cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
alecb10
Level 3

Display of localized license files in License Agreemnet panel

I have localized license files in my IA project, and everything seems good with this, except that when these are displayed in the License panel during an install, they don't display correctly.

The files I think are UTF-16 text files, and when English, French, German etc license files are displayed, every character is double-spaced, plus there are some odd characters displayed here and there.

None of the text editors I have show any problem displaying the file contents.

Am I missing some config step in the designer UI to tell it these aren't ascii files? I've searched but I can't find anything.

Anyone know what might be wrong or how to resolve it?
Labels (1)
0 Kudos
(3) Replies
pv7721
Level 20

What type of files are your EULAs? I use HTML files for GUI mode installation and text files (with ASCII characters only) for text mode installation.
0 Kudos
alecb10
Level 3

The license files are utf-16 encoded, I think.

In hex, the beginning of the file looks like this:
FFFE49006E007400650072

In a text editor (I use Textpad but Notepad behaves the same), the beginning FFFE is ignored, as are the null bytes, so you get Inter (etc.)

In the IA license panel, you get ÿþI n t e r (etc.)

So it seems to expect ascii and doesn't understand the double byte.

Also, in what may be a related issue, when I run the installer, for language choices, I get only English and other western European languages for choices, but I have other locales selected in Project - Locales. For example, I have en, fr, de, ja, tr and zh_CN checked, and locale files are generated for them. But when I run the installer, only English, Deutsch and Francais are presented as choices.

For any of these 3 choices, the manner of display of the license file is the same, ÿþ(letter)(space)(letter) etc.

Any thoughts or clarification would be much appreciated!
0 Kudos
alecb10
Level 3

Hi, a little more research and I think I've found the answer to my question.

Found an article in the IA knowledge base that says that IA requires the use of escaped Unicode documents and strings for display of non-Western characters, and suggests the use of Java's native2ascii binary to make the converson.

Also the FFFE at the head of the file indicates the file is UTF16LE. When that's passed as a parameter to native2ascii, the Western European language license files convert and are displayed properly, and I expect that the non-Western European license files will display correctly also. 🙂
0 Kudos