1973 1974* Fixed http://www.antlr.org/browse/ANTLR-80. 1975 Sem pred states didn't define lookahead vars. 1976 1977* Fixed http://www.antlr.org/browse/ANTLR-91. 1978 When forcing some acyclic DFA to be state tables, they broke. 1979 Forcing all DFA to be state tables should give same results. 1980 1981March 12, 2007 1982 1983* setTokenSource in CommonTokenStream didn't clear tokens list. 1984 setCharStream calls reset in Lexer. 1985 1986* Altered -depend. No longer printing grammar files for multiple input 1987 files with -depend. Doesn't show T__.g temp file anymore. Added 1988 TLexer.tokens. Added .h files if defined. 1989 1990February 11, 2007 1991 1992* Added -depend command-line option that, instead of processing files, 1993 it shows you what files the input grammar(s) depend on and what files 1994 they generate. For combined grammar T.g: 1995 1996 $ java org.antlr.Tool -depend T.g 1997 1998 You get: 1999 2000 TParser.java : T.g 2001 T.tokens : T.g 2002 T__.g : T.g 2003 2004 Now, assuming U.g is a tree grammar ref'd T's tokens: 2005 2006 $ java org.antlr.Tool -depend T.g U.g 2007 2008 TParser.java : T.g 2009 T.tokens : T.g 2010 T__.g : T.g 2011 U.g: T.tokens 2012 U.java : U.g 2013 U.tokens : U.g 2014 2015 Handles spaces by escaping them. Pays attention to -o, -fo and -lib. 2016 Dir 'x y' is a valid dir in current dir. 2017 2018 $ java org.antlr.Tool -depend -lib /usr/local/lib -o 'x y' T.g U.g 2019 x\ y/TParser.java : T.g 2020 x\ y/T.tokens : T.g 2021 x\ y/T__.g : T.g 2022 U.g: /usr/local/lib/T.tokens 2023 x\ y/U.java : U.g 2024 x\ y/U.tokens : U.g 2025 2026 You have API access via org.antlr.tool.BuildDependencyGenerator class: 2027 getGeneratedFileList(), getDependenciesFileList(). You can also access 2028 the output template: getDependencies(). The file 2029 org/antlr/tool/templates/depend.stg contains the template. You can 2030 modify as you want. File objects go in so you can play with path etc... 2031 2032February 10, 2007 2033 2034* no more .gl files generated. All .g all the time. 2035 2036* changed @finally to be @after and added a finally clause to the 2037 exception stuff. I also removed the superfluous "exception" 2038 keyword. Here's what the new syntax looks like: 2039 2040 a 2041 @after { System.out.println("ick"); } 2042 : 'a' 2043 ; 2044 catch[RecognitionException e] { System.out.println("foo"); } 2045 catch[IOException e] { System.out.println("io"); } 2046 finally { System.out.println("foobar"); } 2047 2048 @after executes after bookkeeping to set $rule.stop, $rule.tree but 2049 before scopes pop and any memoization happens. Dynamic scopes and 2050 memoization are still in generated finally block because they must 2051 exec even if error in rule. The @after action and tree setting 2052 stuff can technically be skipped upon syntax error in rule. [Later 2053 we might add something to finally to stick an ERROR token in the 2054 tree and set the return value.] Sequence goes: set $stop, $tree (if 2055 any), @after (if any), pop scopes (if any), memoize (if needed), 2056 grammar finally clause. Last 3 are in generated code's finally 2057 clause. 2058 20593.0b6 - January 31, 2007 2060 2061January 30, 2007 2062 2063* Fixed bug in IntervalSet.and: it returned the same empty set all the time 2064 rather than new empty set. Code altered the same empty set. 2065 2066* Made analysis terminate faster upon a decision that takes too long; 2067 it seemed to keep doing work for a while. Refactored some names 2068 and updated comments. Also made it terminate when it realizes it's 2069 non-LL(*) due to recursion. just added terminate conditions to loop 2070 in convert(). 2071 2072* Sometimes fatal non-LL(*) messages didn't appear; instead you got 2073 "antlr couldn't analyze", which is actually untrue. I had the 2074 order of some prints wrong in the DecisionProbe. 2075 2076* The code generator incorrectly detected when it could use a fixed, 2077 acyclic inline DFA (i.e., using an IF). Upon non-LL(*) decisions 2078 with predicates, analysis made cyclic DFA. But this stops 2079 the computation detecting whether they are cyclic. I just added 2080 a protection in front of the acyclic DFA generator to avoid if 2081 non-LL(*). Updated comments. 2082 2083January 23, 2007 2084 2085* Made tree node streams use adaptor to create navigation nodes. 2086 Thanks to Emond Papegaaij. 2087 2088January 22, 2007 2089 2090* Added lexer rule properties: start, stop 2091 2092January 1, 2007 2093 2094* analysis failsafe is back on; if a decision takes too long, it bails out 2095 and uses k=1 2096 2097January 1, 2007 2098 2099* += labels for rules only work for output option; previously elements 2100 of list were the return value structs, but are now either the tree or 2101 StringTemplate return value. You can label different rules now 2102 x+=a x+=b. 2103 2104December 30, 2006 2105 2106* Allow \" to work correctly in "..." template. 2107 2108December 28, 2006 2109 2110* errors that are now warnings: missing AST label type in trees. 2111 Also "no start rule detected" is warning. 2112 2113* tree grammars also can do rewrite=true for output=template. 2114 Only works for alts with single node or tree as alt elements. 2115 If you are going to use $text in a tree grammar or do rewrite=true 2116 for templates, you must use in your main: 2117 2118 nodes.setTokenStream(tokens); 2119 2120* You get a warning for tree grammars that do rewrite=true and 2121 output=template and have -> for alts that are not simple nodes 2122 or simple trees. new unit tests in TestRewriteTemplates at end. 2123 2124December 27, 2006 2125 2126* Error message appears when you use -> in tree grammar with 2127 output=template and rewrite=true for alt that is not simple 2128 node or tree ref. 2129 2130* no more $stop attribute for tree parsers; meaningless/useless. 2131 Removed from TreeRuleReturnScope also. 2132 2133* rule text attribute in tree parser must pull from token buffer. 2134 Makes no sense otherwise. added getTokenStream to TreeNodeStream 2135 so rule $text attr works. CommonTreeNodeStream etc... now let 2136 you set the token stream so you can access later from tree parser. 2137 $text is not well-defined for rules like 2138 2139 slist : stat+ ; 2140 2141 because stat is not a single node nor rooted with a single node. 2142 $slist.text will get only first stat. I need to add a warning about 2143 this... 2144 2145* Fixed http://www.antlr.org/browse/ANTLR-76 for Java. 2146 Enhanced TokenRewriteStream so it accepts any object; converts 2147 to string at last second. Allows you to rewrite with StringTemplate 2148 templates now :) 2149 2150* added rewrite option that makes -> template rewrites do replace ops for 2151 TokenRewriteStream input stream. In output=template and rewrite=true mode 2152 same as before 'cept that the parser does 2153 2154 ((TokenRewriteStream)input).replace( 2155 ((Token)retval.start).getTokenIndex(), 2156 input.LT(-1).getTokenIndex(), 2157 retval.st); 2158 2159 after each rewrite so that the input stream is altered. Later refs to 2160 $text will have rewrites. Here's a sample test program for grammar Rew. 2161 2162 FileReader groupFileR = new FileReader("Rew.stg"); 2163 StringTemplateGroup templates = new StringTemplateGroup(groupFileR); 2164 ANTLRInputStream input = new ANTLRInputStream(System.in); 2165 RewLexer lexer = new RewLexer(input); 2166 TokenRewriteStream tokens = new TokenRewriteStream(lexer); 2167 RewParser parser = new RewParser(tokens); 2168 parser.setTemplateLib(templates); 2169 parser.program(); 2170 System.out.println(tokens.toString()); 2171 groupFileR.close(); 2172 2173December 26, 2006 2174 2175* BaseTree.dupTree didn't dup recursively. 