They are not so different than that ones you wrote to IIS.
I made some progress since my first message and I found out that even a simgle .WML file stored as nonXML on Tamino (e.g. http://localhost/tamino/mydb/xslcoll/xsldoc/wap.wml) produces the same error when running on Nokia WAP Toolkit although it is running ok on IE 5.5.
In my zone.properties file, there is a Passthru parameter to content type as follows:
# Initialize the content type of the response document servlet.styler.initArgs=contenttype=text/xml
I found out which content type I have to set in zone.properties file in order to use Passthru and wml: text/vnd.wap.wml. If I use application/x-www-urlencoding (or something like this) it works as well.
My new problem is when I load it from a wap emulator (I’m using Nokia, WinWAP and M3Gate), WinWAP replies a “no wml content”. After the message error, there is a response given from Tamino transformed with my stylesheet in wml format. When I view the source code, the content type is changed to application/xml. If I cut and paste this response in a .wml file and load it on WinWAP, it works as I wanted to!
On jserv.log file, It seems that the process was ok, because all passthru settings are correct and there are no errors.
Someone have any idea about what is changing the content type and why is it correct when I load it from a .wml file?
Hi all, Don’t really know if it can be configured somehow but Tamino (version 2.3) does keep information about content type of non-xml objects stored. Thus when you retrieve a non-xml object it will be sent to your browser as application/octet-stream. Explorer is able to repair this in some cases as .pdf’s, .gif’s, .jpg’s … but other browsers will give you an ‘invalid or unspected content type’ error. Does any one know if this happens in newer versions too? I don’t know how the passthru servlet work but i guess it will not be able to determine the content type from the transformed content. We solved that coding our own ‘passthru’ where you have to give query, xsl file and response content type as parameters. Good luck.
OOOPS!!! Sorry for the mistake in the previous post. It should say …but Tamino (version 2.3) does NOT keep information about content type of non-xml objects stored.
Hi Fernando. That’s the idea. The browser makes a request as follows. http://myhost/myservlet?query=myquery&xsl=myxsl&mimetype=text/vnd.wap.wml The servlet gets the tamino response to the query as an xml, transforms this using the xsl, and sends back the result using the mime type choosen. I cannot give you details about how the servlet is coded as was a colleage who did most of the work. It uses some java classes from javax.xml.transform.* We fist started with this servlet when we found many troubles trying to retrieve pdf’s and power point docs stored in tamino. Good Luck