This Bugzilla instance is a read-only archive of historic NetBeans bug reports. To report a bug in NetBeans please follow the project's instructions for reporting issues.
Build 200407270649, j2sdk1.4.2_05 Summary is selfexplaining. Try to write e.g. FFF fff = ""; in the source editor. The line should be underscored, but it isn't. - REGRESSION
Probably caused by last gjast integration.
Tom, for now I rolled back the changes in gjast, javacore and parser related to the new scanner and deprecation support, since tonight we will have a q-build and it seems they caused regressions that you won't be able to fix this week (since you are on vacation). For this reason I am also closing this bug - it should be reopened/verified once the changes are integrated again. Checking in external/gjast.jar.scrambled; /cvs/java/external/gjast.jar.scrambled,v <-- gjast.jar.scrambled new revision: 1.54; previous revision: 1.53 done Processing log script arguments... More commits to come... Checking in javacore/src/org/netbeans/modules/javacore/parser/TokenIterator.java; /cvs/java/javacore/src/org/netbeans/modules/javacore/parser/TokenIterator.java,v <-- TokenIterator.java new revision: 1.6; previous revision: 1.5 done Processing log script arguments... More commits to come... Checking in javacore/src/org/netbeans/modules/javacore/scanning/JavaUpdater.java; /cvs/java/javacore/src/org/netbeans/modules/javacore/scanning/JavaUpdater.java,v <-- JavaUpdater.java new revision: 1.5; previous revision: 1.4 done Processing log script arguments... More commits to come... Checking in parser/src/org/netbeans/lib/java/parser/Factory.java; /cvs/java/parser/src/org/netbeans/lib/java/parser/Factory.java,v <-- Factory.java new revision: 1.4; previous revision: 1.3 done Checking in parser/src/org/netbeans/lib/java/parser/JScanner.java; /cvs/java/parser/src/org/netbeans/lib/java/parser/JScanner.java,v <-- JScanner.java new revision: 1.4; previous revision: 1.3 done Checking in parser/src/org/netbeans/lib/java/parser/Scanner.java; /cvs/java/parser/src/org/netbeans/lib/java/parser/Scanner.java,v <-- Scanner.java new revision: 1.5; previous revision: 1.4 done Checking in parser/src/org/netbeans/lib/java/parser/ScannerToken.java; /cvs/java/parser/src/org/netbeans/lib/java/parser/ScannerToken.java,v <-- ScannerToken.java new revision: 1.4; previous revision: 1.3 done Checking in parser/src/org/netbeans/lib/java/parser/Token.java; /cvs/java/parser/src/org/netbeans/lib/java/parser/Token.java,v <-- Token.java new revision: 1.4; previous revision: 1.3 done Checking in parser/src/org/netbeans/lib/java/parser/scanner.l; /cvs/java/parser/src/org/netbeans/lib/java/parser/scanner.l,v <-- scanner.l new revision: 1.5; previous revision: 1.4 done
*** Issue 46512 has been marked as a duplicate of this issue. ***
Today's build seems to be broken. Error annotations doesn't work, only this message is written to console: "(source unavailable)" This commit seems to be responsible: User: tball Date: 04/08/04 23:10:21 Modified: parser/src/org/netbeans/lib/java/parser Factory.java JScanner.java Scanner.java ScannerToken.java Token.java scanner.l Log: Resubmission of fast scanner changes: tests pass, and now Jesse's project scans without errors due to handling of null sourceLevel values in the fast scanner. Tom, can you check it? Thanks.
The problem is due to a change in javac where it now uses source buffer offsets as positions instead of packed integers (a change requested by the NetBeans team two years ago). The javac Log class (its error reporter) now has to re-read the source to determine the line and column of an error, but the editor doesn't pass in a full file path when doing error checking. This prevented Log from reporting line and column numbers in errors. The fix I am now testing adds a method to Log so that the gjast error checker (called indirectly by the editor) can give Log access to the source buffer so it doesn't need to open any files. Since I also updated Log to use chars instead of bytes, I am also waiting on approval from the javac team that they will incorporate these changes.
Got approval from javac team for proposed and tested changes.
Verified
Still does not work for semantic errors (for syntactic it does). I had a class where I used identifier Utilities which was unresolved. No error annotation was displayed. So I looked at the console window and saw the error was found by the parser but the line number was missing. This was the output: cannot find symbol symbol : variable Utilities location: class org.netbeans.modules.javacore.parser.WildCardRef Note that for syntactic errors it is a bit different: org/netbeans/modules/javacore/parser/WildCardRef.java:49: ';' expected kjh return Utilities.compareObjects(bound, ref.bound);
Jan Becicka tried to write regression tests for error annotations to make the verification easier, but he is not able to reproduce this bug in the test project. So it seems that sometimes it works. We are only able to reproduce it manually (perhaps, you Tom, as you will investigate it, will be able to provide a suitable testcase). Here are the steps to reproduce it: 1) open refactoring project and all required projects 2) open org.netbeans.modules.javacore.parser.WildCardRef 3) delete the last import statement in that file (import org.openide.util.Utilities;) -> there should be an error annotation on line 50 (return Utilities.compareObjects(bound, ref.bound);), but it is not there and the message in the console says: (source unavailable) cannot find symbol symbol : variable Utilities location: class org.netbeans.modules.javacore.parser.WildCardRef
Cached source buffers so that error logger can figure out line and column positions.
*** Issue 47125 has been marked as a duplicate of this issue. ***
Fix required for NB 4.0 Beta 1. Please integrate asap. Thanks.
Done. Checking in gjast.jar.scrambled; /cvs/java/external/gjast.jar.scrambled,v <-- gjast.jar.scrambled new revision: 1.57.2.1; previous revision: 1.57 done
filed another issue 47425
*** Issue 47442 has been marked as a duplicate of this issue. ***
Reorganization of java component