You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@pdfbox.apache.org by Walter Kehl <wa...@outlook.com> on 2014/12/12 11:53:02 UTC

Downloadind a pdf file doesn't work

Hi all,

 

I have the following situation:

 

I am loading with PdfBox files from the internet with the call

PDDocument document = PDDocument.load( inputStream );

 

So far it has worked nicely, but I have problems with this file :
http://esa.un.org/unpd/wup/PressRelease/WUP2014_PressRelease.pdf

 

After I load it, it is empty, and the call document.getNumberOfPages()
returns 0.

However when I download the file manually and then load it into PdfBox, then
everything is fine.

 

Any idea what could be happening? I am currently using PdfBox 1.8.5.

 

Thanks and Best Regards

Walter

 

 

 


Re: Downloadind a pdf file doesn't work

Posted by Jack Bush <ne...@yahoo.com.au>.
Unsubscribe 

     On Saturday, 20 December 2014, 5:21, Walter Kehl <wa...@outlook.com> wrote:
   

 Hi Tilman,

thanks for your response. When I use your code (I am now also at version
1.8.8) I get the correct result. I think the problem has to do with the
stream object which gets used.

I am using (for various reasons) the http client from the Apache
HttpComponents library. This client creates an input stream of type
EofSensorInputStream whereas the URL.openStream method from your code
returns an HttpInputStream object. Maybe the EofSensorInputStream closes the
stream too early, because the following scenario happens:

- I download the pdf file with my Apache client. The resulting file can be
opened without a problem with a PDF viewer. Also a file compare doesn't show
any difference to a manually downloaded version.
- But when I then read in this file into Pdf box, PDFBox cannot read it
either and returns this warning: "End-of-File, expected line". 

But things are getting are involved here and I am wondering whether it is
worth spending more time on this issue...


Best regards
Walter




-----Original Message-----
From: Tilman Hausherr [mailto:THausherr@t-online.de] 
Sent: Samstag, 13. Dezember 2014 20:26
To: users@pdfbox.apache.org
Subject: Re: Downloadind a pdf file doesn't work

Oops, meant to say 1.8.8 is the current one.

Tilman

Am 13.12.2014 um 19:42 schrieb Tilman Hausherr:
> Hi,
>
> So you're using a "special" http client...
>
> Anyway, here's what I just did with the 1.8.9 version:
>
>        URL url = new
> URL("http://esa.un.org/unpd/wup/PressRelease/WUP2014_PressRelease.pdf");
>        InputStream is = url.openStream();
>        PDDocument doc = PDDocument.load(is);
>        System.out.println("pages: " + doc.getNumberOfPages());
>
> All output I get is
>
>    pages: 2
>
> Btw the two "errors" you mention are warnings about malformed PDFs. 
> However there's really a length 66346 in your file and I don't get 
> that warning. This means that somehow you're not getting the exact 