2176 2177December 24, 2006 2178 2179* Cleaned up some comments and removed field treeNode 2180 from MismatchedTreeNodeException class. It is "node" in 2181 RecognitionException. 2182 2183* Changed type from Object to BitSet for expecting fields in 2184 MismatchedSetException and MismatchedNotSetException 2185 2186* Cleaned up error printing in lexers and the messages that it creates. 2187 2188* Added this to TreeAdaptor: 2189 /** Return the token object from which this node was created. 2190 * Currently used only for printing an error message. 2191 * The error display routine in BaseRecognizer needs to 2192 * display where the input the error occurred. If your 2193 * tree of limitation does not store information that can 2194 * lead you to the token, you can create a token filled with 2195 * the appropriate information and pass that back. See 2196 * BaseRecognizer.getErrorMessage(). 2197 */ 2198 public Token getToken(Object t); 2199 2200December 23, 2006 2201 2202* made BaseRecognizer.displayRecognitionError nonstatic so people can 2203 override it. Not sure why it was static before. 2204 2205* Removed state/decision message that comes out of no 2206 viable alternative exceptions, as that was too much. 2207 removed the decision number from the early exit exception 2208 also. During development, you can simply override 2209 displayRecognitionError from BaseRecognizer to add the stuff 2210 back in if you want. 2211 2212* made output go to an output method you can override: emitErrorMessage() 2213 2214* general cleanup of the error emitting code in BaseRecognizer. Lots 2215 more stuff you can override: getErrorHeader, getTokenErrorDisplay, 2216 emitErrorMessage, getErrorMessage. 2217 2218December 22, 2006 2219 2220* Altered Tree.Parser.matchAny() so that it skips entire trees if 2221 node has children otherwise skips one node. Now this works to 2222 skip entire body of function if single-rooted subtree: 2223 ^(FUNC name=ID arg=ID .) 2224 2225* Added "reverse index" from node to stream index. Override 2226 fillReverseIndex() in CommonTreeNodeStream if you want to change. 2227 Use getNodeIndex(node) to find stream index for a specific tree node. 2228 See getNodeIndex(), reverseIndex(Set tokenTypes), 2229 reverseIndex(int tokenType), fillReverseIndex(). The indexing 2230 costs time and memory to fill, but pulling stuff out will be lots 2231 faster as it can jump from a node ptr straight to a stream index. 2232 2233* Added TreeNodeStream.get(index) to make it easier for interpreters to 2234 jump around in tree node stream. 2235 2236* New CommonTreeNodeStream buffers all nodes in stream for fast jumping 2237 around. It now has push/pop methods to invoke other locations in 2238 the stream for building interpreters. 2239 2240* Moved CommonTreeNodeStream to UnBufferedTreeNodeStream and removed 2241 Iterator implementation. moved toNodesOnlyString() to TestTreeNodeStream 2242 2243* [BREAKS ANY TREE IMPLEMENTATION] 2244 made CommonTreeNodeStream work with any tree node type. TreeAdaptor 2245 now implements isNil so must add; trivial, but does break back 2246 compatibility. 2247 2248December 17, 2006 2249 2250* Added traceIn/Out methods to recognizers so that you can override them; 2251 previously they were in-line print statements. The message has also 2252 been slightly improved. 2253 2254* Factored BuildParseTree into debug package; cleaned stuff up. Fixed 2255 unit tests. 2256 2257December 15, 2006 2258 2259* [BREAKS ANY TREE IMPLEMENTATION] 2260 org.antlr.runtime.tree.Tree; needed to add get/set for token start/stop 2261 index so CommonTreeAdaptor can assume Tree interface not CommonTree 2262 implementation. Otherwise, no way to create your own nodes that satisfy 2263 Tree because CommonTreeAdaptor was doing 2264 2265 public int getTokenStartIndex(Object t) { 2266 return ((CommonTree)t).startIndex; 2267 } 2268 2269 Added to Tree: 2270 2271 /** What is the smallest token index (indexing from 0) for this node 2272 * and its children? 2273 */ 2274 int getTokenStartIndex(); 2275 2276 void setTokenStartIndex(int index); 2277 2278 /** What is the largest token index (indexing from 0) for this node 2279 * and its children? 2280 */ 2281 int getTokenStopIndex(); 2282 2283 void setTokenStopIndex(int index); 2284 2285December 13, 2006 2286 2287* Added org.antlr.runtime.tree.DOTTreeGenerator so you can generate DOT 2288 diagrams easily from trees. 2289 2290 CharStream input = new ANTLRInputStream(System.in); 2291 TLexer lex = new TLexer(input); 2292 CommonTokenStream tokens = new CommonTokenStream(lex); 2293 TParser parser = new TParser(tokens); 2294 TParser.e_return r = parser.e(); 2295 Tree t = (Tree)r.tree; 2296 System.out.println(t.toStringTree()); 2297 DOTTreeGenerator gen = new DOTTreeGenerator(); 2298 StringTemplate st = gen.toDOT(t); 2299 System.out.println(st); 2300 2301* Changed the way mark()/rewind() work in CommonTreeNode stream to mirror 2302 more flexible solution in ANTLRStringStream. Forgot to set lastMarker 2303 anyway. Now you can rewind to non-most-recent marker. 2304 2305December 12, 2006 2306 2307* Temp lexer now end in .gl (T__.gl, for example) 2308 2309* TreeParser suffix no longer generated for tree grammars 2310 2311* Defined reset for lexer, parser, tree parser; rewinds the input stream also 2312 2313December 10, 2006 2314 2315* Made Grammar.abortNFAToDFAConversion() abort in middle of a DFA. 2316 2317December 9, 2006 2318 2319* fixed bug in OrderedHashSet.add(). It didn't track elements correctly. 2320 2321December 6, 2006 2322 2323* updated build.xml for future Ant compatibility, thanks to Matt Benson. 2324 2325* various tests in TestRewriteTemplate and TestSyntacticPredicateEvaluation 2326 were using the old 'channel' vs. new '$channel' notation. 2327 TestInterpretedParsing didn't pick up an earlier change to CommonToken. 2328 Reported by Matt Benson. 2329 2330* fixed platform dependent test failures in TestTemplates, supplied by Matt 2331 Benson. 2332 2333November 29, 2006 2334 2335* optimized semantic predicate evaluation so that p||!p yields true. 2336 2337November 22, 2006 2338 2339* fixed bug that prevented var = $rule.some_retval from working in anything 2340 but the first alternative of a rule or subrule. 2341 2342* attribute names containing digits were not allowed, this is now fixed, 2343 allowing attributes like 'name1' but not '1name1'. 2344 2345November 19, 2006 2346 2347* Removed LeftRecursionMessage and apparatus because it seems that I check 2348 for left recursion upfront before analysis and everything gets specified as 2349 recursion cycles at this point. 2350 2351November 16, 2006 2352 2353* TokenRewriteStream.replace was not passing programName to next method. 2354 2355November 15, 2006 2356 2357* updated DOT files for DFA generation to make smaller circles. 2358 2359* made epsilon edges italics in the NFA diagrams. 