forgetest1
/
antlr
forked from jasder/antlr
1
0
Fork 0
Code Issues Pull Requests 1 Releases Wiki Activity
7,372 Commits 4 Branches 0 Tags 82 MiB
Commit Graph

2369 Commits

Author SHA1 Message Date
Terence Parr 2b3508d57e Merge pull request #1662 from bhamiltoncx/no-more-char-casts 
Get rid of all (char) casts
 2017-02-17 13:21:16 -08:00
Ivan Kochurkin fab6172749 Added "action" to badWords set for Go runtime. 
Some spaced replaced with tabs.
 2017-02-17 23:31:51 +03:00
Ben Hamilton 88ecb3d9fa Get rid of all (char) casts 2017-02-16 09:55:08 -08:00
Ben Hamilton c35bbabc16 PR #1658: Also write stderr to file 2017-02-16 09:44:40 -08:00
Terence Parr 6c878b0709 Merge pull request #1653 from bhamiltoncx/codegen-unicode-escapes 
Emit language-specific Unicode escapes in codegen
 2017-02-15 16:48:50 -08:00
Terence Parr 287f17b796 Merge pull request #1652 from bhamiltoncx/csharp-lexer-parser-output-file 
C# support for lexer and parser output to a file
 2017-02-13 15:10:54 -08:00
Terence Parr 9b7e2a81ce Merge pull request #1651 from bhamiltoncx/python-lexer-parser-output-file 
Python 2 and Python 3 support for lexer and parser output to a file
 2017-02-13 15:10:19 -08:00
Ben Hamilton c7e7550a12 Fix CSharpTarget.java 2017-02-13 08:56:57 -08:00
Ben Hamilton 0049d6d9ae Emit language-specific Unicode escapes when generating code containing non-ASCII Unicode values 2017-02-10 14:31:56 -08:00
Terence Parr ae5250685d Merge pull request #1637 from bhamiltoncx/char-streams 
Make BaseJavaTest and TestRig work with Unicode values > U+FFFF
 2017-02-10 14:11:34 -08:00
Ben Hamilton 6611c1ae27 C# runtime and test harness support for writing test output to file 2017-02-10 13:37:44 -08:00
Ben Hamilton 2c447a2b75 Python 2 and Python 3 support for lexer and parser output to a file instead of stdout 2017-02-10 13:11:24 -08:00
Ivan Kochurkin 23f59ad99d Removed trailing comma in channelNames for Python 2 & 3, JavaScript. 2017-01-31 00:34:10 +03:00
Ivan Kochurkin 7b8ba3550b channelNames support in LexerInterpreter (C#, C++, Java, Swift). 2017-01-31 00:34:10 +03:00
Ivan Kochurkin ccb325db42 Removed trailing comma in channel names. 2017-01-31 00:32:21 +03:00
Ivan Kochurkin 9998db82ce Added channel constants to Go runtime. 2017-01-31 00:32:21 +03:00
Ivan Kochurkin f9fa84213c Swift: mode constants written to file as tokens and channels. 2017-01-31 00:32:21 +03:00
Ivan Kochurkin 956dc0547f C++: mode constants written to file as tokens and channels. 2017-01-31 00:32:21 +03:00
Ivan Kochurkin 1fe1216e26 C#: mode constants written to file as tokens and channels. 2017-01-31 00:32:21 +03:00
Ivan Kochurkin 00d0ba129b Mode constants written to file as tokens and channels. 
I mean:

```Java
public static final int M1=1, M2=2;
```