> file. Maybe save what you're downloading with your "http client" and 
> compare it with that you download with a browser. Or try what I did 
> and see if it works.
>
> What version are you using? 1.8.9 is the current one.
>
> Tilman
>
> Am 13.12.2014 um 18:41 schrieb Walter Kehl:
>> Hi John, Tilman,
>>
>> thanks for the reply. Here is some additional information:
>>
>> - the http client I am using to get the input stream already has a 
>> user agent set. Also I have downloaded with PDF box already lots of 
>> PDF files where there never was a problem.
>> - when I try to load the document remotely from the URL, I get the 
>> following error messages:
>>    18:34:32 WARN  BaseParser          :: Specified stream length 
>> 66346 is
>> wrong. Fall back to reading stream until 'endstream'.
>>    18:34:35 WARN  XrefTrailerResolver  :: Did not found XRef object 
>> at specified startxref position 0
>> - I have written the input stream directly to a file and it was a 
>> valid PDF.
>> It could load it both with an external tool and with PDFBox.
>>
>> Yes, of course I could always download a file first to a temp file 
>> and then load it into PDFBox. But I think the direct way is more 
>> elegant and faster.
>> I have also debugged a little bit into the code and to me it doesn't 
>> look like PDFBox uses a temporary file, but rather reads directly 
>> from the input stream.... but I might be wrong.
>>
>> Anyway, thanks for providing such a good free software!
>>
>> Best
>> Walter
>>
>> -----Original Message-----
>> From: John Hewson [mailto:john@jahewson.com]
>> Sent: Freitag, 12. Dezember 2014 18:57
>> To: users@pdfbox.apache.org
>> Subject: Re: Downloadind a pdf file doesn't work
>>
>> Good point Tilman. Walter, try saving writing the InputStream to a 
>> File and check that it's a valid PDF.
>>
>> -- John
>>
>>> On 12 Dec 2014, at 09:50, Tilman Hausherr <TH...@t-online.de>
>>> wrote:
>>>
>>> This sounds more like a http problem. Try setting a user agent like 
>>> a
>> browser.
>>> https://stackoverflow.com/questions/2529682/setting-user-agent-of-a-
>>> ja
>>> va-urlconnection
>>>
>>> Tilman
>>>
>>> Am 12.12.2014 um 11:53 schrieb Walter Kehl:
>>>> Hi all,
>>>>
>>>>  I have the following situation:
>>>>
>>>>  I am loading with PdfBox files from the internet with the call
>>>>
>>>> PDDocument document = PDDocument.load( inputStream );
>>>>
>>>>  So far it has worked nicely, but I have problems with this file :
>>>> http://esa.un.org/unpd/wup/PressRelease/WUP2014_PressRelease.pdf
>>>>
>>>>  After I load it, it is empty, and the call
>>>> document.getNumberOfPages() returns 0.
>>>>
>>>> However when I download the file manually and then load it into 
>>>> PdfBox, then everything is fine.
>>>>
>>>>  Any idea what could be happening? I am currently using PdfBox 1.8.5.
>>>>
>>>>  Thanks and Best Regards
>>>>
>>>> Walter
>>>>
>>>>
>