2360 23613.0b5 - November 15, 2006 2362 2363The biggest thing is that your grammar file names must match the grammar name 2364inside (your generated class names will also be different) and we use 2365$channel=HIDDEN now instead of channel=99 inside lexer actions. 2366Should be compatible other than that. Please look at complete list of 2367changes. 2368 2369November 14, 2006 2370 2371* Force token index to be -1 for CommonIndex in case not set. 2372 2373November 11, 2006 2374 2375* getUniqueID for TreeAdaptor now uses identityHashCode instead of hashCode. 2376 2377November 10, 2006 2378 2379* No grammar nondeterminism warning now when wildcard '.' is final alt. 2380 Examples: 2381 2382 a : A | B | . ; 2383 2384 A : 'a' 2385 | . 2386 ; 2387 2388 SL_COMMENT 2389 : '//' (options {greedy=false;} : .)* '\r'? '\n' 2390 ; 2391 2392 SL_COMMENT2 2393 : '//' (options {greedy=false;} : 'x'|.)* '\r'? '\n' 2394 ; 2395 2396 2397November 8, 2006 2398 2399* Syntactic predicates did not get hoisting properly upon non-LL(*) decision. Other hoisting issues fixed. Cleaned up code. 2400 2401* Removed failsafe that check to see if I'm spending too much time on a single DFA; I don't think we need it anymore. 2402 2403November 3, 2006 2404 2405* $text, $line, etc... were not working in assignments. Fixed and added 2406 test case. 2407 2408* $label.text translated to label.getText in lexer even if label was on a char 2409 2410November 2, 2006 2411 2412* Added error if you don't specify what the AST type is; actions in tree 2413 grammar won't work without it. 2414 2415 $ cat x.g 2416 tree grammar x; 2417 a : ID {String s = $ID.text;} ; 2418 2419 ANTLR Parser Generator Early Access Version 3.0b5 (??, 2006) 1989-2006 2420 error: x.g:0:0: (152) tree grammar x has no ASTLabelType option 2421 2422November 1, 2006 2423 2424* $text, $line, etc... were not working properly within lexer rule. 2425 2426October 32, 2006 2427 2428* Finally actions now execute before dynamic scopes are popped it in the 2429 rule. Previously was not possible to access the rules scoped variables 2430 in a finally action. 2431 2432October 29, 2006 2433 2434* Altered ActionTranslator to emit errors on setting read-only attributes 2435 such as $start, $stop, $text in a rule. Also forbid setting any attributes 2436 in rules/tokens referenced by a label or name. 2437 Setting dynamic scopes's attributes and your own parameter attributes 2438 is legal. 2439 2440October 27, 2006 2441 2442* Altered how ANTLR figures out what decision is associated with which 2443 block of grammar. Makes ANTLRWorks correctly find DFA for a block. 2444 2445October 26, 2006 2446 2447* Fixed bug where EOT transitions led to no NFA configs in a DFA state, 2448 yielding an error in DFA table generation. 2449 2450* renamed action.g to ActionTranslator.g 2451 the ActionTranslator class is now called ActionTranslatorLexer, as ANTLR 2452 generates this classname now. Fixed rest of codebase accordingly. 2453 2454* added rules recognizing setting of scopes' attributes to ActionTranslator.g 2455 the Objective C target needed access to the right-hand side of the assignment 2456 in order to generate correct code 2457 2458* changed ANTLRCore.sti to reflect the new mandatory templates to support the above 2459 namely: scopeSetAttributeRef, returnSetAttributeRef and the ruleSetPropertyRef_* 2460 templates, with the exception of ruleSetPropertyRef_text. we cannot set this attribute 2461 2462October 19, 2006 2463 2464* Fixed 2 bugs in DFA conversion that caused exceptions. 2465 altered functionality of getMinElement so it ignores elements<0. 2466 2467October 18, 2006 2468 2469* moved resetStateNumbersToBeContiguous() to after issuing of warnings; 2470 an internal error in that routine should make more sense as issues 2471 with decision will appear first. 2472 2473* fixed cut/paste bug I introduced when fixed EOF in min/max 2474 bug. Prevented C grammar from working briefly. 2475 2476October 17, 2006 2477 2478* Removed a failsafe that seems to be unnecessary that ensure DFA didn't 2479 get too big. It was resulting in some failures in code generation that 2480 led me on quite a strange debugging trip. 2481 2482October 16, 2006 2483 2484* Use channel=HIDDEN not channel=99 to put tokens on hidden channel. 2485 2486October 12, 2006 2487 2488* ANTLR now has a customizable message format for errors and warnings, 2489 to make it easier to fulfill requirements by IDEs and such. 2490 The format to be used can be specified via the '-message-format name' 2491 command line switch. The default for name is 'antlr', also available 2492 at the moment is 'gnu'. This is done via StringTemplate, for details 2493 on the requirements look in org/antlr/tool/templates/messages/formats/ 2494 2495* line numbers for lexers in combined grammars are now reported correctly. 2496 2497September 29, 2006 2498 2499* ANTLRReaderStream improperly checked for end of input. 2500 2501September 28, 2006 2502 2503* For ANTLRStringStream, LA(-1) was off by one...gave you LA(-2). 2504 25053.0b4 - August 24, 2006 2506 2507* error when no rules in grammar. doesn't crash now. 2508 2509* Token is now an interface. 2510 2511* remove dependence on non runtime classes in runtime package. 2512 2513* filename and grammar name must be same Foo in Foo.g. Generates FooParser, 2514 FooLexer, ... Combined grammar Foo generates Foo$Lexer.g which generates 2515 FooLexer.java. tree grammars generate FooTreeParser.java 2516 2517August 24, 2006 2518 2519* added C# target to lib, codegen, templates 2520 2521August 11, 2006 2522 2523* added tree arg to navigation methods in treeadaptor 2524 2525August 07, 2006 2526 2527* fixed bug related to (a|)+ on end of lexer rules. crashed instead 2528 of warning. 2529 2530* added warning that interpreter doesn't do synpreds yet 2531 2532* allow different source of classloader: 2533ClassLoader cl = Thread.currentThread().getContextClassLoader(); 2534if ( cl==null ) { 2535 cl = this.getClass().getClassLoader(); 2536} 2537 2538 2539July 26, 2006 2540 2541* compressed DFA edge tables significantly. All edge tables are 2542 unique. The transition table can reuse arrays. Look like this now: 2543 2544 public static readonly DFA30_transition0 = 2545 new short[] { 46, 46, -1, 46, 46, -1, -1, -1, -1, -1, -1, -1,...}; 2546 public static readonly DFA30_transition1 = 2547 new short[] { 21 }; 2548 public static readonly short[][] DFA30_transition = { 2549 DFA30_transition0, 2550 DFA30_transition0, 2551 DFA30_transition1, 2552 ... 2553 }; 2554 2555* If you defined both a label like EQ and '=', sometimes the '=' was 2556 used instead of the EQ label. 2557 2558* made headerFile template have same arg list as outputFile for consistency 2559 2560* outputFile, lexer, genericParser, parser, treeParser templates 2561 reference cyclicDFAs attribute which was no longer used after I 2562 started the new table-based DFA. I made cyclicDFADescriptors 2563 argument to outputFile and headerFile (only). I think this is 2564 correct as only OO languages will want the DFA in the recognizer. 2565 At the top level, C and friends can use it. Changed name to use 2566 cyclicDFAs again as it's a better name probably. Removed parameter 2567 from the lexer, ... For example, my parser template says this now: 2568 2569 <cyclicDFAs:cyclicDFA()> <! dump tables for all DFA !