Problem description: Every report execution, launched from inside the report editor, stops on error with the message (french version): Une exception de type com.sun.star.lang.WrappedTargetException est survenue. caught a java.lang.IncompatibleClassChangeError LO Version: 4.1.0.0.beta2 Build ID: 33224f4f11a05cfad2249e812fcc2975fbb61f6 Java version "1.7.0_25" Java(TM) SE Runtime Environment (build 1.7.0_25-b16) Steps to reproduce: 1. Start a database 2. Click on an old report Current behavior: cannot open "report", caught an java.lang.IncompatibleClassChangeError Expected behavior: execute the report Operating System: Windows 7 Version: 4.1.0.0.beta2 Last worked in: 3.6.6.2 release
Sound like it could be the same issue as in bug 65951. Creating cross-linkage.
(In reply to comment #1) > Sound like it could be the same issue as in bug 65951. Creating > cross-linkage. Couldn't be the same. It is reported for 4.1.0.0.beta2. The crash exists only with Lo 4.0.4.2.
Just installed Version: 4.1.0.1 Build ID: 1b3956717a60d6ac35b133d7b0a0f5eb55e9155 All reports are working again, on simple invocation or inside the report editor. Many Thanks and kudos to all involved developers !
(In reply to comment #3) > Just installed Version: 4.1.0.1 > Build ID: 1b3956717a60d6ac35b133d7b0a0f5eb55e9155 > > All reports are working again, on simple invocation or inside the report > editor. > Many Thanks and kudos to all involved developers ! So we could set this to RESOLVED and WORKSFORME.
After installing 4.1.0.4 I get exactly this behaviour using Java 1.7.0_25 or 1.6.0_45 described in Bug 65937 : Running a report always ends with the error message "caught a java.lan.incompatibleClasssChangeError" I checked with reports that worked in previous LO versions as well as with newly created reports on a MYSQL database (tested with ODBC as well as JDBC connection)
(In reply to comment #5) > After installing 4.1.0.4 I get exactly this behaviour using Java 1.7.0_25 or > 1.6.0_45 described in Bug 65937 : Running a report always ends with the > error message "caught a java.lan.incompatibleClasssChangeError" > > I checked with reports that worked in previous LO versions as well as with > newly created reports on a MYSQL database (tested with ODBC as well as JDBC > connection) All reports work for me in LO 4.1.0.4, Windows-XP and Windows-7, Java version "1.7.0_25" so I cannot confirm the reopening.
Created attachment 83301 [details] this is the error message I receive every time I execute any report Though the error persists with my installation. I'm able to configure a report but not to execute it. Version: 4.1.0.4 Build ID: 89ea49ddacd9aa532507cbf852f2bb22b1ace28 Windows 7 Ultimate x64 Service Pack 1 32GB RAM, AMD FX-8350 Java 1.7.0_25 (32 bit selected; parallel installed also 64 bit java) Report Builder 1.2.3
(In reply to comment #7) > Created attachment 83301 [details] > this is the error message I receive every time I execute any report > > Though the error persists with my installation. I'm able to configure a > report but not to execute it. > > Version: 4.1.0.4 > Build ID: 89ea49ddacd9aa532507cbf852f2bb22b1ace28 > > Windows 7 Ultimate x64 Service Pack 1 > 32GB RAM, AMD FX-8350 > > Java 1.7.0_25 (32 bit selected; parallel installed also 64 bit java) > > Report Builder 1.2.3 After removing "Report Builder 1.2.3" from the list of extensions now everything works. Obviously Report Builder is an internal feature now and no longer considered an extension. Maybe the entry was an old one that stayed there while migrating across several versions of LibreOffice. So - it seems that solved my problem and should be a hint to everyone experiencing the same strange behavior.
(In reply to comment #8) > > After removing "Report Builder 1.2.3" from the list of extensions now > everything works. Obviously Report Builder is an internal feature now and no > longer considered an extension. Maybe the entry was an old one that stayed > there while migrating across several versions of LibreOffice. > So - it seems that solved my problem and should be a hint to everyone > experiencing the same strange behavior. So we could close this bug-report again.