Re: [fpc-devel] wrong rtti default value in the fixes_2_2 branch (dont know about trunk)
Michael Van Canneyt wrote: I fixed the bug in trunk. Please do some tests in Lazarus with the 12114 revision of the compiler. If all works still OK and the testsuites don't give any regressions, I'll merge it to the fix branch. Here nothing bad happen - at least I had not note. If you have no related tracker issues then maybe you will merge your fix? Best regards, Paul Ishenin. ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-devel
Re: [fpc-devel] wrong rtti default value in the fixes_2_2 branch (dont know about trunk)
On Fri, 21 Nov 2008, Paul Ishenin wrote: Michael Van Canneyt wrote: I fixed the bug in trunk. Please do some tests in Lazarus with the 12114 revision of the compiler. If all works still OK and the testsuites don't give any regressions, I'll merge it to the fix branch. Here nothing bad happen - at least I had not note. If you have no related tracker issues then maybe you will merge your fix? I will do so tonight. Michael. ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-devel
Re: [fpc-devel] wrong rtti default value in the fixes_2_2 branch (dont know about trunk)
On Fri, 21 Nov 2008, Paul Ishenin wrote: Michael Van Canneyt wrote: I fixed the bug in trunk. Please do some tests in Lazarus with the 12114 revision of the compiler. If all works still OK and the testsuites don't give any regressions, I'll merge it to the fix branch. Here nothing bad happen - at least I had not note. If you have no related tracker issues then maybe you will merge your fix? Merged. Michael. ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-devel
[fpc-devel] wrong rtti default value in the fixes_2_2 branch (dont know about trunk)
Hello, FPC developers' list Here is a small example attached which shows that rtti return wrong default values for properties derived from ancestor. Please fix it - our object inspector dont like such behavior :) Best regards, Paul Ishenin. program rtti; {$mode objfpc}{$H+} {$apptype console} uses {$IFDEF UNIX}{$IFDEF UseCThreads} cthreads, {$ENDIF}{$ENDIF} Classes, typinfo { you can add units after this }; type { TSomeBaseClass } TSomeBaseClass = class(TPersistent) private FSomeProperty: Integer; public property SomeProperty: Integer read FSomeProperty write FSomeProperty default 10; end; { TSomeDerivedClass } TSomeDerivedClass = class(TSomeBaseClass) private FOwnProperty: Integer; published property SomeProperty; property OwnProperty: Integer read FOwnProperty write FOwnProperty default 11; end; var DC: TSomeDerivedClass; Info: PPropInfo; begin DC := TSomeDerivedClass.Create; Info := GetPropInfo(DC, 'SomeProperty'); WriteLn(Info^.Default); Info := GetPropInfo(DC, 'OwnProperty'); WriteLn(Info^.Default); end. ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-devel
Re: [fpc-devel] wrong rtti default value in the fixes_2_2 branch (dont know about trunk)
On 15 Nov 2008, at 09:29, Paul Ishenin wrote: Here is a small example attached which shows that rtti return wrong default values for properties derived from ancestor. Please fix it - our object inspector dont like such behavior :) Please submit a bug report. I guess it's probably caused by the merge of r11682. Jonas ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-devel
Re: [fpc-devel] wrong rtti default value in the fixes_2_2 branch (dont know about trunk)
On Sat, 15 Nov 2008, Jonas Maebe wrote: On 15 Nov 2008, at 09:29, Paul Ishenin wrote: Here is a small example attached which shows that rtti return wrong default values for properties derived from ancestor. Please fix it - our object inspector dont like such behavior :) Please submit a bug report. I guess it's probably caused by the merge of r11682. No, because the bug also exists in trunk. Michael. ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-devel
Re: [fpc-devel] wrong rtti default value in the fixes_2_2 branch (dont know about trunk)
On Sat, 15 Nov 2008, Michael Van Canneyt wrote: On Sat, 15 Nov 2008, Jonas Maebe wrote: On 15 Nov 2008, at 09:29, Paul Ishenin wrote: Here is a small example attached which shows that rtti return wrong default values for properties derived from ancestor. Please fix it - our object inspector dont like such behavior :) Please submit a bug report. I guess it's probably caused by the merge of r11682. No, because the bug also exists in trunk. I fixed the bug in trunk. Please do some tests in Lazarus with the 12114 revision of the compiler. If all works still OK and the testsuites don't give any regressions, I'll merge it to the fix branch. Michael. ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-devel