> 2570 2571* made all token ref token types go thru code gen's 2572 getTokenTypeAsTargetLabel() 2573 2574* no more computing DFA transition tables for acyclic DFA. 2575 2576July 25, 2006 2577 2578* fixed a place where I was adding syn predicates into rewrite stuff. 2579 2580* turned off invalid token index warning in AW support; had a problem. 2581 2582* bad location event generated with -debug for synpreds in autobacktrack mode. 2583 2584July 24, 2006 2585 2586* changed runtime.DFA so that it treats all chars and token types as 2587 char (unsigned 16 bit int). -1 becomes '\uFFFF' then or 65535. 2588 2589* changed MAX_STATE_TRANSITIONS_FOR_TABLE to be 65534 by default 2590 now. This means that all states can use a table to do transitions. 2591 2592* was not making synpreds on (C)* type loops with backtrack=true 2593 2594* was copying tree stuff and actions into synpreds with backtrack=true 2595 2596* was making synpreds on even single alt rules / blocks with backtrack=true 2597 25983.0b3 - July 21, 2006 2599 2600* ANTLR fails to analyze complex decisions much less frequently. It 2601 turns out that the set of decisions for which ANTLR fails (times 2602 out) is the same set (so far) of non-LL(*) decisions. Morever, I'm 2603 able to detect this situation quickly and report rather than timing 2604 out. Errors look like: 2605 2606 java.g:468:23: [fatal] rule concreteDimensions has non-LL(*) 2607 decision due to recursive rule invocations in alts 1,2. Resolve 2608 by left-factoring or using syntactic predicates with fixed k 2609 lookahead or use backtrack=true option. 2610 2611 This message only appears when k=*. 2612 2613* Shortened no viable alt messages to not include decision 2614 description: 2615 2616[compilationUnit, declaration]: line 8:8 decision=<<67:1: declaration 2617: ( ( fieldDeclaration )=> fieldDeclaration | ( methodDeclaration )=> 2618methodDeclaration | ( constructorDeclaration )=> 2619constructorDeclaration | ( classDeclaration )=> classDeclaration | ( 2620interfaceDeclaration )=> interfaceDeclaration | ( blockDeclaration )=> 2621blockDeclaration | emptyDeclaration );>> state 3 (decision=14) no 2622viable alt; token=[@1,184:187='java',<122>,8:8] 2623 2624 too long and hard to read. 2625 2626July 19, 2006 2627 2628* Code gen bug: states with no emanating edges were ignored by ST. 2629 Now an empty list is used. 2630 2631* Added grammar parameter to recognizer templates so they can access 2632 properties like getName(), ... 2633 2634July 10, 2006 2635 2636* Fixed the gated pred merged state bug. Added unit test. 2637 2638* added new method to Target: getTokenTypeAsTargetLabel() 2639 2640July 7, 2006 2641 2642* I was doing an AND instead of OR in the gated predicate stuff. 2643 Thanks to Stephen Kou! 2644 2645* Reduce op for combining predicates was insanely slow sometimes and 2646 didn't actually work well. Now it's fast and works. 2647 2648* There is a bug in merging of DFA stop states related to gated 2649 preds...turned it off for now. 2650 26513.0b2 - July 5, 2006 2652 2653July 5, 2006 2654 2655* token emission not properly protected in lexer filter mode. 2656 2657* EOT, EOT DFA state transition tables should be init'd to -1 (only 2658 was doing this for compressed tables). Fixed. 2659 2660* in trace mode, exit method not shown for memoized rules 2661 2662* added -Xmaxdfaedges to allow you to increase number of edges allowed 2663 for a single DFA state before it becomes "special" and can't fit in 2664 a simple table. 2665 2666* Bug in tables. Short are signed so min/max tables for DFA are now 2667 char[]. Bizarre. 2668 2669July 3, 2006 2670 2671* Added a method to reset the tool error state for current thread. 2672 See ErrorManager.java 2673 2674* [Got this working properly today] backtrack mode that let's you type 2675 in any old crap and ANTLR will backtrack if it can't figure out what 2676 you meant. No errors are reported by antlr during analysis. It 2677 implicitly adds a syn pred in front of every production, using them 2678 only if static grammar LL(*) analysis fails. Syn pred code is not 2679 generated if the pred is not used in a decision. 2680 2681 This is essentially a rapid prototyping mode. 2682 2683* Added backtracking report to the -report option 2684 2685* Added NFA->DFA conversion early termination report to the -report option 2686 2687* Added grammar level k and backtrack options to -report 2688 2689* Added a dozen unit tests to test autobacktrack NFA construction. 2690 2691* If you are using filter mode, you must manually use option 2692 memoize=true now. 2693 2694July 2, 2006 2695 2696* Added k=* option so you can set k=2, for example, on whole grammar, 2697 but an individual decision can be LL(*). 2698 2699* memoize option for grammars, rules, blocks. Remove -nomemo cmd-line option 2700 2701* but in DOT generator for DFA; fixed. 2702 2703* runtime.DFA reported errors even when backtracking 2704 2705July 1, 2006 2706 2707* Added -X option list to help 2708 2709* Syn preds were being hoisted into other rules, causing lots of extra 2710 backtracking. 2711 2712June 29, 2006 2713 2714* unnecessary files removed during build. 2715 2716* Matt Benson updated build.xml 2717 2718* Detecting use of synpreds in analysis now instead of codegen. In 2719 this way, I can avoid analyzing decisions in synpreds for synpreds 2720 not used in a DFA for a real rule. This is used to optimize things 2721 for backtrack option. 2722 2723* Code gen must add _fragment or whatever to end of pred name in 2724 template synpredRule to avoid having ANTLR know anything about 2725 method names. 2726 2727* Added -IdbgST option to emit ST delimiters at start/stop of all 2728 templates spit out. 2729 2730June 28, 2006 2731 2732* Tweaked message when ANTLR cannot handle analysis. 2733 27343.0b1 - June 27, 2006 2735 2736June 24, 2006 2737 2738* syn preds no longer generate little static classes; they also don't 2739 generate a whole bunch of extra crap in the rules built to test syn 2740 preds. Removed GrammarFragmentPointer class from runtime. 2741 2742June 23-24, 2006 2743 2744* added output option to -report output. 2745 2746* added profiling info: 2747 Number of rule invocations in "guessing" mode 2748 number of rule memoization cache hits 2749 number of rule memoization cache misses 2750 2751* made DFA DOT diagrams go left to right not top to bottom 2752 2753* I try to recursive overflow states now by resolving these states 2754 with semantic/syntactic predicates if they exist. The DFA is then 2755 deterministic rather than simply resolving by choosing first 2756 nondeterministic alt. I used to generated errors: 2757 2758~/tmp $ java org.antlr.Tool -dfa t.g 2759ANTLR Parser Generator Early Access Version 3.0b2 (July 5, 2006) 1989-2006 2760t.g:2:5: Alternative 1: after matching input such as A A A A A decision cannot predict what comes next due to recursion overflow to b from b 2761t.g:2:5: Alternative 2: after matching input such as A A A A A decision cannot predict what comes next due to recursion overflow to b from b 2762 2763 Now, I uses predicates if available and emits no warnings. 2764 2765* made sem preds share accept states. Previously, multiple preds in a 2766decision forked new accepts each time for each nondet state. 2767 2768June 19, 2006 2769 2770* Need parens around the prediction expressions in templates. 