   

RE: Downloadind a pdf file doesn't work

Posted by Walter Kehl <wa...@outlook.com>.
Hi Tilman,

thanks for your response. When I use your code (I am now also at version
1.8.8) I get the correct result. I think the problem has to do with the
stream object which gets used.

I am using (for various reasons) the http client from the Apache
HttpComponents library. This client creates an input stream of type
EofSensorInputStream whereas the URL.openStream method from your code
returns an HttpInputStream object. Maybe the EofSensorInputStream closes the
stream too early, because the following scenario happens:

- I download the pdf file with my Apache client. The resulting file can be
opened without a problem with a PDF viewer. Also a file compare doesn't show
any difference to a manually downloaded version.
- But when I then read in this file into Pdf box, PDFBox cannot read it
either and returns this warning: "End-of-File, expected line". 

But things are getting are involved here and I am wondering whether it is
worth spending more time on this issue...


Best regards
Walter




-----Original Message-----
From: Tilman Hausherr [mailto:THausherr@t-online.de] 
Sent: Samstag, 13. Dezember 2014 20:26
To: users@pdfbox.apache.org
Subject: Re: Downloadind a pdf file doesn't work

Oops, meant to say 1.8.8 is the current one.

Tilman

Am 13.12.2014 um 19:42 schrieb Tilman Hausherr:
> Hi,
>
> So you're using a "special" http client...
>
> Anyway, here's what I just did with the 1.8.9 version:
>
>         URL url = new
> URL("http://esa.un.org/unpd/wup/PressRelease/WUP2014_PressRelease.pdf");
>         InputStream is = url.openStream();
>         PDDocument doc = PDDocument.load(is);
>         System.out.println("pages: " + doc.getNumberOfPages());
>
> All output I get is
>
>     pages: 2
>
> Btw the two "errors" you mention are warnings about malformed PDFs. 
> However there's really a length 66346 in your file and I don't get 
> that warning. This means that somehow you're not getting the exact 
> file. Maybe save what you're downloading with your "http client" and 
> compare it with that you download with a browser. Or try what I did 
> and see if it works.
>
> What version are you using? 1.8.9 is the current one.
>
> Tilman
>
> Am 13.12.2014 um 18:41 schrieb Walter Kehl:
>> Hi John, Tilman,
>>
>> thanks for the reply. Here is some additional information:
>>
>> - the http client I am using to get the input stream already has a 
>> user agent set. Also I have downloaded with PDF box already lots of 
>> PDF files where there never was a problem.
>> - when I try to load the document remotely from the URL, I get the 
>> following error messages:
>>    18:34:32 WARN  BaseParser           :: Specified stream length 
>> 66346 is
>> wrong. Fall back to reading stream until 'endstream'.
>>    18:34:35 WARN  XrefTrailerResolver  :: Did not found XRef object 
>> at specified startxref position 0
>> - I have written the input stream directly to a file and it was a 
>> valid PDF.
>> It could load it both with an external tool and with PDFBox.
>>
>> Yes, of course I could always download a file first to a temp file 
>> and then load it into PDFBox. But I think the direct way is more 
>> elegant and faster.
>> I have also debugged a little bit into the code and to me it doesn't 
>> look like PDFBox uses a temporary file, but rather reads directly 
>> from the input stream.... but I might be wrong.
>>
>> Anyway, thanks for providing such a good free software!
>>
>> Best
>> Walter
>>
>> -----Original Message-----
>> From: John Hewson [mailto:john@jahewson.com]
>> Sent: Freitag, 12. Dezember 2014 18:57
>> To: users@pdfbox.apache.org
>> Subject: Re: Downloadind a pdf file doesn't work
>>
>> Good point Tilman. Walter, try saving writing the InputStream to a 
>> File and check that it's a valid PDF.
>>
>> -- John
>>
>>> On 12 Dec 2014, at 09:50, Tilman Hausherr <TH...@t-online.de>
>>> wrote:
>>>
>>> This sounds more like a http problem. Try setting a user agent like 
>>> a
>> browser.
>>> https://stackoverflow.com/questions/2529682/setting-user-agent-of-a-
>>> ja
>>> va-urlconnection
>>>
>>> Tilman
>>>
>>> Am 12.12.2014 um 11:53 schrieb Walter Kehl:
>>>> Hi all,
>>>>
>>>>   I have the following situation:
>>>>
>>>>   I am loading with PdfBox files from the internet with the call
>>>>
>>>> PDDocument document = PDDocument.load( inputStream );
>>>>
>>>>   So far it has worked nicely, but I have problems with this file :
>>>> http://esa.un.org/unpd/wup/PressRelease/WUP2014_PressRelease.pdf
>>>>
>>>>   After I load it, it is empty, and the call
>>>> document.getNumberOfPages() returns 0.
>>>>
>>>> However when I download the file manually and then load it into 
>>>> PdfBox, then everything is fine.
>>>>
>>>>   Any idea what could be happening? I am currently using PdfBox 1.8.5.
>>>>
>>>>   Thanks and Best Regards
>>>>
>>>> Walter
>>>>
>>>>
>


Re: Downloadind a pdf file doesn't work

Posted by Tilman Hausherr <TH...@t-online.de>.
Oops, meant to say 1.8.8 is the current one.

