File does not begin with '%PDF-'
I have a servlet that returns a PDF over SSL (using Struts and fop).
HttpServletResponse Res;
Res.setContentType("application/pdf");
Res.setHeader("Content-disposition","inline;filename=\""+MyFile+".pdf\"");
Here I get the OutputStream from Res (Res.getOutputStream() ) and stream out the PDF
I was getting IE security warnings and while going through various sites and from this link "http://support.microsoft.com/default.aspx?scid=kb;en-us;321532", I used Res.setHeader("Accept-Ranges", "bytes") in my code.
But now, some screens work perfectly fine without the warnings while for some screens I get the error "File does not begin with %PDF".
Anyone faced this ?
Thanks.
# 1
According to the HTTP RFC, the filename should not have quotes around it. Just use
Res.setContentType("application/pdf");
Res.setHeader("Content-disposition","inline;filename="+MyFile+".pdf");
Res.setHeader("Accept-Ranges", "bytes");
I doubt that this will fix your problem, but it is worth a try at least.
# 2
Will give it a shot now .. thx for suggesting ...
# 3
Unfortunately, this did not fix it :(
# 4
Maybe this is a radical suggestion, but have you considered accepting the message at face value and finding out why the data you are sending doesn't start with those characters?
# 5
Hmm well let me tell that this occurs only with IE and not with Firefox but unfortunately our users use IE.
Also, if I remove the Res.setHeader("Accept-Ranges", "bytes");
then I do not get this error and only the IE security warning appears. Accepting that loads the PDF correctly.
I read it somewhere that IE makes multiple GETs to the server while requesting for the PDF. Although, it would be requesting for the content-type which I am specifically setting to my response, not sure where things are going wrong.
# 6
One more addition to this ..If I upgrade my acrobat to the latest, then everything seems alright but again that is not an option here since we can't ask users to upgrade the acrobat.
# 7
> Hmm well let me tell that this occurs only with IE
> and not with Firefox but unfortunately our users use IE.
Ah, I see. There's no accounting for taste, is there?
> Also, if I remove the Res.setHeader("Accept-Ranges",
> "bytes");
> then I do not get this error and only the IE security
> warning appears. Accepting that loads the PDF
> correctly.
That header is making a statement to the client that your web server has the ability to... um, I don't know exactly what. Maybe you could look it up. But at any rate it appears that your web server doesn't actually have that ability, hence the problem. (My guess.)
> I read it somewhere that IE makes multiple GETs to
> the server while requesting for the PDF. Although, it
> would be requesting for the content-type which I am
> specifically setting to my response, not sure where
> things are going wrong.
Right. Or else it's Acrobat that is requesting a particular range of bytes (if that's what the header actually means) and your web server can't support that request properly. And maybe the new version of Acrobat has a better way of doing things that doesn't require this. (More guessing.)
Edit: And maybe that Microsoft support page assumed your web server was IIS and it does support the Accept-Ranges header, but yours isn't.
Message was edited by:
DrClap
# 8
> > Hmm well let me tell that this occurs only with IE
> > and not with Firefox but unfortunately our users
> use IE.
>
> Ah, I see. There's no accounting for taste, is
> there?
>
> > Also, if I remove the
> Res.setHeader("Accept-Ranges",
> > "bytes");
> > then I do not get this error and only the IE
> security
> > warning appears. Accepting that loads the PDF
> > correctly.
>
> That header is making a statement to the client that
> your web server has the ability to... um, I don't
> know exactly what. Maybe you could look it up. But at
> any rate it appears that your web server doesn't
> actually have that ability, hence the problem. (My
> guess.)
>
> > I read it somewhere that IE makes multiple GETs to
> > the server while requesting for the PDF. Although,
> it
> > would be requesting for the content-type which I
> am
> > specifically setting to my response, not sure
> where
> > things are going wrong.
>
> Right. Or else it's Acrobat that is requesting a
> particular range of bytes (if that's what the header
> actually means) and your web server can't support
> that request properly. And maybe the new version of
> Acrobat has a better way of doing things that doesn't
> require this. (More guessing.)
>
I guess you are right here. The web server doesn't have the ability to understand the "Accept-Ranges = bytes" and moreover as you said the MS site would be self assuming everyone has IIS only.
> Edit: And maybe that Microsoft support page assumed
> your web server was IIS and it does support the
> Accept-Ranges header, but yours isn't.
>
> Message was edited by:
> DrClap
Message was edited by:
Deal_NoDeal
# 9
Well a correction here ... I think my web server is able to understand the "Accept-Ranges" since some pages begin to work just fine but this problem occurs only with few other pages.
# 10
Anyone got this working ?