2771 2772* Referencing $ID.text in an action forced bad code gen in lexer rule ID. 2773 2774* Fixed a bug in how predicates are collected. The definition of 2775 "last predicated alternative" was incorrect in the analysis. Further, 2776 gated predicates incorrectly missed a case where an edge should become 2777 true (a tautology). 2778 2779* Removed an unnecessary input.consume() reference in the runtime/DFA class. 2780 2781June 14, 2006 2782 2783* -> ($rulelabel)? didn't generate proper code for ASTs. 2784 2785* bug in code gen (did not compile) 2786a : ID -> ID 2787 | ID -> ID 2788 ; 2789Problem is repeated ref to ID from left side. Juergen pointed this out. 2790 2791* use of tokenVocab with missing file yielded exception 2792 2793* (A|B)=> foo yielded an exception as (A|B) is a set not a block. Fixed. 2794 2795* Didn't set ID1= and INT1= for this alt: 2796 | ^(ID INT+ {System.out.print(\"^(\"+$ID+\" \"+$INT+\")\");}) 2797 2798* Fixed so repeated dangling state errors only occur once like: 2799t.g:4:17: the decision cannot distinguish between alternative(s) 2,1 for at least one input sequence 2800 2801* tracking of rule elements was on (making list defs at start of 2802 method) with templates instead of just with ASTs. Turned off. 2803 2804* Doesn't crash when you give it a missing file now. 2805 2806* -report: add output info: how many LL(1) decisions. 2807 2808June 13, 2006 2809 2810* ^(ROOT ID?) Didn't work; nor did any other nullable child list such as 2811 ^(ROOT ID* INT?). Now, I check to see if child list is nullable using 2812 Grammar.LOOK() and, if so, I generate an "IF lookahead is DOWN" gate 2813 around the child list so the whole thing is optional. 2814 2815* Fixed a bug in LOOK that made it not look through nullable rules. 2816 2817* Using AST suffixes or -> rewrite syntax now gives an error w/o a grammar 2818 output option. Used to crash ;) 2819 2820* References to EOF ended up with improper -1 refs instead of EOF in output. 2821 2822* didn't warn of ambig ref to $expr in rewrite; fixed. 2823list 2824 : '[' expr 'for' type ID 'in' expr ']' 2825 -> comprehension(expr={$expr.st},type={},list={},i={}) 2826 ; 2827 2828June 12, 2006 2829 2830* EOF works in the parser as a token name. 2831 2832* Rule b:(A B?)*; didn't display properly in AW due to the way ANTLR 2833 generated NFA. 2834 2835* "scope x;" in a rule for unknown x gives no error. Fixed. Added unit test. 2836 2837* Label type for refs to start/stop in tree parser and other parsers were 2838 not used. Lots of casting. Ick. Fixed. 2839 2840* couldn't refer to $tokenlabel in isolation; but need so we can test if 2841 something was matched. Fixed. 2842 2843* Lots of little bugs fixed in $x.y, %... translation due to new 2844 action translator. 2845 2846* Improperly tracking block nesting level; result was that you couldn't 2847 see $ID in action of rule "a : A+ | ID {Token t = $ID;} | C ;" 2848 2849* a : ID ID {$ID.text;} ; did not get a warning about ambiguous $ID ref. 2850 2851* No error was found on $COMMENT.text: 2852 2853COMMENT 2854 : '/*' (options {greedy=false;} : . )* '*/' 2855 {System.out.println("found method "+$COMMENT.text);} 2856 ; 2857 2858 $enclosinglexerrule scope does not exist. Use text or setText() here. 2859 2860June 11, 2006 2861 2862* Single return values are initialized now to default or to your spec. 2863 2864* cleaned up input stream stuff. Added ANTLRReaderStream, ANTLRInputStream 2865 and refactored. You can specify encodings now on ANTLRFileStream (and 2866 ANTLRInputStream) now. 2867 2868* You can set text local var now in a lexer rule and token gets that text. 2869 start/stop indexes are still set for the token. 2870 2871* Changed lexer slightly. Calling a nonfragment rule from a 2872 nonfragment rule does not set the overall token. 2873 2874June 10, 2006 2875 2876* Fixed bug where unnecessary escapes yield char==0 like '\{'. 2877 2878* Fixed analysis bug. This grammar didn't report a recursion warning: 2879x : y X 2880 | y Y 2881 ; 2882y : L y R 2883 | B 2884 ; 2885 The DFAState.equals() method was messed up. 2886 2887* Added @synpredgate {...} action so you can tell ANTLR how to gate actions 2888 in/out during syntactic predicate evaluation. 2889 2890* Fuzzy parsing should be more efficient. It should backtrack over a rule 2891 and then rewind and do it again "with feeling" to exec actions. It was 2892 actually doing it 3x not 2x. 2893 2894June 9, 2006 2895 2896* Gutted and rebuilt the action translator for $x.y, $x::y, ... 2897 Uses ANTLR v3 now for the first time inside v3 source. :) 2898 ActionTranslator.java 2899 2900* Fixed a bug where referencing a return value on a rule didn't work 2901 because later a ref to that rule's predefined properties didn't 2902 properly force a return value struct to be built. Added unit test. 2903 2904June 6, 2006 2905 2906* New DFA mechanisms. Cyclic DFA are implemented as state tables, 2907 encoded via strings as java cannot handle large static arrays :( 2908 States with edges emanating that have predicates are specially 2909 treated. A method is generated to do these states. The DFA 2910 simulation routine uses the "special" array to figure out if the 2911 state is special. See March 25, 2006 entry for description: 2912 http://www.antlr.org/blog/antlr3/codegen.tml. analysis.DFA now has 2913 all the state tables generated for code gen. CyclicCodeGenerator.java 2914 disappeared as it's unneeded code. :) 2915 2916* Internal general clean up of the DFA.states vs uniqueStates thing. 2917 Fixed lookahead decisions no longer fill uniqueStates. Waste of 2918 time. Also noted that when adding sem pred edges, I didn't check 2919 for state reuse. Fixed. 2920 2921June 4, 2006 2922 2923* When resolving ambig DFA states predicates, I did not add the new states 2924 to the list of unique DFA states. No observable effect on output except 2925 that DFA state numbers were not always contiguous for predicated decisions. 2926 I needed this fix for new DFA tables. 2927 29283.0ea10 - June 2, 2006 2929 2930June 2, 2006 2931 2932* Improved grammar stats and added syntactic pred tracking. 2933 2934June 1, 2006 2935 2936* Due to a type mismatch, the DebugParser.recoverFromMismatchedToken() 2937 method was not called. Debug events for mismatched token error 2938 notification were not sent to ANTLRWorks probably 2939 2940* Added getBacktrackingLevel() for any recognizer; needed for profiler. 2941 2942* Only writes profiling data for antlr grammar analysis with -profile set 2943 2944* Major update and bug fix to (runtime) Profiler. 2945 2946May 27, 2006 2947 2948* Added Lexer.skip() to force lexer to ignore current token and look for 2949 another; no token is created for current rule and is not passed on to 2950 parser (or other consumer of the lexer). 2951 2952* Parsers are much faster now. I removed use of java.util.Stack for pushing 2953 follow sets and use a hardcoded array stack instead. Dropped from 2954 5900ms to 3900ms for parse+lex time parsing entire java 1.4.2 source. Lex 2955 time alone was about 1500ms. Just looking at parse time, we get about 2x 2956 speed improvement. :) 2957 2958May 26, 2006 2959 2960* Fixed NFA construction so it generates NFA for (A*)* such that ANTLRWorks 2961 can display it properly. 