Tilman

Am 13.12.2014 um 19:42 schrieb Tilman Hausherr:
> Hi,
>
> So you're using a "special" http client...
>
> Anyway, here's what I just did with the 1.8.9 version:
>
>         URL url = new 
> URL("http://esa.un.org/unpd/wup/PressRelease/WUP2014_PressRelease.pdf");
>         InputStream is = url.openStream();
>         PDDocument doc = PDDocument.load(is);
>         System.out.println("pages: " + doc.getNumberOfPages());
>
> All output I get is
>
>     pages: 2
>
> Btw the two "errors" you mention are warnings about malformed PDFs. 
> However there's really a length 66346 in your file and I don't get 
> that warning. This means that somehow you're not getting the exact 
> file. Maybe save what you're downloading with your "http client" and 
> compare it with that you download with a browser. Or try what I did 
> and see if it works.
>
> What version are you using? 1.8.9 is the current one.
>
> Tilman
>
> Am 13.12.2014 um 18:41 schrieb Walter Kehl:
>> Hi John, Tilman,
>>
>> thanks for the reply. Here is some additional information:
>>
>> - the http client I am using to get the input stream already has a user
>> agent set. Also I have downloaded with PDF box already lots of PDF files
>> where there never was a problem.
>> - when I try to load the document remotely from the URL, I get the 
>> following
>> error messages:
>>    18:34:32 WARN  BaseParser           :: Specified stream length 
>> 66346 is
>> wrong. Fall back to reading stream until 'endstream'.
>>    18:34:35 WARN  XrefTrailerResolver  :: Did not found XRef object at
>> specified startxref position 0
>> - I have written the input stream directly to a file and it was a 
>> valid PDF.
>> It could load it both with an external tool and with PDFBox.
>>
>> Yes, of course I could always download a file first to a temp file 
>> and then
>> load it into PDFBox. But I think the direct way is more elegant and 
>> faster.
>> I have also debugged a little bit into the code and to me it doesn't 
>> look
>> like PDFBox uses a temporary file, but rather reads directly from the 
>> input
>> stream.... but I might be wrong.
>>
>> Anyway, thanks for providing such a good free software!
>>
>> Best
>> Walter
>>
>> -----Original Message-----
>> From: John Hewson [mailto:john@jahewson.com]
>> Sent: Freitag, 12. Dezember 2014 18:57
>> To: users@pdfbox.apache.org
>> Subject: Re: Downloadind a pdf file doesn't work
>>
>> Good point Tilman. Walter, try saving writing the InputStream to a 
>> File and
>> check that it's a valid PDF.
>>
>> -- John
>>
>>> On 12 Dec 2014, at 09:50, Tilman Hausherr <TH...@t-online.de> 
>>> wrote:
>>>
>>> This sounds more like a http problem. Try setting a user agent like a
>> browser.
>>> https://stackoverflow.com/questions/2529682/setting-user-agent-of-a-ja
>>> va-urlconnection
>>>
>>> Tilman
>>>
>>> Am 12.12.2014 um 11:53 schrieb Walter Kehl:
>>>> Hi all,
>>>>
>>>>   I have the following situation:
>>>>
>>>>   I am loading with PdfBox files from the internet with the call
>>>>
>>>> PDDocument document = PDDocument.load( inputStream );
>>>>
>>>>   So far it has worked nicely, but I have problems with this file :
>>>> http://esa.un.org/unpd/wup/PressRelease/WUP2014_PressRelease.pdf
>>>>
>>>>   After I load it, it is empty, and the call
>>>> document.getNumberOfPages() returns 0.
>>>>
>>>> However when I download the file manually and then load it into
>>>> PdfBox, then everything is fine.
>>>>
>>>>   Any idea what could be happening? I am currently using PdfBox 1.8.5.
>>>>
>>>>   Thanks and Best Regards
>>>>
>>>> Walter
>>>>
>>>>
>


Re: Downloadind a pdf file doesn't work

Posted by Tilman Hausherr <TH...@t-online.de>.
Hi,

So you're using a "special" http client...

Anyway, here's what I just did with the 1.8.9 version:

         URL url = new 
URL("http://esa.un.org/unpd/wup/PressRelease/WUP2014_PressRelease.pdf");
         InputStream is = url.openStream();
         PDDocument doc = PDDocument.load(is);
         System.out.println("pages: " + doc.getNumberOfPages());

All output I get is

     pages: 2

Btw the two "errors" you mention are warnings about malformed PDFs. 
However there's really a length 66346 in your file and I don't get that 
warning. This means that somehow you're not getting the exact file. 
Maybe save what you're downloading with your "http client" and compare 
it with that you download with a browser. Or try what I did and see if 
it works.

What version are you using? 1.8.9 is the current one.

