utf-7
as you might already know utf-7 is not a supported java (and hence cf) charset. it does however exist in the wild, mainly as part of bounced email systems and sometimes used in webmail like hotmail (well mainly hotmail, i've never seen it anywhere else to tell you the truth) as well as MS Exchange. folks have been complaining off and on about this for years, many mistakenly blaming macromedia for a sun java bug. votes have piled up in sun's java bugparade but alas and alack, nothing's been done about it. until now. there's a very persistent thread (its been running since feb-2004) in the cf support forums concerning this issue. a few days ago somebody (gdbezona) posted a link to an opensource utf-7 charset JCharset. if you drop that jar (jcharset.jar) into the cfinstall/runtime/jre/lib dir and stop/restart cf server ervice, cf will pick up that utf-7 charset fine. we've exercised this jar pretty thoroughly over the last two days and it has yet to blow up in our faces. it works with cfpop/cfmail/cfile and shows up in the server's available charsets via our charset CFC.if you're experiencing this issue, you might want to give this thing a whirl.
posted by Paul Hastings @ 5:00 PM
6 Comments:
I just ran into this for the first time about a month ago. I was doing just as you described, writing a mail bounce script, for automating a email list. I wrote the script on my cf7 mac install and it worked great. Deployed it to a client working with cf6 and got the error.
Upgrading them to cf7 did the trick in fixing the problem so everything ran again.
You are certainly the chartset king, but is there something in cf7 to help deal with this? I only ran into the problem while doing a cfpop action="getall"
mm "fixed" this by trapping the bad charset error and returning a message to that effect in the mail body. this fix just came out the a week or two ago.
just to be clear, by "this fix" i mean jcharset.
Funny that such a large company would leave their clients in such a mess with this issue for so long. It took me altogether about one day's work to implement JCharset's UTF-7 support - they should have done this long ago and saved everyone alot of time and frustration!
first let me say thanks for the effort, that jar works a treat & will probably save cf deevlopers quite a bit.
sun seems to drag their feet on some i18n issues. it took years for them to fix the brazilian numeric formatting. beyond that there's plenty of places where core java differs from the CLDR data too.
I wasn't referring to Sun in this case...
(They're doing a whole lot for the open community!)
Post a Comment
<< Home