2962 2963May 25, 2006 2964 2965* added abort method to Grammar so AW can terminate the conversion if it's 2966 taking too long. 2967 2968May 24, 2006 2969 2970* added method to get left recursive rules from grammar without doing full 2971 grammar analysis. 2972 2973* analysis, code gen not attempted if serious error (like 2974 left-recursion or missing rule definition) occurred while reading 2975 the grammar in and defining symbols. 2976 2977* added amazing optimization; reduces analysis time by 90% for java 2978 grammar; simple IF statement addition! 2979 29803.0ea9 - May 20, 2006 2981 2982* added global k value for grammar to limit lookahead for all decisions unless 2983overridden in a particular decision. 2984 2985* added failsafe so that any decision taking longer than 2 seconds to create 2986the DFA will fall back on k=1. Use -ImaxtimeforDFA n (in ms) to set the time. 2987 2988* added an option (turned off for now) to use multiple threads to 2989perform grammar analysis. Not much help on a 2-CPU computer as 2990garbage collection seems to peg the 2nd CPU already. :( Gotta wait for 2991a 4 CPU box ;) 2992 2993* switched from #src to // $ANTLR src directive. 2994 2995* CommonTokenStream.getTokens() looked past end of buffer sometimes. fixed. 2996 2997* unicode literals didn't really work in DOT output and generated code. fixed. 2998 2999* fixed the unit test rig so it compiles nicely with Java 1.5 3000 3001* Added ant build.xml file (reads build.properties file) 3002 3003* predicates sometimes failed to compile/eval properly due to missing (...) 3004 in IF expressions. Forced (..) 3005 3006* (...)? with only one alt were not optimized. Was: 3007 3008 // t.g:4:7: ( B )? 3009 int alt1=2; 3010 int LA1_0 = input.LA(1); 3011 if ( LA1_0==B ) { 3012 alt1=1; 3013 } 3014 else if ( LA1_0==-1 ) { 3015 alt1=2; 3016 } 3017 else { 3018 NoViableAltException nvae = 3019 new NoViableAltException("4:7: ( B )?", 1, 0, input); 3020 throw nvae; 3021 } 3022 3023is now: 3024 3025 // t.g:4:7: ( B )? 3026 int alt1=2; 3027 int LA1_0 = input.LA(1); 3028 if ( LA1_0==B ) { 3029 alt1=1; 3030 } 3031 3032 Smaller, faster and more readable. 3033 3034* Allow manual init of return values now: 3035 functionHeader returns [int x=3*4, char (*f)()=null] : ... ; 3036 3037* Added optimization for DFAs that fixed a codegen bug with rules in lexer: 3038 EQ : '=' ; 3039 ASSIGNOP : '=' | '+=' ; 3040 EQ is a subset of other rule. It did not given an error which is 3041 correct, but generated bad code. 3042 3043* ANTLR was sending column not char position to ANTLRWorks. 3044 3045* Bug fix: location 0, 0 emitted for synpreds and empty alts. 3046 3047* debugging event handshake how sends grammar file name. Added getGrammarFileName() to recognizers. Java.stg generates it: 3048 3049 public String getGrammarFileName() { return "<fileName>"; } 3050 3051* tree parsers can do arbitrary lookahead now including backtracking. I 3052 updated CommonTreeNodeStream. 3053 3054* added events for debugging tree parsers: 3055 3056 /** Input for a tree parser is an AST, but we know nothing for sure 3057 * about a node except its type and text (obtained from the adaptor). 3058 * This is the analog of the consumeToken method. Again, the ID is 3059 * the hashCode usually of the node so it only works if hashCode is 3060 * not implemented. 3061 */ 3062 public void consumeNode(int ID, String text, int type); 3063 3064 /** The tree parser looked ahead */ 3065 public void LT(int i, int ID, String text, int type); 3066 3067 /** The tree parser has popped back up from the child list to the 3068 * root node. 3069 */ 3070 public void goUp(); 3071 3072 /** The tree parser has descended to the first child of a the current 3073 * root node. 3074 */ 3075 public void goDown(); 3076 3077* Added DebugTreeNodeStream and DebugTreeParser classes 3078 3079* Added ctor because the debug tree node stream will need to ask quesitons about nodes and since nodes are just Object, it needs an adaptor to decode the nodes and get text/type info for the debugger. 3080 3081public CommonTreeNodeStream(TreeAdaptor adaptor, Tree tree); 3082 3083* added getter to TreeNodeStream: 3084 public TreeAdaptor getTreeAdaptor(); 3085 3086* Implemented getText/getType in CommonTreeAdaptor. 3087 3088* Added TraceDebugEventListener that can dump all events to stdout. 3089 3090* I broke down and make Tree implement getText 3091 3092* tree rewrites now gen location debug events. 3093 3094* added AST debug events to listener; added blank listener for convenience 3095 3096* updated debug events to send begin/end backtrack events for debugging 3097 3098* with a : (b->b) ('+' b -> ^(PLUS $a b))* ; you get b[0] each time as 3099 there is no loop in rewrite rule itself. Need to know context that 3100 the -> is inside the rule and hence b means last value of b not all 3101 values. 3102 3103* Bug in TokenRewriteStream; ops at indexes < start index blocked proper op. 3104 3105* Actions in ST rewrites "-> ({$op})()" were not translated 3106 3107* Added new action name: 3108 3109@rulecatch { 3110catch (RecognitionException re) { 3111 reportError(re); 3112 recover(input,re); 3113} 3114catch (Throwable t) { 3115 System.err.println(t); 3116} 3117} 3118Overrides rule catch stuff. 3119 3120* Isolated $ refs caused exception 3121 31223.0ea8 - March 11, 2006 3123 3124* added @finally {...} action like @init for rules. Executes in 3125 finally block (java target) after all other stuff like rule memoization. 3126 No code changes needs; ST just refs a new action: 3127 <ruleDescriptor.actions.finally> 3128 3129* hideous bug fixed: PLUS='+' didn't result in '+' rule in lexer 3130 3131* TokenRewriteStream didn't do toString() right when no rewrites had been done. 3132 3133* lexer errors in interpreter were not printed properly 3134 3135* bitsets are dumped in hex not decimal now for FOLLOW sets 3136 3137* /* epsilon */ is not printed now when printing out grammars with empty alts 3138 3139* Fixed another bug in tree rewrite stuff where it was checking that elements 3140 had at least one element. Strange...commented out for now to see if I can remember what's up. 3141 3142* Tree rewrites had problems when you didn't have x+=FOO variables. Rules 3143 like this work now: 3144 3145 a : (x=ID)? y=ID -> ($x $y)?; 3146 3147* filter=true for lexers turns on k=1 and backtracking for every token 3148 alternative. Put the rules in priority order. 3149 3150* added getLine() etc... to Tree to support better error reporting for 3151 trees. Added MismatchedTreeNodeException. 3152 3153* $templates::foo() is gone. added % as special template symbol. 3154 %foo(a={},b={},...) ctor (even shorter than $templates::foo(...)) 3155 %({name-expr})(a={},...) indirect template ctor reference 3156 3157 The above are parsed by antlr.g and translated by codegen.g 3158 The following are parsed manually here: 3159 3160 %{string-expr} anonymous template from string expr 3161 %{expr}.y = z; template attribute y of StringTemplate-typed expr to z 3162 %x.y = z; set template attribute y of x (always set never get attr) 3163 to z [languages like python without ';' must still use the 3164 ';' which the code generator is free to remove during code gen] 3165 3166* -> ({expr})(a={},...) notation for indirect template rewrite. 3167 expr is the name of the template. 