Umlauts in Select Components ListBox

Archive: Umlauts in Select Components ListBox

F. Heidenreich
14th November 2001 06:44 UTC

Umlauts in Select Components ListBox
Hi there,

I have a small problem with NSIS 1.65: when I use umlauts to describe a section in my nsi-script, I get a y with two points above, in front of every umlaut in the section-string that appears in the select components listbox. Has anyone an idea how to fix this. I've looked to the code but I didn't found the reason for this problem.

Thanks for any suggestions
~ Flo

F. Heidenreich
14th November 2001 16:32 UTC

Umlauts in Select Components ListBox - solution?
Hi there again,

I spent some time on the debugger an I noticed that
CEXEBuild::preprocess_string() add these strange characters in front of the umlauts.
The value of i is 252 on the umlaut according to the character u and thats why the lines:


if (i >= VAR_CODES_START)
  *out++ = (char)255;

add the y with the two points above.
The two lines mentioned above are the 'Test for characters extending into the variable codes', but I think (but I'm not sure!) this test should be only performed on in-strings with the length of 1.
So I hacked


if (strlen(in) == 1 && i >= VAR_CODES_START)
      *out++ = (char)255;

and it works fine for me :)

Did I overlook something?

Thanks for any suggestions

~ Flo

justin
14th November 2001 19:19 UTC

Actually this is a bug I introduced. The routine add_string() in build.cpp was not passing the 'process' parameter to add_string_main(). Which was causing this string to be processed, when it should not have been. I will put the fix in 1.66.

-Justin

F. Heidenreich
15th November 2001 11:50 UTC

Hi Justin,

thank you for the upcoming fix (I've already suspected, that my fix isn't on the right place :) ) and thank you very much for NSIS!

Regards,
~ Flo