DO NOT REPLY [Bug 28170] - PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1

2004-05-04 Thread bugzilla
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://issues.apache.org/bugzilla/show_bug.cgi?id=28170.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://issues.apache.org/bugzilla/show_bug.cgi?id=28170

PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1





--- Additional Comments From [EMAIL PROTECTED]  2004-05-04 10:32 ---
We serve our PDFs with a servlet which had worked for a long time with all 
Tomcat4 versions and Acrobat versions 4/5 and also Acrobat 6.0.0 but not Tomcat 
4.1.29 and Acrobat 6.0.1.

With Tomcat 4.1.30 it works again.

This report is only valid for IE of course (tested only with IE6).

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



DO NOT REPLY [Bug 28170] - PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1

2004-04-05 Thread bugzilla
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://issues.apache.org/bugzilla/show_bug.cgi?id=28170.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://issues.apache.org/bugzilla/show_bug.cgi?id=28170

PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1





--- Additional Comments From [EMAIL PROTECTED]  2004-04-05 14:31 ---
Thanks guys.

The servlet checks out:

GET /sample.pdf HTTP/1.0

HTTP/1.1 200 OK
Last-Modified: Fri, 02 Apr 2004 18:38:42 GMT
Content-disposition: inline; filename=sample.pdf
Content-Type: application/pdf
Content-Length: 32111
Date: Mon, 05 Apr 2004 14:28:36 GMT
Server: Apache-Coyote/1.1
Connection: close

I can understand all the motivations behind all of this. Ours is that JSPs are a lot 
more convenient to 
develop than servlets, but so it goes.

Peace,
Dave

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



DO NOT REPLY [Bug 28170] - PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1

2004-04-05 Thread bugzilla
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://issues.apache.org/bugzilla/show_bug.cgi?id=28170.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://issues.apache.org/bugzilla/show_bug.cgi?id=28170

PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1





--- Additional Comments From [EMAIL PROTECTED]  2004-04-05 18:30 ---
Created an attachment (id=11139)
WEB-INF directory for PDF serving servlet test

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



DO NOT REPLY [Bug 28170] - PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1

2004-04-02 Thread bugzilla
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://issues.apache.org/bugzilla/show_bug.cgi?id=28170.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://issues.apache.org/bugzilla/show_bug.cgi?id=28170

PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1





--- Additional Comments From [EMAIL PROTECTED]  2004-04-02 20:44 ---
Created an attachment (id=11107)
Test Case for Bug

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



DO NOT REPLY [Bug 28170] - PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1

2004-04-02 Thread bugzilla
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://issues.apache.org/bugzilla/show_bug.cgi?id=28170.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://issues.apache.org/bugzilla/show_bug.cgi?id=28170

PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1

[EMAIL PROTECTED] changed:

   What|Removed |Added

 Status|NEW |RESOLVED
 Resolution||INVALID



--- Additional Comments From [EMAIL PROTECTED]  2004-04-02 20:53 ---
This is actually normal. JSP is textual data, so a charset will be sent to the
client. Acrobat is broken and doesn't like that (and this is the end of the
story). You should either:
- use a servlet to generate the PDF (this seems a better match)
- hack Jasper a little bit

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



DO NOT REPLY [Bug 28170] - PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1

2004-04-02 Thread bugzilla
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://issues.apache.org/bugzilla/show_bug.cgi?id=28170.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://issues.apache.org/bugzilla/show_bug.cgi?id=28170

PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1





--- Additional Comments From [EMAIL PROTECTED]  2004-04-02 21:09 ---
I read  this supposedly fixed bug 24970: charset appended to content-type even if not 
text/*

Do you care to comment on this? This is the problem. Some comment about how you guys 
fixed the 
extra characters for images (but not application mimetypes?) I have not generated any 
text up to that 
point in my jsp and would have thought that setting the content type would cause your 
connectors to 
behave appropriately.

If I am not following the standards then please point me to the document that proves 
it. I thought that 
Tomcat was a Reference Implementation.

Also my test case works with everyone else and I know that MS IE is a bad actor with 
its 2X or 3X 
request mimetype snooping, but you guys certainly have had to deal with that for years.

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



DO NOT REPLY [Bug 28170] - PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1

2004-04-02 Thread bugzilla
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://issues.apache.org/bugzilla/show_bug.cgi?id=28170.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://issues.apache.org/bugzilla/show_bug.cgi?id=28170

PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1





--- Additional Comments From [EMAIL PROTECTED]  2004-04-02 21:13 ---
Well, it's what I said. Bug 24970 is about the connector always setting the
charset every time.
OTOH, JSPs, since they are text, always append the charset in the application
layer (here, it's the default HTTP encoding).