instead of

```Java
public static final int M1=1;
public static final int M2=2;
```
 2017-01-31 00:32:21 +03:00
Ivan Kochurkin e9e83a69e7 Added channel constants to Python 2 & 3 runtimes. 
Improved indentation.
 2017-01-31 00:32:21 +03:00
Ivan Kochurkin def7e6afc8 Added channel constants to JavaScript runtime. 
Improved indentation.
 2017-01-31 00:32:21 +03:00
Ivan Kochurkin c3d96eb5d7 Added channelNames to all runtimes. 2017-01-31 00:32:21 +03:00
Ben Hamilton 0f52b7c0d9 CharStreams 2017-01-30 10:09:56 -08:00
Terence Parr 0901851719 Merge pull request #1585 from KvanTTT/left-recursive-rule-labels 
The final fix (hopefully) for alternative labels check in left recursive rules
 2017-01-29 13:34:32 -08:00
Gary Miller 7cbef5617c removed lower case formatting on types. Case has meaning in Go, needs preserving. 2017-01-11 22:33:35 +11:00
Gary Miller 810000c23e removed superClass import 2017-01-11 22:21:29 +11:00
Ivan Kochurkin d71c2157ce Fixed @sharwell note about comment. 2017-01-08 15:08:48 +03:00
Ivan Kochurkin 4a5717162e Fixed indents for else statements. 
Fixed correct token in checkForTypeMismatch.
 2017-01-07 18:37:17 +03:00
Ivan Kochurkin a558334a01 Code cleared, added comments. 2017-01-07 18:26:22 +03:00
Ivan Kochurkin 385e96784d Initial fix for labels in left recursive rules: https://github.com/antlr/antlr4/pull/1570 & https://github.com/antlr/antlr4/issues/1543 2017-01-07 18:16:22 +03:00
Sam Harwell 587ea3646c Don't use list labels for elements that appear twice in a block set 2017-01-06 00:21:42 -06:00
Sam Harwell 204c281772 Fix optional element analysis 
* Properly handle elements that are optional in some alts but not others
* Properly handle block sets (a group of terminals producing a SetTransition)
* Properly handle OPTIONAL subrule
 2017-01-06 00:19:43 -06:00
Sam Harwell 9009a1b989 Simplify logic leaving a subrule in ElementFrequenciesVisitor 2017-01-04 11:46:47 -06:00
Sam Harwell eb53e70e39 Implement support for optional getters 
This analysis is required for proper code generation in the TypeScript target
when strict null checks are enabled. It also applies to targets intending to
differentiate optional values from required values.
 2017-01-04 09:43:20 -06:00
Sam Harwell 6b007b445c Disable label checks for left recursive rules 
See #1543
 2017-01-02 14:14:06 -06:00
Mike Lischke c874ab68fb Merge branch 'master' into optimizations 2017-01-01 17:40:48 +01:00
Mike Lischke a60fcfed0f Removed the need of a shadow s0 copy in the DFA class. 2017-01-01 17:39:44 +01:00
Terence Parr aa909ddcb1 Merge pull request #1563 from parrt/bump-to-4.6.1 
bump version to 4.6.1 in all files.
 2016-12-28 14:09:33 -08:00
parrt 2629433830 bump version to 4.6.1 in all files. 2016-12-28 12:03:43 -08:00
parrt 36ee17449f Fixes #1556. 
```
beast:/tmp $ a4.6 T.g4
org/antlr/v4/parse/GrammarTreeVisitor.g: node from line 2:7 no viable alternative at input '..'
org/antlr/v4/parse/GrammarTreeVisitor.g: node from line 2:7 no viable alternative at input '..'
org/antlr/v4/parse/GrammarTreeVisitor.g: node from line 2:7 no viable alternative at input '..'
org/antlr/v4/parse/GrammarTreeVisitor.g: node from line 2:7 no viable alternative at input '..'
org/antlr/v4/parse/GrammarTreeVisitor.g: node from line 2:7 no viable alternative at input '..'
context [/report INTERNAL_ERROR] 1:17 attribute arg isn't defined
error(20):  internal error:
beast:/tmp $ a4.6.1 T.g4
error(181): T.g4:2:4: token ranges not allowed in parser: 'A'..'Z'
```
 2016-12-28 11:47:48 -08:00
Sam Harwell 7a83cd4910 Fix code generation for LL1OptionalBlock bypass alternative 
See #1545
 2016-12-23 16:12:43 -06:00
parrt 3ce00ec933 fix some typos Sam noticed. 2016-12-22 09:43:18 -08:00
Mike Lischke 367f09dcbe Removed command line option for exportMacro. 
This is now handled via a grammar option.
 2016-12-21 09:11:31 +01:00
Mike Lischke a9dc3ebd86 Merge branch 'master_upstream' 2016-12-19 16:47:39 +01:00
parrt 23313ee44e [maven-release-plugin] prepare for next development iteration 2016-12-15 14:25:41 -08:00
parrt aacd2a2c95 [maven-release-plugin] prepare release 4.6 2016-12-15 14:25:36 -08:00
parrt 31726311a8 add exportMacro to list of valid grammar options. verified it works in grammar and -D on command line. 2016-12-15 13:25:02 -08:00
Mike Lischke d1b429d94c Added a new setting required by the C++ runtime. 
In order to export generated classes in DLLs we need a way to specify the __declspec setting. This is is usually done via a macro that takes the import or export value. The new parameter (`-export-macro`) allows to specify this, which increases the flexibility of the generated classes.

