Dealing with accent marks

[prince_mallow]

I have not been updated on how to deal with accent marks... Most of the sites I work with are in french and I've been manually changing all the é to their anscii equivalent with the ampersign and number.. Do I really still have to change these manually or is there a magic phrase I can put into html so that I don't have to worry about the all the different accent characters?

[eruna]

Can you use find and replace? Dreamweaver is pretty good at this. It will do it site wide. Other applications can also do it.

[mikemckinney]

You may find one of these articles to be of interest:

http://www.sitepoint.com/article/gui...acter-encoding
http://www.sitepoint.com/article/typ...unctuation-web

mikem

[AutisticCuckoo]

It all depends on which character encoding you use. UTF-8 or ISO 8859-1 should work well for French.

As long as you use an encoding that allows a literal representation of those accented characters, there's no need whatsoever to replace them with entity references or numeric character references.

[C. Ankerstjerne]

In addition to what Tommy wrote, I personally prefer the named entity references, rather than the numbered ones, i.e. é rather than é and È rather than È (mainly because they're easier to remember).

[Stomme poes]

While I wouldn't change content that was being sent to me, when I am building pages, even with UTF-8 in the document, in the meta tag and on the server, someone somewhere is going to have some goofy machine that ignores all that for some reason and gets ??? instead. So, at least for things like headers, menus, footers, if not the content, I still use the ascii (the hex actually).

I also use them in place of the "character entities" since, when I started with my fake XHTML, I was taught that XML doesn't allow most of those, so &copy became & #169; and so-on.

But that's me, it's a bit of extra work, and I likely wouldn't have to do it as even the WIndows users here have some charset that understands é as e with grave accent.

Fortunately, as I understand it, UTF-8 is supposed to be the default for a UA if it doesn't know which charset a site is using?

[C. Ankerstjerne]

You can't count on the default, unfortunately. For some reason, my machine defaults to japanese, when encountering unescaped UTF-8 with the wrong content-header.

[AutisticCuckoo]

In addition to what Tommy wrote, I personally prefer the named entity references, rather than the numbered ones

That's fine, as long as you don't use XHTML and you're willing to take the (negligible) risk of problems in really old browsers.

Fortunately, as I understand it, UTF-8 is supposed to be the default for a UA if it doesn't know which charset a site is using?

I think most browsers use Windows-1252 (or possibly ISO 8859-1) as the default encoding, since that's usually the encoding in point-and-click publishing tools used by non-savvy authors who don't set up their servers properly.

[prince_mallow]

Wow thanks for all the info... So I guess it's really split down the middle; I think I may try the UTF code for a while and see how that reacts on different systems..