tomcat-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Adam Hardy <ahardy.str...@cyberspaceroad.com>
Subject Re: 5.0.16 and 5.0.18 international character support
Date Fri, 20 Feb 2004 18:06:15 GMT
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:ahardy.struts@cyberspaceroad.com] 
> 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:ahardy.struts@cyberspaceroad.com] 
>>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:fiol.bonnin@terra.es] 
>>>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:ahardy.struts@cyberspaceroad.com] 
>>>>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: tomcat-user-unsubscribe@jakarta.apache.org
>>>For additional commands, e-mail: tomcat-user-help@jakarta.apache.org
>>>
>>>
>>
>>
>>
> 
> 


-- 
struts 1.1 + tomcat 5.0.16 + java 1.4.2
Linux 2.4.20 Debian


---------------------------------------------------------------------
To unsubscribe, e-mail: tomcat-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tomcat-user-help@jakarta.apache.org


Mime
View raw message