Unicode doesn't show in error dialogs

[mlerch]

Unicode does not show in error dialogs, it simply shows question marks. It seems to only support the current character set. It also does not show in the Viewer. I saw a thread from way back in 2005 suggesting the Viewer now supports it. I don't see it.

To repeat: simply put an edit box on a form, then on a button click:

raise exception.create(edit1.text);

In the Edit box, enter characters which are not from the current character set. only characters from the current Windows char set display properly.

Thanks,
Mark

[Alex]

Hi.

There is no full unicode support in old Delphi versions for the reason that the Delphi itself does not support unicode in RTL classes (only ANSI).
The current EurekaLog version for Delphi 2009 is still using ansi-strings (i.e. all string types was just appended with "Ansi" prefix and all WinAPI functions was added with "A" suffix). That is done to provide you EurekaLog version for D2009 ASAP. There was no possibility of proper porting such complex tool in the short time.

The real conversion for Delphi 2009 and full support of unicode (by using UnicodeString and unicode log files) is coming in v7 version, which is under development now.

EurekaLog supports unicode log files in the sense that you could view, say, japanese and russian log files in Viewer. This is done by storing charset in log file and switching it when viewing report in Viewer.

Note, that as Viewer is usual Delphi application - it inherits all limitations of ANSI Delphi versions. I.e. it can't work in full unicode mode. It means that you could not view japanese AND russian characters at the same time in the same log.

The Viewer v4 (that is the Viewer for v7 of EurekaLog) will be compiled in D2009 and will have full unicode support. There is probability that there may be two separate versions of Viewer (i.e. one for old 9x systems and one for unicode NT).

[mlerch]

Thanks Alexander - looking forward to v7.

[mlerch]

Alex any news on v7?

Hi.

There is no full unicode support in old Delphi versions for the reason that the Delphi itself does not support unicode in RTL classes (only ANSI).
The current EurekaLog version for Delphi 2009 is still using ansi-strings (i.e. all string types was just appended with "Ansi" prefix and all WinAPI functions was added with "A" suffix). That is done to provide you EurekaLog version for D2009 ASAP. There was no possibility of proper porting such complex tool in the short time.

The real conversion for Delphi 2009 and full support of unicode (by using UnicodeString and unicode log files) is coming in v7 version, which is under development now.

EurekaLog supports unicode log files in the sense that you could view, say, japanese and russian log files in Viewer. This is done by storing charset in log file and switching it when viewing report in Viewer.

Note, that as Viewer is usual Delphi application - it inherits all limitations of ANSI Delphi versions. I.e. it can't work in full unicode mode. It means that you could not view japanese AND russian characters at the same time in the same log.

The Viewer v4 (that is the Viewer for v7 of EurekaLog) will be compiled in D2009 and will have full unicode support. There is probability that there may be two separate versions of Viewer (i.e. one for old 9x systems and one for unicode NT).