I've just found this weird bug with TeXnicCenter (TXC). I have two versions installed :
TXC2.0 beta 1 (64 bit) (referred to below as 2b1)
TXC1.0 RC1 (referred to as 1RC)
My PC is running Win 7 pro in French.
EDIT : a quick summary, before the long description. When I create a project in 2b1, the accents aren't displayed properly. If I open the project in 1RC, within TXC, the accents are also displayed incorrectly. Adding new text with no accents in 1RC changes nothing, but adding text with accents in 1RC causes all new text added in either version to be displayed correctly. Finally, I did similar tests creating just a tex file in 2b1, and a project in 1RC, and I observed no problem.
I've done my best to describe step by step what happens, but it's a bit long, so please bear with me.
In 2b1, I create a new project, using the general/empty project template. I then type the following code :
When compiling, I get the same output as in step 6. At this stage, if I open the .tex in notepad, it now looks like the code pasted just above.
From this point onwards, I can add text, with or without accents, in either version, and all the new text behaves normally (ie the accents are displayed properly).
Now, if I create just a .tex file (not a project) in 2b1, with the following code :
Finally, if I create a project in 1RC, and open it in 2b1, there is no problem displaying accents.
I'm sorry this post is so long. Can somebody help me out here. This took me a long time to work out: I started out on 2b1, and had no idea why the accents weren't being displayed properly! Also, I'm about to start writing my PhD thesis in French, and I really can't afford to suddenly have all my accents replaced by weird characters !
Forget TXC1 RC1 completely. It doesn't manage Unicode. For European languages like French you can use a semi-automatic determination for the input encoding by the selinput package from the oberdiek bundle. The following lines replace inputenc.
scalpas wrote:[…] Thorsten, re your comment about RC1, I've worked with it for years and never had any trouble with accents. […]
It is simply outdated and can't even be downloaded any more. Furthermore it has several deficiencies which cause problems. And these do not only concern the encoding.
scalpas wrote:[…] And another question: is there any particular reason to choose either ANSI or Unicode?
Unicode provides all characters you can think of in different languages, especially Asian languages, Arabic, Hebrew and much more. ANSI is based on ASCII, thus very limited. ANSI represents the past, Unicode is the present and the future. The upcoming typesetting engines XeTeX and LuaTeX have native Unicode support.