Starting with V0.9.18, WarpIN handles all character data internally as Unicode. If you do not know what Unicode is about, www.unicode.org will give you all the information you will ever need and more.

However, WarpIN's Unicode support is totally transparent to the user:

  1. Since the OS/2 file-system API's do not support Unicode, WarpIN's file and compression engine will always operate with the default codepage that the WarpIN process was started with. This is normally the first codepage specified with the CODEPAGE setting in CONFIG.SYS, which is 850 on most systems. This ensures that all file names etc. will always be processed with the user's default codepage.

    However, to make sure that WarpIN's file operations will keep working even if the user chooses to change his default code page (between package installation and de-installation), it is strongly recommended that package creators never use non-ASCII characters (above ASCII 127) in the packages' file names. Otherwise file verification and de-installation of WarpIN might fail if the user chooses to alter the system code page at one point. (But then this is a general problem with OS/2 that WarpIN cannot easily fix.)

  2. The WarpIN database now stores all data in Unicode (encoded as UTF-8). For compatibility with packages that were stored with older versions, WarpIN will transparently translate older data from the codepage that it was stored with to Unicode.

  3. The frontend (i.e. the windows that you see in either "Install" or "Database" mode) will always use the system codepage. All Unicode data from the install scripts and the database will be converted to display properly. This might however result in some data not being displayed correctly if some Unicode code points cannot be displayed on your system, e.g. if you are trying to install a package that was created in Asian or Russian languages.