The C++ target documentation has been extended to describe build specific things, includig this new parameter.
 2016-12-15 13:11:10 -08:00
parrt 160825a86f minor reformatting. 2016-12-15 11:41:56 -08:00
parrt e0460f103b simplify; reduce indentation. 2016-12-15 11:32:48 -08:00
parrt b503838a59 simplify 2016-12-15 11:22:42 -08:00
parrt 642d06ebb2 fix typo 2016-12-15 11:15:56 -08:00
Ivan Kochurkin 5d3b99b2ee Fixed error message for EMPTY_STRINGS_AND_SETS_NOT_ALLOWED. 2016-12-15 11:08:09 -08:00
Ivan Kochurkin 75d5db5941 Implemented CHARACTERS_COLLISION_IN_SET handling. 
Implemented EMPTY_STRINGS_AND_SETS_NOT_ALLOWED for empty sets.
Improved INVALID_ESCAPE_SEQUENCE, fixes #1077
 2016-12-15 11:08:09 -08:00
Ivan Kochurkin 37148ff7fa Added CHARACTERS_COLLISION_IN_SET warning. 
EMPTY_STRINGS_NOT_ALLOWED rename to EMPTY_STRINGS_AND_SETS_NOT_ALLOWED.
 2016-12-15 11:08:09 -08:00
Mike Lischke 428ff28f35 Added a new setting required by the C++ runtime. 
In order to export generated classes in DLLs we need a way to specify the __declspec setting. This is is usually done via a macro that takes the import or export value. The new parameter (`-export-macro`) allows to specify this, which increases the flexibility of the generated classes.

