>>> GDP per capita >>> ((Dollars >> - the opening round bracket, which is normally on the LHS of the text, >> must go on the RHS of the text, due to dir=RTL >> - the text then renders LTR as the bidi rules for ascii is LTR. >> - the closing round bracket depends on the bidi rules for the previous >> character (which was LTR), so it displays LTR too.
> Thanks. Am I right in assuming, then, that if I use a RTL language here > instead of the English ASCII the parentheses will be displayed correctly? Generally speaking, yes it should layout correctly. Note that some languages can be written LTR or RTL, so will depend on the 'script' being used... but that is probably too much to worry about... I only speak English, so I'm not really an expert on the issue... One technique I use is to put the english text into Google Translate, the cut'n'paste the resultant text. This works really well if you test it using a local html file, with its encoding set to utf8. If being implemented on the server side, this simplicity depend will on your server-side language.... eg: Perl supports utf8 natively, so (provided your Linux terminals support utf8), you can cut from the browser, then paste it into a string variable. Java on the other hand, is hard work and if being built using Windows tools, will probably cause more problems than it solves. Aside: I wrote some translation-related code a few years ago... this may give a bit more background on the complexities of web-app translations: http://search.cpan.org/~mathew/Locale-MakePhrase-0.5/lib/Locale/MakePhrase/OSDC2004.pod hops this helps, Mathew Robertson ******************************************************************* List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm Help: memberh...@webstandardsgroup.org *******************************************************************