Tilman

Am 13.12.2014 um 18:41 schrieb Walter Kehl:
> Hi John, Tilman,
>
> thanks for the reply. Here is some additional information:
>
> - the http client I am using to get the input stream already has a user
> agent set. Also I have downloaded with PDF box already lots of PDF files
> where there never was a problem.
> - when I try to load the document remotely from the URL, I get the following
> error messages:
>    18:34:32 WARN  BaseParser           :: Specified stream length 66346 is
> wrong. Fall back to reading stream until 'endstream'.
>    18:34:35 WARN  XrefTrailerResolver  :: Did not found XRef object at
> specified startxref position 0
> - I have written the input stream directly to a file and it was a valid PDF.
> It could load it both with an external tool and with PDFBox.
>
> Yes, of course I could always download a file first to a temp file and then
> load it into PDFBox. But I think the direct way is more elegant and faster.
> I have also debugged a little bit into the code and to me it doesn't look
> like PDFBox uses a temporary file, but rather reads directly from the input
> stream.... but I might be wrong.
>
> Anyway, thanks for providing such a good free software!
>
> Best
> Walter
>
> -----Original Message-----
> From: John Hewson [mailto:john@jahewson.com]
> Sent: Freitag, 12. Dezember 2014 18:57
> To: users@pdfbox.apache.org
> Subject: Re: Downloadind a pdf file doesn't work
>
> Good point Tilman. Walter, try saving writing the InputStream to a File and
> check that it's a valid PDF.
>
> -- John
>
>> On 12 Dec 2014, at 09:50, Tilman Hausherr <TH...@t-online.de> wrote:
>>
>> This sounds more like a http problem. Try setting a user agent like a
> browser.
>> https://stackoverflow.com/questions/2529682/setting-user-agent-of-a-ja
>> va-urlconnection
>>
>> Tilman
>>
>> Am 12.12.2014 um 11:53 schrieb Walter Kehl:
>>> Hi all,
>>>
>>>   
>>> I have the following situation:
>>>
>>>   
>>> I am loading with PdfBox files from the internet with the call
>>>
>>> PDDocument document = PDDocument.load( inputStream );
>>>
>>>   
>>> So far it has worked nicely, but I have problems with this file :
>>> http://esa.un.org/unpd/wup/PressRelease/WUP2014_PressRelease.pdf
>>>
>>>   
>>> After I load it, it is empty, and the call
>>> document.getNumberOfPages() returns 0.
>>>
>>> However when I download the file manually and then load it into
>>> PdfBox, then everything is fine.
>>>
>>>   
>>> Any idea what could be happening? I am currently using PdfBox 1.8.5.
>>>
>>>   
>>> Thanks and Best Regards
>>>
>>> Walter
>>>
>>>   
>>>   
>>>   
>>>


RE: Downloadind a pdf file doesn't work

Posted by Walter Kehl <wa...@outlook.com>.
Hi John, Tilman, 

thanks for the reply. Here is some additional information:

- the http client I am using to get the input stream already has a user
agent set. Also I have downloaded with PDF box already lots of PDF files
where there never was a problem. 
- when I try to load the document remotely from the URL, I get the following
error messages:
  18:34:32 WARN  BaseParser           :: Specified stream length 66346 is
wrong. Fall back to reading stream until 'endstream'.
  18:34:35 WARN  XrefTrailerResolver  :: Did not found XRef object at
specified startxref position 0
- I have written the input stream directly to a file and it was a valid PDF.
It could load it both with an external tool and with PDFBox. 

Yes, of course I could always download a file first to a temp file and then
load it into PDFBox. But I think the direct way is more elegant and faster.
I have also debugged a little bit into the code and to me it doesn't look
like PDFBox uses a temporary file, but rather reads directly from the input
stream.... but I might be wrong.

Anyway, thanks for providing such a good free software!

Best
Walter

-----Original Message-----
From: John Hewson [mailto:john@jahewson.com] 
Sent: Freitag, 12. Dezember 2014 18:57
To: users@pdfbox.apache.org
Subject: Re: Downloadind a pdf file doesn't work

Good point Tilman. Walter, try saving writing the InputStream to a File and
check that it's a valid PDF.

