|
From: | Brynne and Russ Jorgensen |
Subject: | Re: Several problems with 2.5.27 version on Windows ME |
Date: | Sat, 11 Jun 2005 22:47:07 -0600 |
User-agent: | Mozilla/5.0 (Windows; U; Win 9x 4.90; en-US; rv:1.6) Gecko/20040113 |
Jan Nieuwenhuizen wrote:
Brynne and Russ Jorgensen writes:I'm not sure I understand, it seems that the test is for operating system flavour, not for unicode functionalityYes, that is right.So, in that case, we'd better include something like http://nsis.sourceforge.net/archive/viewpage.php?pageid=261I thought that you said the unicode version of lilypad would workActually, you need to link the unicode version of lilypad against a special libraryOk, I'm not going to be doing that, but I do take patches against CVS module lilypond/installer/windows/lilypad. Jan.
Jan,I figured out a way to have a single executable contain both the unicode and ascii versions of lilypad and it chooses at run time which one is appropriate. Thought you might like it better than having two executables and trying to make sure the right one gets installed. Conceptually, I compile the source once for -DUNICODE and once for -UUNICODE, link it all together, and WinMain() chooses which to use. Sounds simpler than it really was - I ended up putting all of lilypad into a C++ class so that I could solve the name clash problem by using a different class name when compiling -DUNICODE and -UUNICODE. But, it does work, and I think it turned out to be relatively elegant. If you're interested in using this, let me know how to get the source to you - since I changed c-files to cpp-files, I just tarballed the whole thing which is too big to attach.
I also fixed the problem with parsing the command line so now right-click "Edit source..." on a .ly file works.
-Russ
[Prev in Thread] | Current Thread | [Next in Thread] |