MakeNSIS can now generate Unicode or Ansi installers based on a script attribute. SCons generates both Ansi and Unicode stubs and plugins.

The official plugins are now stored in architecture specific subdirectories under NSIS\Plugins. !AddPluginDir also gained a new (optional) architecture flag because MakeNSIS now stores separate plugin information for each target architecture. Storing plugins in the root of the Plugins directory is no longer supported.

MinGW does not implement the unicode CRT startup functions so the entry point functions and linker parameters had to be changed. The unicode tools use the ansi entry point and a small helper function that calls into the real code: _tmain has full argc+argv emulation while wWinMain does not pass the command line parameters. The stubs do not use any CRT functions and have no CRT or unicode helper code, they call our entry point directly.



git-svn-id: https://svn.code.sf.net/p/nsis/code/NSIS/trunk@6269 212acab6-be3b-0410-9dea-997c60f758d6
This commit is contained in:
anders_k 2012-10-13 01:47:50 +00:00
parent 8f330bbbdf
commit 7cc150c464
73 changed files with 936 additions and 713 deletions

View file

@ -152,8 +152,8 @@ The NSIS script format and the format used for interface dialogs are easy, docum
\H{intro-unicode} Unicode installers
Starting with MakeNSIS v?.??, you can choose to create Unicode installers by using \R{atargetminimalos}{TargetMinimalOS} with a value greater than or equal to 5.0.
These installers will work only under Windows 2000 or more recent (depending on the value you chose), but they will allow you to display your installer in any Unicode language supported by the OS.
Starting with MakeNSIS v3.0 you can choose to create Unicode installers by setting the \R{aunicodetarget}{Unicode} attribute.
These installers will not work on Windows 95/98/ME but they will allow you to display your installer in any Unicode language supported by the OS.
When building a Unicode installer, NSIS variables can hold Unicode characters (0000-FFFF). There should be no need to modify your existing scripts.
If you want to read/write Unicode files, specific instructions have been added to read/write UTF-16LE strings from/to disk.