The C++ target documentation has been extended to describe build specific things, includig this new parameter.
 2016-12-15 16:44:26 +01:00
parrt f38ccba94a Fixes #1514 2016-12-14 12:26:32 -08:00
parrt a3aa610e3d get recognizer name properly. Fixes #958 2016-12-13 11:05:39 -08:00
Terence Parr ec5f2ecf31 Merge pull request #1498 from parrt/fix-1299 
Fixes #1299. matching a set didn't properly notify error strategy of a successful match
 2016-12-13 08:30:49 -08:00
parrt a69299c507 set javadoc to deploy phase. 2016-12-13 08:17:13 -08:00
parrt 068a13e0f8 Update all other targets 2016-12-12 18:09:55 -08:00
parrt b11ff9636a Fixes #1299. matching a set didn't properly notify error strategy of a successful match. 2016-12-12 17:54:53 -08:00
parrt 2739a0c9ef Fixes #1492. only create javadoc if we use `mvn -DskipTests javadoc:jar install` 2016-12-12 17:33:39 -08:00
Eric Vergnaud 39de8896f8 Merge branch 'master' into realign-csharp-runtime 2016-12-12 21:54:26 +08:00
Terence Parr 08a8e30864 Merge pull request #1470 from gaulouis/issue-1457 
Issue 1457
 2016-12-11 15:14:36 -08:00
parrt 42086eff6e Fix #980 2016-12-11 14:56:11 -08:00
Eric Vergnaud 0170c9a064 Good progress with parser 2016-12-12 02:00:09 +08:00
Eric Vergnaud 19d70cd777 TestLexerExec tests pass 2016-12-12 00:52:09 +08:00
parrt 0dae6359ec Fixes #1007. 2016-12-10 18:29:09 -08:00
parrt 52cecbe9e1 generates tool javadoc jar now 2016-12-10 18:21:10 -08:00
parrt 98900144b2 attempting to get src/javadoc back. 2016-12-10 18:09:37 -08:00
parrt 731603b489 maven shade and bundle don't work together so i dropped bundle for main complete jar. 2016-12-10 17:31:03 -08:00
parrt c0e7c49a9e update jar version, add implementation manifest entries. Fixes #1036 2016-12-10 16:43:10 -08:00
parrt 9b758cb668 update versions 2016-12-10 16:07:35 -08:00
parrt 0de8a9c414 simplify tool pom 2016-12-10 15:32:39 -08:00
parrt e65154c161 Fix #1073. Better token output. 2016-12-10 14:34:50 -08:00
parrt 0eb27279de fix LL1 analyzer 2016-12-09 18:07:47 -08:00
gaulouis 258b1a4a38 Redo dynamic cast 2016-12-09 21:37:52 +01:00
gaulouis e9859bcd2f Fix unnecessary cast 2016-12-09 18:43:07 +01:00
gaulouis ea53dac7b7 Fix antlr4:: namespace pollution in Lexer/Parser headers 2016-12-09 18:35:55 +01:00
Terence Parr 2dddd007e4 Merge pull request #1462 from pboyer/lafix 
Fix for #1459
 2016-12-08 12:19:14 -08:00
Terence Parr 11117e02d4 Merge pull request #1458 from KvanTTT/mixed_type_labels_1409 
Check labels for tokens with different types
 2016-12-08 11:47:09 -08:00
parrt 2c99f187a0 make it read "Copyright (c) 2012-2016 The ANTLR Project" 2016-12-08 11:43:06 -08:00
parrt 1b1a9862c4 contributors->authors 2016-12-08 11:43:06 -08:00
parrt 61e8eba4fc fix typo 2016-12-08 11:43:06 -08:00
parrt 891915de84 Update copyright on java, xml files. Make BSD license conform to standard. 2016-12-08 11:40:30 -08:00
Peter Boyer 0cb32c0620 Fix for repeat declaration of la_ 2016-12-07 21:03:03 -05:00
Ivan Kochurkin bdc1911e6e Removed not used LabelType. 2016-12-07 22:40:27 +03:00
Ivan Kochurkin d5dda9f998 Added check for labels with different token types. 
Spaces -> Tabs.
 2016-12-07 22:39:27 +03:00
Terence Parr abed9e733a Merge pull request #1449 from KvanTTT/duplicated_and_nonsensical_commands_1388 
Duplicated and incompatible commands
 2016-12-05 13:40:10 -08:00
Ivan Kochurkin e4c765aa27 Added comment. 2016-12-04 15:11:25 +03:00
Ivan Kochurkin bf3bff16e4 Misprints fixed, code style improved (removed unnecessary else). 2016-12-04 15:06:07 +03:00
Ivan Kochurkin cb16e369fe Misprint in comment. 2016-12-04 04:05:55 +03:00
Ivan Kochurkin 0e1b5d9d7a Merge branch 'command_wrong_id_1411' into duplicated_and_nonsensical_commands_1388 
# Conflicts:
#	tool/src/org/antlr/v4/tool/ErrorType.java
 2016-12-04 03:57:11 +03:00
Terence Parr f11a380609 Merge pull request #1417 from KvanTTT/command_wrong_id_1411 
ANTLR generates a wrong lexer code if type, channel or mode commands use inproper identifier
 2016-12-03 16:55:23 -08:00
Ivan Kochurkin 4e2f50aef8 Added new error types for not recognized token, mode or channel name. 2016-12-04 03:47:28 +03:00
Ivan Kochurkin b8585b7eb5 Fixed some missed cases for checkCommands. 2016-12-04 02:32:39 +03:00
Terence Parr 8ead51d8ff Merge pull request #1423 from wxio/master 
Fix for redeclared error is generated Go code.
 2016-12-01 10:51:40 -08:00
Peter Boyer f00e346ccd Add additional _alt decl 2016-11-30 11:18:26 -05:00
Peter Boyer ced8fba5bc Update Go.stg 2016-11-29 23:50:39 -05:00
Peter Boyer 39c77ca20f Potential fix for antlr/antlr4 #1397 2016-11-29 23:45:29 -05:00