3168 3169* $x[i]::y and $x[-i]::y notation for accesssing absolute scope stack 3170 indexes and relative negative scopes. $x[-1]::y is the y attribute 3171 of the previous scope (stack top - 1). 3172 3173* filter=true mode for lexers; can do this now...upon mismatch, just 3174 consumes a char and tries again: 3175lexer grammar FuzzyJava; 3176options {filter=true;} 3177 3178FIELD 3179 : TYPE WS? name=ID WS? (';'|'=') 3180 {System.out.println("found var "+$name.text);} 3181 ; 3182 3183* refactored char streams so ANTLRFileStream is now a subclass of 3184 ANTLRStringStream. 3185 3186* char streams for lexer now allowed nested backtracking in lexer. 3187 3188* added TokenLabelType for lexer/parser for all token labels 3189 3190* line numbers for error messages were not updated properly in antlr.g 3191 for strings, char literals and <<...>> 3192 3193* init action in lexer rules was before the type,start,line,... decls. 3194 3195* Tree grammars can now specify output; I've only tested output=templat 3196 though. 3197 3198* You can reference EOF now in the parser and lexer. It's just token type 3199 or char value -1. 3200 3201* Bug fix: $ID refs in the *lexer* were all messed up. Cleaned up the 3202 set of properties available... 3203 3204* Bug fix: .st not found in rule ref when rule has scope: 3205field 3206scope { 3207 StringTemplate funcDef; 3208} 3209 : ... 3210 {$field::funcDef = $field.st;} 3211 ; 3212it gets field_stack.st instead 3213 3214* return in backtracking must return retval or null if return value. 3215 3216* $property within a rule now works like $text, $st, ... 3217 3218* AST/Template Rewrites were not gated by backtracking==0 so they 3219 executed even when guessing. Auto AST construction is now gated also. 3220 3221* CommonTokenStream was somehow returning tokens not text in toString() 3222 3223* added useful methods to runtime.BitSet and also to CommonToken so you can 3224 update the text. Added nice Token stream method: 3225 3226 /** Given a start and stop index, return a List of all tokens in 3227 * the token type BitSet. Return null if no tokens were found. This 3228 * method looks at both on and off channel tokens. 3229 */ 3230 public List getTokens(int start, int stop, BitSet types); 3231 3232* literals are now passed in the .tokens files so you can ref them in 3233 tree parses, for example. 3234 3235* added basic exception handling; no labels, just general catches: 3236 3237a : {;}A | B ; 3238 exception 3239 catch[RecognitionException re] { 3240 System.out.println("recog error"); 3241 } 3242 catch[Exception e] { 3243 System.out.println("error"); 3244 } 3245 3246* Added method to TokenStream: 3247 public String toString(Token start, Token stop); 3248 3249* antlr generates #src lines in lexer grammars generated from combined grammars 3250 so error messages refer to original file. 3251 3252* lexers generated from combined grammars now use originally formatting. 3253 3254* predicates have $x.y stuff translated now. Warning: predicates might be 3255 hoisted out of context. 3256 3257* return values in return val structs are now public. 3258 3259* output=template with return values on rules was broken. I assume return values with ASTs was broken too. Fixed. 3260 32613.0ea7 - December 14, 2005 3262 3263* Added -print option to print out grammar w/o actions 3264 3265* Renamed BaseParser to be BaseRecognizer and even made Lexer derive from 3266 this; nice as it now shares backtracking support code. 3267 3268* Added syntactic predicates (...)=>. See December 4, 2005 entry: 3269 3270 http://www.antlr.org/blog/antlr3/lookahead.tml 3271 3272 Note that we have a new option for turning off rule memoization during 3273 backtracking: 3274 3275 -nomemo when backtracking don't generate memoization code 3276 3277* Predicates are now tested in order that you specify the alts. If you 3278 leave the last alt "naked" (w/o pred), it will assume a true pred rather 3279 than union of other preds. 3280 3281* Added gated predicates "{p}?=>" that literally turn off a production whereas 3282disambiguating predicates are only hoisted into the predictor when syntax alone 3283is not sufficient to uniquely predict alternatives. 3284 3285A : {p}? => "a" ; 3286B : {!p}? => ("a"|"b")+ ; 3287 3288* bug fixed related to predicates in predictor 3289lexer grammar w; 3290A : {p}? "a" ; 3291B : {!p}? ("a"|"b")+ ; 3292DFA is correct. A state splits for input "a" on the pred. 3293Generated code though was hosed. No pred tests in prediction code! 3294I added testLexerPreds() and others in TestSemanticPredicateEvaluation.java 3295 3296* added execAction template in case we want to do something in front of 3297 each action execution or something. 3298 3299* left-recursive cycles from rules w/o decisions were not detected. 3300 3301* undefined lexer rules were not announced! fixed. 3302 3303* unreachable messages for Tokens rule now indicate rule name not alt. E.g., 3304 3305 Ruby.lexer.g:24:1: The following token definitions are unreachable: IVAR 3306 3307* nondeterminism warnings improved for Tokens rule: 3308 3309Ruby.lexer.g:10:1: Multiple token rules can match input such as ""0".."9"": INT, FLOAT 3310As a result, tokens(s) FLOAT were disabled for that input 3311 3312 3313* DOT diagrams didn't show escaped char properly. 3314 3315* Char/string literals are now all 'abc' not "abc". 3316 3317* action syntax changed "@scope::actionname {action}" where scope defaults 3318 to "parser" if parser grammar or combined grammar, "lexer" if lexer grammar, 3319 and "treeparser" if tree grammar. The code generation targets decide 3320 what scopes are available. Each "scope" yields a hashtable for use in 3321 the output templates. The scopes full of actions are sent to all output 3322 file templates (currently headerFile and outputFile) as attribute actions. 3323 Then you can reference <actions.scope> to get the map of actions associated 3324 with scope and <actions.parser.header> to get the parser's header action 3325 for example. This should be very flexible. The target should only have 3326 to define which scopes are valid, but the action names should be variable 3327 so we don't have to recompile ANTLR to add actions to code gen templates. 3328 3329 grammar T; 3330 options {language=Java;} 3331 @header { package foo; } 3332 @parser::stuff { int i; } // names within scope not checked; target dependent 3333 @members { int i; } 3334 @lexer::header {head} 3335 @lexer::members { int j; } 3336 @headerfile::blort {...} // error: this target doesn't have headerfile 3337 @treeparser::members {...} // error: this is not a tree parser 3338 a 3339 @init {int i;} 3340 : ID 3341 ; 3342 ID : 'a'..'z'; 3343 3344 For now, the Java target uses members and header as a valid name. Within a 3345 rule, the init action name is valid. 3346 3347* changed $dynamicscope.value to $dynamicscope::value even if value is defined 3348 in same rule such as $function::name where rule function defines name. 3349 3350* $dynamicscope gets you the stack 3351 3352* rule scopes go like this now: 3353 3354 rule 3355 scope {...} 3356 scope slist,Symbols; 3357 : ... 