Hi
There is a 20+ years old database in a company.
The server is
Ubuntu 12.04 using
Firebird 2.5.
And all the applications are running on
Windows 10.
The database has been created without specifying any charset.
SELECT rdb$character_set_name
FROM rdb$database
; => "NONE"
There is an application written in
Lazarus 1.4.4 using the old
FBLib library from Alessandro Batisti
https://github.com/graemeg/firebird-libWhen connecting, this application specifies a charset of
ISO8859_1DB := TFBLDatabase.Create(nil);
DB.CharacterSet := 'ISO8859_1';
...
So far so good, but...
I want to migrate this application from
Lazarus 1.4.4 to
Lazarus 2.0.12.
No matter what charset I use when connecting, the result is the same : all characters with accents are replaced with question marks.
Exemple :
H?l?ne
J?r?me
Jo?l
I also tried using the TIBConnection from the Lazarus components but the result is exactly the same.
It looks like something has changed between
Lazarus 1.4 and
Lazarus 2.x regarding character encoding...?
I guess I will need to recreate this (big) database and find a way to convert the existing data from its original charset to UTF8.
Is it possible to convert the charset while restoring a backup ?
Is there another solution ?
Many thanks for your answers
Marc