Denis Blank c8279a1d89 Fix a dangerous c-cast in the cpp codegen 2016-11-29 17:14:02 +01:00
Ivan Kochurkin 9aab338021 Reverted old error types and numbers. 2016-11-29 19:12:43 +03:00
Ivan Kochurkin 44aab113f3 Added method checkCommands. 
commands check micro optimization and code clearing.
 2016-11-29 18:48:03 +03:00
Ivan Kochurkin f76ca22763 MODE_CONFLICTS_WITH_TOKEN error restored. 2016-11-29 18:13:57 +03:00
Ivan Kochurkin 8fe4def4ed Warnings for incompatible commands, added unit-test, fixes #1388. 
Improved test for duplicated commands (added pushMode, popMode).
 2016-11-29 16:03:33 +03:00
Ivan Kochurkin ed7f6f70da pushMode and popMode commands are not duplicated. 2016-11-29 16:00:41 +03:00
Gary Miller 763669ffc7 Fix for redeclared error is generated Go code. Create block so _x and _m var are created in a local scope. 2016-11-29 12:35:48 +11:00
Ivan Kochurkin bd1111ffa3 Fixed issue with duplicated commands in lexer rule. 2016-11-28 19:48:35 +03:00
Ivan Kochurkin a1def96500 getConstantValue replaced with getModeConstantValue, getTokenConstantValue and getChannelConstantValue methods. 
ErrorType simplified, tests fixed.
Fixes #1411, added unit-test.
 2016-11-28 14:41:04 +03:00
parrt 3506dcef1d add @Override to hush warnings. Manually did some of https://github.com/antlr/antlr4/pull/1066 2016-11-25 12:52:44 -08:00
Sam Harwell f196971565 Fix failure to identify prefix rules that also recurse 
Fixes #1402
 2016-11-24 14:38:34 -06:00
parrt 6c2d4e6225 Fixes #1211 2016-11-22 15:04:28 -08:00
parrt dc2c9b039e Fixes #1188 2016-11-22 14:21:07 -08:00
parrt 4e6e7edf0d Fixes #1317 2016-11-21 13:45:05 -08:00
Terence Parr 8ed1105290 Merge pull request #1353 from marcohu/fix_1233 
Grammar dependency management for Maven plugin
 2016-11-21 11:21:58 -08:00
parrt d580f2184f Use JFrame instead of JDialog to get "expand window" buttons 2016-11-21 10:42:52 -08:00
Marco Hunsicker cfc538bfdc Merge branch 'master' of https://github.com/antlr/antlr4 into fix_1233 to solve 
Conflicts:
	contributors.txt
 2016-11-21 19:23:48 +01:00
parrt 66be087c38 merge old and new travis; improve so it runs C++ on osx; update BaseSwiftTest so it mirrors other targets; remove unneeded dep in pom to some json stuff 2016-11-20 19:50:54 -08:00
Marco Hunsicker a077d3d072 Merge branch 'master' of https://github.com/antlr/antlr4 into fix_1233 2016-11-20 12:49:39 +01:00
hanjoes e0b326677b Migrating the current Swift tests to the new testing framework. 2016-11-19 18:30:48 -08:00
janyou cddb8a6ff2 Update Swift.stg 
add namedActions
 2016-11-19 18:30:48 -08:00
janyou abf2a1883c Update SwiftTarget.java 
"4.5.1"->"4.5.3"
 2016-11-19 18:30:48 -08:00
janyou fc3fa24464 Update Swift.stg 2016-11-19 18:30:48 -08:00
janyou 61248c8a4d Update Swift.stg 2016-11-19 18:30:48 -08:00
Ewan Mellor 6d6327aa9d Move the Token declarations into an enum. 
This avoids the naming conflict in the tests between the use of ATN as a
token name and as a type name.
 2016-11-19 18:30:48 -08:00
Ewan Mellor 869bcfe429 Mark matchWildcard call with a try, since it can throw. 2016-11-19 18:30:48 -08:00
Ewan Mellor 6d5d1f1c23 Add build system changes needed for Swift backend. 2016-11-19 18:30:48 -08:00
Ewan Mellor c989064e32 Added (part of) a Swift backend. 
This comes from https://github.com/janyou/ANTLR-Swift-Target and is
marked Copyright (c) 2016 janyou on top of the BSD license and Copyrights
for Terence Parr and Sam Harwell derived from the original ANTLR source.
 2016-11-19 18:30:48 -08:00
Terence Parr 10ce5d2625 Merge pull request #1311 from sebkur/allow_path_in_token_vocab 
Allow paths as tokenVocab option
 2016-11-19 15:30:19 -08:00
parrt f5d3e4cfe6 Improve robustness and correctness of two functions used by intellij antlr4 plugin to show ambig and lookahead parse trees. 2016-11-19 13:06:49 -08:00
parrt d0620a65f5 Merge branch 'issue#1295' of git://github.com/harriman/antlr4-updates into harriman-issue#1295 2016-11-19 12:05:32 -08:00
parrt 5153154658 Merge branch 'fix_typos' of git://github.com/sebkur/antlr4 into sebkur-fix_typos 2016-11-19 11:52:17 -08:00
Marco Hunsicker 81efb0e366 Merge branch 'master' of https://github.com/antlr/antlr4 into fix_1233 2016-11-18 17:53:54 +01:00
Marco Hunsicker 7d83a3aa9c More cleanup; use save Collection comparison 2016-11-17 23:29:27 +01:00
parrt 5adf79121f revert Go "fix" 2016-11-17 11:16:22 -08:00
Terence Parr cf5a79a5b2 Merge pull request #1362 from sridharxp/stgwotempvar 
Stgwotempvar
 2016-11-17 19:26:34 +01:00
Kurt Harriman ab8d17472f ANTLR4 generated incorrect code to access return value of recursive rule 
Incorrect code was generated for $e.v in a rule like this one:

    e returns [int v] ::=
           INT       {$v = $INT.int;}
        | '(' e ')'  {$v = $e.v;}
        ;

After parsing "(99)" the result would have v == 0 instead of 99.
 2016-11-16 22:38:11 -08:00
parrt 4f3c7738f3 ignore :: in C++ arg declarations; add test. 2016-11-16 12:34:14 -08:00
parrt 5aae2edb24 support x:T notation in rule arguments, update documentation, improve [...] scope parsing testing 2016-11-16 11:54:23 -08:00
hanjoes 03fa75efba differentiate prefix/postfix type decl. 2016-11-16 11:44:39 -08:00
sridharxp 6e09f08348 InvokeRule without Fix without Temporary variable 2016-11-16 13:14:38 +05:30
parrt 08e9776a4a shift template-based runtime test to legacy, new mechanism uses annotation to get multi-line strings into java. all tests pass minus about 15 in C# 2016-11-15 11:08:19 -08:00
Marco Hunsicker be585f6a8f Make graph accessible to the outside 2016-11-14 00:29:51 +01:00
parrt da342af01e added sync() per other targets to LL(1) decisions 2016-11-07 14:48:20 -08:00
parrt f8bec27d6d bring Go target templates up to same args as others. 2016-11-07 11:07:46 -08:00
parrt 3d0678508b reuse code to builld named actions. all file model objects have it now. 2016-11-07 11:07:23 -08:00
parrt 51d38110be trying a refactoring to test if i can push back 2016-11-07 10:55:48 -08:00
Mike Lischke 1c3a5879e6 Set required runtime version to 4.6. 2016-11-07 11:22:46 +01:00
Mike Lischke 92b032820e Merge branch 'master_upstream' 2016-11-07 11:19:12 +01:00
Mike Lischke 3b3fe3da57 Merge branch 'master_upstream' 
Had to resolve a number of conflicts + added code required for the C++ target to generate header files, after the recent changes in the main repo.
 2016-11-07 11:16:02 +01:00
parrt ffaedd209d Set version in code to 4.6. 2016-11-06 11:21:06 -08:00
parrt e5ec462e22 Fixes #1337. Add's sync() call back to LL(1) decisions; all targets updated (causing failed test in Go). Altered test for new output of errors. regen'd tests. 2016-11-06 09:45:40 -08:00
parrt 95c850e5f8 set version in prep for 4.6 2016-11-05 16:48:36 -07:00
Peter Boyer fc35818c52 Update import statements after merge 2016-11-05 16:36:53 -04:00
parrt 0799e838bb weird. missed a few files. these are to simplify get filename functionality. 2016-11-05 09:32:04 -07:00
parrt abdb0b6a46 Merge branch 'model-updates-for-new-targets' of github.com:parrt/antlr4 into reduce-model-changes 2016-11-05 09:29:25 -07:00
parrt 2c98b5d325 incorporate in a single commit all changes needed to support generating header files. many fewer diffs with C++ target codegen now. 2016-11-05 09:25:51 -07:00
parrt 6d022379bc Pull in parrt's branch 'model-updates-for-new-targets' which appears to have freshened some non Go stuff as well. Simplify Go filename construction. 2016-11-05 08:48:37 -07:00
parrt 40e19e5d37 refactor so getRecognizerFileName() etc... delegate to Target; default is same behavior. adds "is header" arg as well in prep for C++ 2016-11-05 08:08:08 -07:00
parrt 75794a588b track different types of StructDecl elements so Go can do setters/getters. 2016-11-05 07:35:57 -07:00
parrt c0cedc424b track different types of StructDecl elements so Go can do setters/getters. 2016-11-05 07:35:39 -07:00
Mike Lischke e71e95080a Merge branch 'master_upstream' 
Also updated the list of ignored tests in the Cpp.test.stg file (lke it is done for the other targets).
 2016-11-05 15:07:59 +01:00
parrt 8b8094c044 appears to be an ST bug that doesn't see arrays of len 0 as missing; convert array to List. 2016-11-05 06:55:09 -07:00
parrt e0521400cb rm need for isRuleContext in templates; refer to underlying grammar object. 2016-11-05 06:28:58 -07:00
parrt 85f2a87840 rm recognizerName field where possible 2016-11-04 16:07:45 -07:00
parrt 95dbf7b44b look for go in path during tests 2016-11-04 16:06:41 -07:00
Sebastian Kürten 2f628a1106 Fix another typo 2016-11-01 11:56:46 +01:00
Sebastian Kürten 8b16ff180a Fix a few typos in Javadoc 2016-11-01 11:46:04 +01:00
kai stammerjohann f3c55be33c fix RulePropertyRef_stopHeaderHeader typo 2016-11-01 09:17:46 +01:00
Mike Lischke f794820aaf Reworked mutex handling + updated C++ template for a var init 
- Mutexes have been consolidated. Instead of one per DFA (which can easily get to hundreds of them) we only have one mutex in the Recognizer class and all other parties use this for serialization. It's only about protected the DFA anyway, which is stored in a recognizer (lexer/parser).
- ATNState::getStateType() returns a size_t value now (actually an enum).
- Replaced checks via RTTI for transitions by the (serialization) type of the transition, for simplicity.
- Added some missing initialization for fields in certain ATN state classes.
- Fixed mem leak in DFA by shadowing the s0 field. That way still have a ref to the self created instance, even is s0 was replaced later.
- Added variable init in code generation for a rule context declaration (e.g. for labels).
 2016-10-29 14:46:38 +02:00
Sebastian Kürten 961f087d7a Allow paths as tokenVocab option 2016-10-25 10:31:04 +02:00
Peter Boyer 2444386091 Remove JavaScript PORT_DEBUG, re-enable runtime tests for other language targets (#69) 
* Re-enable runtime tests for other language targets