3358 ; 3359 3360* Created RuleReturnScope as a generic rule return value. Makes it easier 3361 to do this: 3362 RuleReturnScope r = parser.program(); 3363 System.out.println(r.getTemplate().toString()); 3364 3365* $template, $tree, $start, etc... 3366 3367* $r.x in current rule. $r is ignored as fully-qualified name. $r.start works too 3368 3369* added warning about $r referring to both return value of rule and dynamic scope of rule 3370 3371* integrated StringTemplate in a very simple manner 3372 3373Syntax: 3374-> template(arglist) "..." 3375-> template(arglist) <<...>> 3376-> namedTemplate(arglist) 3377-> {free expression} 3378-> // empty 3379 3380Predicate syntax: 3381a : A B -> {p1}? foo(a={$A.text}) 3382 -> {p2}? foo(a={$B.text}) 3383 -> // return nothing 3384 3385An arg list is just a list of template attribute assignments to actions in curlies. 3386 3387There is a setTemplateLib() method for you to use with named template rewrites. 3388 3389Use a new option: 3390 3391grammar t; 3392options {output=template;} 3393... 3394 3395This all should work for tree grammars too, but I'm still testing. 3396 3397* fixed bugs where strings were improperly escaped in exceptions, comments, etc.. For example, newlines came out as newlines not the escaped version 3398 33993.0ea6 - November 13, 2005 3400 3401* turned off -debug/-profile, which was on by default 3402 3403* completely refactored the output templates; added some missing templates. 3404 3405* dramatically improved infinite recursion error messages (actually 3406 left-recursion never even was printed out before). 3407 3408* wasn't printing dangling state messages when it reanalyzes with k=1. 3409 3410* fixed a nasty bug in the analysis engine dealing with infinite recursion. 3411 Spent all day thinking about it and cleaned up the code dramatically. 3412 Bug fixed and software is more powerful and I understand it better! :) 3413 3414* improved verbose DFA nodes; organized by alt 3415 3416* got much better random phrase generation. For example: 3417 3418 $ java org.antlr.tool.RandomPhrase simple.g program 3419 int Ktcdn ';' method wh '(' ')' '{' return 5 ';' '}' 3420 3421* empty rules like "a : ;" generated code that didn't compile due to 3422 try/catch for RecognitionException. Generated code couldn't possibly 3423 throw that exception. 3424 3425* when printing out a grammar, such as in comments in generated code, 3426 ANTLR didn't print ast suffix stuff back out for literals. 3427 3428* This never exited loop: 3429 DATA : (options {greedy=false;}: .* '\n' )* '\n' '.' ; 3430 and now it works due to new default nongreedy .* Also this works: 3431 DATA : (options {greedy=false;}: .* '\n' )* '.' ; 3432 3433* Dot star ".*" syntax didn't work; in lexer it is nongreedy by 3434 default. In parser it is on greedy but also k=1 by default. Added 3435 unit tests. Added blog entry to describe. 3436 3437* ~T where T is the only token yielded an empty set but no error 3438 3439* Used to generate unreachable message here: 3440 3441 parser grammar t; 3442 a : ID a 3443 | ID 3444 ; 3445 3446 z.g:3:11: The following alternatives are unreachable: 2 3447 3448 In fact it should really be an error; now it generates: 3449 3450 no start rule in grammar t (no rule can obviously be followed by EOF) 3451 3452 Per next change item, ANTLR cannot know that EOF follows rule 'a'. 3453 3454* added error message indicating that ANTLR can't figure out what your 3455 start rule is. Required to properly generate code in some cases. 3456 3457* validating semantic predicates now work (if they are false, they 3458 throw a new FailedPredicateException 3459 3460* two hideous bug fixes in the IntervalSet, which made analysis go wrong 3461 in a few cases. Thanks to Oliver Zeigermann for finding lots of bugs 3462 and making suggested fixes (including the next two items)! 3463 3464* cyclic DFAs are now nonstatic and hence can access instance variables 3465 3466* labels are now allowed on lexical elements (in the lexer) 3467 3468* added some internal debugging options 3469 3470* ~'a'* and ~('a')* were not working properly; refactored antlr.g grammar 3471 34723.0ea5 - July 5, 2005 3473 3474* Using '\n' in a parser grammar resulted in a nonescaped version of '\n' in the token names table making compilation fail. I fixed this by reorganizing/cleaning up portion of ANTLR that deals with literals. See comment org.antlr.codegen.Target. 3475 3476* Target.getMaxCharValue() did not use the appropriate max value constant. 3477 3478* ALLCHAR was a constant when it should use the Target max value def. set complement for wildcard also didn't use the Target def. Generally cleaned up the max char value stuff. 3479 3480* Code gen didn't deal with ASTLabelType properly...I think even the 3.0ea7 example tree parser was broken! :( 3481 3482* Added a few more unit tests dealing with escaped literals 3483 34843.0ea4 - June 29, 2005 3485 3486* tree parsers work; added CommonTreeNodeStream. See simplecTreeParser 3487 example in examples-v3 tarball. 3488 3489* added superClass and ASTLabelType options 3490 3491* refactored Parser to have a BaseParser and added TreeParser 3492 3493* bug fix: actions being dumped in description strings; compile errors 3494 resulted 3495 34963.0ea3 - June 23, 2005 3497 3498Enhancements 3499 3500* Automatic tree construction operators are in: ! ^ ^^ 3501 3502* Tree construction rewrite rules are in 3503 -> {pred1}? rewrite1 3504 -> {pred2}? rewrite2 3505 ... 3506 -> rewriteN 3507 3508 The rewrite rules may be elements like ID, expr, $label, {node expr} 3509 and trees ^( <root> <children> ). You have have (...)?, (...)*, (...)+ 3510 subrules as well. 3511 3512 You may have rewrites in subrules not just at outer level of rule, but 3513 any -> rewrite forces auto AST construction off for that alternative 3514 of that rule. 3515 3516 To avoid cycles, copy semantics are used: 3517 3518 r : INT -> INT INT ; 3519 3520 means make two new nodes from the same INT token. 3521 3522 Repeated references to a rule element implies a copy for at least one 3523 tree: 3524 3525 a : atom -> ^(atom atom) ; // NOT CYCLE! (dup atom tree) 3526 3527* $ruleLabel.tree refers to tree created by matching the labeled element. 3528 3529* A description of the blocks/alts is generated as a comment in output code 3530 3531* A timestamp / signature is put at top of each generated code file 3532 35333.0ea2 - June 12, 2005 3534 3535Bug fixes 3536 3537* Some error messages were missing the stackTrace parameter 3538 3539* Removed the file locking mechanism as it's not cross platform 3540 3541* Some absolute vs relative path name problems with writing output 3542 files. Rules are now more concrete. -o option takes precedence 3543 // -o /tmp /var/lib/t.g => /tmp/T.java 3544 // -o subdir/output /usr/lib/t.g => subdir/output/T.java 3545 // -o . /usr/lib/t.g => ./T.java 3546 // -o /tmp subdir/t.g => /tmp/subdir/t.g 3547 // If they didn't specify a -o dir so just write to location 3548 // where grammar is, absolute or relative 3549 3550* does error checking on unknown option names now 3551 3552* Using just language code not locale name for error message file. I.e., 3553 the default (and for any English speaking locale) is en.stg not en_US.stg 3554 anymore. 3555 3556* The error manager now asks the Tool to panic rather than simply doing 3557 a System.exit(). 3558 3559* Lots of refactoring concerning grammar, rule, subrule options. Now 3560 detects invalid options. 3561 35623.0ea1 - June 1, 2005 3563 3564Initial early access release 3565