I use a *.resx file for the localization purpose. The Name - it's a phrase or word in English. The Value - the translation to another language of that phrase. I choose this approach to have a one localization file for the whole application. And anyone who have this file can make translation by themselves.
But in the Visual Studio 2010 resx editor, each record with name which have spaces in it, 开发者_运维百科have a warning: "The resource name is not a valid identifier." Though it compiles and works, but please tell me if I am doing something wrong here.
First of all, the idea of the resx files is to have a separate resource file for each culture. You can provide the new translation by creating a new file with different values for the same keys.
For example, you can create Forms.en-GB.resx, Forms.pl-PL.resx, Forms.de-DE.resx and the appropriate file will be picked up based on the current UI culture without you having to do anything (except ensuring relevant culture is set).
Visual Studio will generate a resource class that contains all your key/value pairs from resource file as properties - that makes it easier to use in code. The warning you get means that the keys you've provided in resource file are not a valid identifiers (they contain spaces). You might want to use _
instead of space in the keys.
If you don't want to use the generated class you can ignore this warning - your resx files are fine and can be used directly. You can remove ResXFileCodeGenerator
from 'Custom Tool' property of your resx file (properties windows) or set 'Access Modifier' to 'No code generation' in resx file editor if you do not need to generate a class, but you will still get the warning.
The strength of localization with resx files is that the culture on your computer decides what language your application should be in. If you keep to one resx file, according to me, you ignore it's power. Instead, try making a resx file for each language you want to integrate. for example: the default language is english, then you have a default page localization.resx where you only keep english sentences. Say you need a French translation, make another resx file called localization.fr-FR.resx. So users who have the fr-FR culture enabled on there computer will have that language on the program without any code specific work. If someone with a culture not integrated in your application starts the program, it will look for it, and if it doesn't find it, it chooses the default, ie english, one.
So to my opinion, don't use 1 resx file for different languages, but use the powers given in the framework.
I suppose it works, but it's not really the strategy you are supposed to follow.
Take a look here; the basic idea is that you take advantage of the controls in .NET to automatically get the correct localisation themselves, so you kind of don't need to worry about doing the translating.
Though, I don't use this all the time, and I do somewhat do as you do, but I tend to use an identifier, so I may have:
UserWelcome Hey, {name}, thanks for dropping by ...
And then I'll translate that. It's helpful because it allows generality in the languages (say, for example, some languages should be greeted formally, and others not, you don't want to be contrained by a direct translation of, "You", say).
Hope this is clear.
If what you've got works, then I suppose that's something, but it's not the "general" way of doing it.
精彩评论