* Fix typo in js code

* Remove PORT_DEBUG statements

* Fix missing

* Fix testAltNum
 2016-10-13 17:06:22 -04:00
parrt 4312b37828 pull test changes in 2016-10-13 10:58:19 -07:00
Peter Boyer 42347c078b Fix merge failure 2016-10-12 09:53:43 -04:00
Peter Boyer 60662c4b70 Fix remaining tests (#64) 
* Fix pred context typo

* Adjust array lengths

* PositionAdjustLexer test fix first steps

* allow virtual method calls from Lexer

* Hopefully fix issues with dup namedActions

* Make parser members

* Fix testParserProperty

* Fix LeftRecursion test failures
 2016-10-12 09:49:42 -04:00
Mike Lischke a4e343806f Some small optimizations. 
- No copy of a DFA state during ATN deserialization.
- Fixed memory leak of locally created s0 state in DFAState.
 2016-10-12 11:06:51 +02:00
Mike Lischke 4897bd8571 Simplified memory management for volatile parsing part. 
There are 2 parts in an ANTLR genrated parser where memory is allocated: the actual parsing (with or w/o creating a parse tree) and the prediction part (via DFA/ATN etc.). The first part is highly volatile as it recreates parse tree instances (the class) on each parser run. In fact also lexer tokens belong to that part, but are already managed via unique pointers. This first part works without any smart pointer now. Instead there is a simple tracker class which holds all created references and frees them when the parser is reset or destroyed. This is a bit less optimal if the parser is set to create no parse tree, as created rule context objects are not freed immediately (like with smart pointers), but during reset. On the other hand this change gives (depending on the input) a nice speed up (0%-100%, after the warm up phase). Additionally memory consumption drops by a good amount.

