Community
Participate
Working Groups
Build ID: M20071023-1652 Steps To Reproduce: Strictly saying this isn't a bug. More information: Currently Eclipse in Windows defaults to windows 1252 encoding, which has characters that are incompatible with UTF-8 - for instance the open/close quotes. On the other hand in all *nix environments it defaults to UTF-8. So files created in Eclipse+Windows are not necessarily cross-platform. One problem we keep running into in our development environment is that developers tend to copy requirement text from MS Word or IE into Eclipse. The text sometimes contains open/close quotes, which cause trouble later when some other tools down the chain assume UTF-8 encoding. We had to request all the developers explicitly configure their eclipse to use UTF-8 encoding. This can clearly be categorized as a config/training issue, but I thought it would be nice if Eclipse could always default to UTF-8, regardless of which platform it runs on.
We won't do this. For a longer discussion see bug 108668. *** This bug has been marked as a duplicate of bug 108668 ***
In case it's not obvious in bug 108668 the key to handling this correctly and smoothly is not so much "training" your developers to set their workbench to all ways use UTF-8 (if that's what you want), but instead for one smart person on your team to set it for each project you share. That way, one person has to do the main work, others "get it for free" when they check projects in and out of source control. And, I would caution against a blind default of UTF-8 ... it's good for most things, but since CVS, to name one example, isn't all that smart about encoding, it can occasionally mess up some things since it pretty much assumes one byte per character. I always advise ISO-8859-1 as the default, since it is a single-byte characterset, it's hard to ever get "messed up". (UTF-8 is variable byte ... from 1 to 3). Good luck.