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



DO NOT REPLY [Bug 28170] - PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1

2004-04-02 Thread bugzilla
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://issues.apache.org/bugzilla/show_bug.cgi?id=28170.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://issues.apache.org/bugzilla/show_bug.cgi?id=28170

PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1





--- Additional Comments From [EMAIL PROTECTED]  2004-04-02 21:16 ---
I'll get back yo you if the servlet doesn't work, but I'd hate to have my developers 
change things if it 
won't.

Regards,
Dave

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



DO NOT REPLY [Bug 28170] - PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1

2004-04-02 Thread bugzilla
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://issues.apache.org/bugzilla/show_bug.cgi?id=28170.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://issues.apache.org/bugzilla/show_bug.cgi?id=28170

PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1





--- Additional Comments From [EMAIL PROTECTED]  2004-04-02 21:22 ---
Write a trivial servlet to test, without the actual PDF body generation to test
it (and use a telnet to see the headers).

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



DO NOT REPLY [Bug 28170] - PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1

2004-04-02 Thread bugzilla
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://issues.apache.org/bugzilla/show_bug.cgi?id=28170.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://issues.apache.org/bugzilla/show_bug.cgi?id=28170

PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1





--- Additional Comments From [EMAIL PROTECTED]  2004-04-02 21:37 ---
In other words, there is a bug/gap/feature in the JSP spec itself that one
really cannot return anything but text of some variety or another from a JSP page.

Essentially prior to filters you had to use a servlet for anything other than
text -- including images, PDFs (which really should not be treated as text for
various reasons), etc.  Even with filters you JSP page must pass something to
the filter which is not the end-binary response whereupon the filter produces
the binary responses -- which is not always convenient...

This is unfortunate (and something I debated with the spec authors -- especially
since no filters existed at the time), but essentially the notions of having to
either require any non-text page authors to strictly watch out for or
auto-discard whitespace produced by JSP source formatting, etc, were both
considered untenable.

At this point, I have to agree the spec authors to *some* degree, though it
would still be nice to just use the familiar JSP page mechanism for binary
output as well, e.g. by setting the page to binary prior to the output writer
being flushed and then being able to get a hold of the output *stream* instead
of writer.

Anyway, to make a long story short -- you'll have deal with the spec implications.

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



DO NOT REPLY [Bug 28170] - PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1

2004-04-02 Thread bugzilla
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://issues.apache.org/bugzilla/show_bug.cgi?id=28170.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://issues.apache.org/bugzilla/show_bug.cgi?id=28170

PDF content through JSP fails in Tomocat 4.1.30, Windows IE and Acrobat Reader 6.0.1





--- Additional Comments From [EMAIL PROTECTED]  2004-04-02 23:27 ---
Thanks, I did the telnet trick to see everything:

GET /badpdf.jsp HTTP/1.0

Returns:
HTTP/1.1 200 OK
Set-Cookie: JSESSIONID=FACB1957A5D101E32866F2466D16D929; Path=/
Content-disposition: inline; filename=sample.pdf
Content-Type: application/pdf;charset=ISO-8859-1
Content-Length: 32111
Date: Fri, 02 Apr 2004 23:08:46 GMT
Server: Apache-Coyote/1.1
Connection: close

While 
GET /sample.pdf HTTP/1.0

Returns:
HTTP/1.1 200 OK
ETag: W/32111-1080931122000
Last-Modified: Fri, 02 Apr 2004 18:38:42 GMT
Content-Type: application/pdf
Content-Length: 32111
Date: Fri, 02 Apr 2004 23:11:06 GMT
Server: Apache-Coyote/1.1
Connection: close

So it looks like servlet time.

About PDFs as Text - As someone who beta tested Acrobat 1.0 and has followed it ever 
since 
although not as closely since 4.0. Done all kinds of things like Java applets launched 
from a 3.0 
plugin, etc. - Since PDF-1.2 (or so) and linearized/zipped content streams PDFs are no 
longer 7-
bit ASCII. So, I can see where a character set choice might confuse any of the 
applications 
upstream of the web server!

Also, Jess's point about worrying about linefeeds is something I'm always checking - 
particularly 
the one(s) which easily can get tacked on after the last % - PDFs (unless linearized) 
need the end 
of the file to find everything - all the objects are found from the end.

Here we are the little guys and upstream we have Sun, MSFT and Adobe doing what they 
will.

Peace.

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]