Everything in the simulartors (and interpreters) remains unchanged. This is the shared prediction part.
 2016-10-11 13:49:02 +02:00
Christian Gudrian cb590334bb the visit methods now return a value 2016-10-06 22:52:19 +02:00
Christian Gudrian eb292adb18 implemented templates for the JavaScript visitors 2016-10-06 22:52:19 +02:00
Peter Boyer 9b76bc6a71 Update Go.stg 2016-10-05 12:04:01 -04:00
Mike Lischke a7b9499230 Code generation + Travis CI fixes. 
- Had to adjust a comparison <= 0 to the new unsigned EOF.
- For testing: extended the runtime tests to all C++ tests.
- Install uuid-dev in Travis CI in order to be able to build the C++ runtime.
 2016-10-03 11:25:02 +02:00
Mike Lischke 8280beb564 Refactoring of further int values, mostly converting them to size_t. 
- Switched most symbolic signed constants to unsigned variants. Redefined EOF in particular to become (size)-1, to avoid having to use signed token type values.
- Introduced INVALID_INDEX for all previous -1 values to indicate e.g. not found indexes etc.
- Added 2 helpers to convert between symbolic and numeric form (mostly for intervals and toString()).
- Removed many no longer needed type casts to size_t.
- Updated templates for these changes.
- Limited runtime tests to C++ tests only, to see how Travis CI copes with that.
 2016-10-02 16:51:57 +02:00
Mike Lischke 4394cd20b0 Merge branch 'master_upstream' 2016-09-29 16:50:57 +02:00
Terence Parr e9aa00e05b Merge pull request #1172 from renatahodovan/default-pass 
Add missing default pass statements to Python targets.
 2016-09-25 11:24:48 -07:00
Christian Gudrian 46df58d5d7 The JavaScript ParseTreeVisitor now implements a depth-first traversal by default. 2016-09-22 08:28:11 +02:00
Mike Lischke b0fc8ea0da Added semicolon generation after some of the header decls. 
Closes issue #40.
 2016-09-06 09:51:31 +02:00
Mike Lischke 6f62821573 More optimizations of the runtime. 
- Lesser use of shared_ptr, e.g. in listeners and some loops.
- Removed useless access methods for children in ParseRuleContext. The child list is public. Fixed initialization for start and stop nodes.
- Simplified parent + child organization in Tree and all derived classes. Instead of using overridable functions in various descendants we have now central parent + child fields in the base tree class (where they belong actually, considering this is about forming a tree). Users have to cast to the appropriate classes if necessary.
- Removed obsolete getChildren() function in Trees helper. We can just return the child vector.
- Changed edges member to an unordered_map, as this is a sparse container. This speeds up certain grammars by 1000% (e.g. highly recursive expression rules) and avoids wasting a lot of memory. This change also simplifies handling significantly.
 2016-08-22 12:12:22 +02:00
Mike Lischke ee2510e654 Output formatting in DefaultErrorStrategy and init of generated member vars. 
- Had to escape tabs + linebreaks in DefaultErrorStrategy when generating a text representation. Also removed a few explicit string instance creations on the way.
- Member vars in parser context classes that take (optional) Token references must be initialized.
- Fixed a warning that copyFrom() would hide a virtual function in a ParserRuleContext.
- Another attempt to limit genrating double semicolons.
 2016-08-10 08:55:17 +02:00
Mike Lischke 8335d32b2d Various smaller changes to fix compilation and other problems. 
- Added new rule to test grammar to get code generation for wildcard capture.
- Updated the Cpp.stg template file for that.
- Made the Unicode hack (auto extend 0xFFFF to 0x10FFFF) dependent on a parameter, so we only use this hack when deserializing an ATN. This avoids trouble with intervals used in other contexts (like string offsets).
- Added a few operator != overloads, to fix compilation after recent changes.
- Simplified operands comparison in SemanticContext (uses the Arrays class now). Some cleanup in that class too.
- The abstract parse tree visitor now uses const& for Any references, to avoid reallocating new instances over and over again.
- The lexer counts syntax errors the same way as the parser does. So we can directly determine if there was any error by simply examining that (which avoids having to use a temporary listener).
 2016-08-04 09:28:13 +02:00
Mike Lischke 8a5f6815fe Rework of the visitor implementation 
The translation from Java generics to templates in C++ lead to the need of virtual template functions, which is not supported by C++. Instead we use now the Any class for results of visits and no longer need templates for that part.
 2016-07-30 20:03:27 +02:00
Will Faught 7f42bc602b Add interface build checks for tree types and generated base listeners 2016-07-18 12:19:31 -07:00
Nathan Burles 9fa56a4cd9 Added lexer channels enum to match Java target 2016-07-07 14:04:01 +01:00
Will Faught a05dea54ec Fix test case build errors 2016-07-06 09:02:42 -07:00
Mike Lischke c462e03126 Merge branch 'master_upstream' 2016-06-26 11:12:23 +02:00
Mike Lischke 2352ff03e1 getSerializedATN() wasn't properly overwritten. 2016-06-26 11:10:31 +02:00
Eric Vergnaud 15430d4bf9 remove unwanted semicolon in generated Python code 2016-06-23 15:06:49 +02:00
Mike Lischke 4a7ec859aa Simplified parse tree listener handling. 
No need to use shared_ptr for management. Listeners are, like the other main classes (parser, lexer, input stream etc.) provided by the application and hence managed there.
 2016-06-17 10:22:19 +02:00
Peter Boyer c8a9d75cfe Merge pull request #21 from wjkohnen/gofmt 
tool/Go: gofmt generated code.
 2016-06-14 16:26:36 -04:00