Um, that's true for any charset that are not part of iso-8859-1. In his case, he could display those characters(even though they are not part of iso-8859-1, I think), so I guess he is lucky that most systems support those characters(mine does too:).
-----Original Message----- From: Adam Hardy [mailto:[EMAIL PROTECTED] Sent: Friday, February 20, 2004 11:06 AM To: Tomcat Users List Subject: Re: 5.0.16 and 5.0.18 international character support Well if you're not worried about it. By the way, if you are writing message resource files and serving the results as UTF-8, then you have to write the files & save them in unicode, and then use native2ascii to convert them into \uXXX for the characters to come out right. If you don't do that, you just see question marks, I think. On 02/20/2004 06:40 PM Trygve Hardersen wrote: > Checked the headers at 5.0.18, and as you said Content-Type is text/html. I > don't use the i18n taglib, and I've tested 15 times that the difference is > really between 14 and 16/18. Model is made up of Session Beans, which use > Hibernate to get the data from a MaxDB database. In other words there is no > relation between data access and my problem (which also relates to message > resources). If anyone has a 5.0.16 or 18 up and running, adding a message > resource containing a ?, ? or ? and checking the result could confirm that > this is not a single case. However, if it works with iso-8859-1, I really > don't care (: > > Thanks a bunch for the help! > milx > > -----Original Message----- > From: Adam Hardy [mailto:[EMAIL PROTECTED] > Sent: Friday, February 20, 2004 6:24 PM > To: Tomcat Users List > Subject: Re: 5.0.16 and 5.0.18 international character support > > It probably also has alot to do with the JSTL i18n taglib. Are you using > that? Or perhaps your database or JDBC driver? > > I would not assume that a minor version upgrade from 14 to 16 would > produce such a major change. There would have been a massive outcry back > then. Although 14 was still beta and 16 is production release, so > perhaps not that many people did it. > > Did you check the http headers? > > > > On 02/20/2004 06:14 PM Trygve Hardersen wrote: > >>Correct, Context-Type is text/html. So what's the conclusion? Tomcat > > 5.0.14 > >>and international characters work fine with utf-8, for later versions use >>iso-8859-1 or other? >>milx >> >> >>-----Original Message----- >>From: Adam Hardy [mailto:[EMAIL PROTECTED] >>Sent: Friday, February 20, 2004 6:00 PM >>To: Tomcat Users List >>Subject: Re: 5.0.16 and 5.0.18 international character support >> >>The point is that the browser sees the HTTP response headers, even if it >>doesn't display them, and it uses them to decide which charset to use to >>display the page, regardless of your xml & meta declarations in the html >>file. >> >>On 02/20/2004 05:57 PM Trygve Hardersen wrote: >> >> >>>I don't get the point of doing this. The problem is not request and >>>response, but the way characters are displayed. >>> >>>Changed the charset to western European, charset=iso-8859-1, and now >>>everything works fine. >>>milx >>> >>>-----Original Message----- >>>From: Antonio Fiol Bonn?n [mailto:[EMAIL PROTECTED] >>>Sent: Friday, February 20, 2004 5:33 PM >>>To: Tomcat Users List >>>Subject: Re: 5.0.16 and 5.0.18 international character support >>> >>>What Adam said was: >>>Look in the response headers to see it. >>> >>>I think he *really* meant the response headers, not the html code. >>> >>>To do that on IE, you need a plugin called ieHTTPheaders or something like >>>that. >>> >>>On Netscape/Mozilla, use LiveHTTPHeaders. >>> >>>Otherwise, if you are not on HTTPS, you can use a network sniffer. >>> >>> >>>Antonio Fiol >>> >>> >>>Trygve Hardersen wrote: >>> >>> >>> >>> >>>>Thanks for the reply. Project never ran on 4.x, developed on 5.x starting >>> >>>>from 5.0.14. Using JSTL 1.1. >>> >>> >>> >>>>5.0.14: >>>><?xml version="1.0" encoding="UTF-8"?> >>>><!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" >>>> "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd";> >>>><html xmlns="http://www.w3.org/1999/xhtml"; lang="en" xml:lang="en"> >>>><meta http-equiv="Content-Type" content="application/xhtml+xml; >>>>charset=utf-8" /> >>>> >>>>5.0.18: >>>><?xml version="1.0" encoding="UTF-8"?> >>>><!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" >>>> "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd";> >>>><html xmlns="http://www.w3.org/1999/xhtml"; lang="en" xml:lang="en"> >>>><meta http-equiv="Content-Type" content="application/xhtml+xml; >>>>charset=utf-8" /> >>>> >>>>That is equal. However, the characters ??? are in plain text in the page, >>>>but IE displays them incorrectly. I can't find any difference in the >> >>source >> >> >>>>of the pages between 5.0.14 and later, which makes me wonder if the >> >>problem >> >> >>>>is IE oriented. Thing is though, that IE displays everything correct on >>>>5.0.14..... Installing Netscape now, Opera does not support xhtml with >>>>script elements. Idea? >>>> >>>>milx >>>> >>>>-----Original Message----- >>>>From: Adam Hardy [mailto:[EMAIL PROTECTED] >>>>Sent: Friday, February 20, 2004 2:43 PM >>>>To: Tomcat Users List >>>>Subject: Re: 5.0.16 and 5.0.18 international character support >>>> >>>>On 02/20/2004 01:17 PM Trygve Hardersen wrote: >>>> >>>> >>>> >>>> >>>> >>>>>I'm having a silly problem with 5.0.16 and 5.0.18, regarding the >>>>>Scandinavian characters ?, ? and ? (probably others to). I've developed > > a > >>>>>project using tomcat and struts, where both message resources and actual >>>>>data in the database contain these characters. Using 5.0.14 and prior, >>>>> >>>>> >>>> >>>>I've >>>> >>>> >>>> >>>> >>>> >>>>>not paid attention to the characters; just used plain text for both >>>>>resources and data. This has worked out just fine, regardless of user >>>>> >>>>> >>>> >>>>locale >>>> >>>> >>>> >>>> >>>> >>>>>(and thereby the lang option of the page), the characters have been >>>>> >>>>> >>>> >>>>rendered >>>> >>>> >>>> >>>> >>>> >>>>>properly. Attempting to stay up-to-date, I upgraded to 5.0.16 and later >>>>>5.0.18, but now the characters are Chinese-like (unreadable for a >>>>>Scandinavian). Anyone knows the cause of this? >>>>> >>>>>The pages are all UTF-8, xhtml and there is no difference in handling of >>>>>message resources and model data. >>>>> >>>>> >>>> >>>>Are you sure about that? Could it be that you actually did have this >>>>problem with tomcat 5.0.x but just didn't notice? When did you upgrade >>> >>>>from tomcat 4.x? >>> >>> >>> >>>>There are issues going from tomcat 4 to 5 that could affect this, but >>>>none that I know of, just from 5.0.14 -> 16. >>>> >>>>Check the character encoding of your pages in the browser. Look in the >>>>response headers to see it. What is it & what do you want it to be? >>>> >>>>Adam >>>> >>>> >>>> >>>> >>> >>> >>> >>>--------------------------------------------------------------------- >>>To unsubscribe, e-mail: [EMAIL PROTECTED] >>>For additional commands, e-mail: [EMAIL PROTECTED] >>> >>> >> >> >> > > -- struts 1.1 + tomcat 5.0.16 + java 1.4.2 Linux 2.4.20 Debian --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