-- John

> On 12 Dec 2014, at 09:50, Tilman Hausherr <TH...@t-online.de> wrote:
> 
> This sounds more like a http problem. Try setting a user agent like a
browser.
> 
> https://stackoverflow.com/questions/2529682/setting-user-agent-of-a-ja
> va-urlconnection
> 
> Tilman
> 
> Am 12.12.2014 um 11:53 schrieb Walter Kehl:
>> Hi all,
>> 
>>  
>> I have the following situation:
>> 
>>  
>> I am loading with PdfBox files from the internet with the call
>> 
>> PDDocument document = PDDocument.load( inputStream );
>> 
>>  
>> So far it has worked nicely, but I have problems with this file :
>> http://esa.un.org/unpd/wup/PressRelease/WUP2014_PressRelease.pdf
>> 
>>  
>> After I load it, it is empty, and the call 
>> document.getNumberOfPages() returns 0.
>> 
>> However when I download the file manually and then load it into 
>> PdfBox, then everything is fine.
>> 
>>  
>> Any idea what could be happening? I am currently using PdfBox 1.8.5.
>> 
>>  
>> Thanks and Best Regards
>> 
>> Walter
>> 
>>  
>>  
>>  
>> 
> 


Re: Downloadind a pdf file doesn't work

Posted by John Hewson <jo...@jahewson.com>.
Good point Tilman… Walter, try saving writing the InputStream to a File and check that it’s a valid PDF.

-- John

> On 12 Dec 2014, at 09:50, Tilman Hausherr <TH...@t-online.de> wrote:
> 
> This sounds more like a http problem. Try setting a user agent like a browser.
> 
> https://stackoverflow.com/questions/2529682/setting-user-agent-of-a-java-urlconnection
> 
> Tilman
> 
> Am 12.12.2014 um 11:53 schrieb Walter Kehl:
>> Hi all,
>> 
>>  
>> I have the following situation:
>> 
>>  
>> I am loading with PdfBox files from the internet with the call
>> 
>> PDDocument document = PDDocument.load( inputStream );
>> 
>>  
>> So far it has worked nicely, but I have problems with this file :
>> http://esa.un.org/unpd/wup/PressRelease/WUP2014_PressRelease.pdf
>> 
>>  
>> After I load it, it is empty, and the call document.getNumberOfPages()
>> returns 0.
>> 
>> However when I download the file manually and then load it into PdfBox, then
>> everything is fine.
>> 
>>  
>> Any idea what could be happening? I am currently using PdfBox 1.8.5.
>> 
>>  
>> Thanks and Best Regards
>> 
>> Walter
>> 
>>  
>>  
>>  
>> 
> 


Re: Downloadind a pdf file doesn't work

Posted by Tilman Hausherr <TH...@t-online.de>.
This sounds more like a http problem. Try setting a user agent like a 
browser.

https://stackoverflow.com/questions/2529682/setting-user-agent-of-a-java-urlconnection

Tilman

Am 12.12.2014 um 11:53 schrieb Walter Kehl:
> Hi all,
>
>   
>
> I have the following situation:
>
>   
>
> I am loading with PdfBox files from the internet with the call
>
> PDDocument document = PDDocument.load( inputStream );
>
>   
>
> So far it has worked nicely, but I have problems with this file :
> http://esa.un.org/unpd/wup/PressRelease/WUP2014_PressRelease.pdf
>
>   
>
> After I load it, it is empty, and the call document.getNumberOfPages()
> returns 0.
>
> However when I download the file manually and then load it into PdfBox, then
> everything is fine.
>
>   
>
> Any idea what could be happening? I am currently using PdfBox 1.8.5.
>
>   
>
> Thanks and Best Regards
>
> Walter
>
>   
>
>   
>
>   
>
>


Re: Downloadind a pdf file doesn't work

Posted by John Hewson <jo...@jahewson.com>.
That looks like a bug. As it happens, PDFBox will just write your InputStream to a temporary File anyway. The workaround is to just save your data to a temporary file yourself and then call PDDocument.load(file).

-- John

> On 12 Dec 2014, at 02:53, Walter Kehl <wa...@outlook.com> wrote:
> 
> Hi all,
> 
> 
> 
> I have the following situation:
> 
> 
> 
> I am loading with PdfBox files from the internet with the call
> 
> PDDocument document = PDDocument.load( inputStream );
> 
> 
> 
> So far it has worked nicely, but I have problems with this file :
> http://esa.un.org/unpd/wup/PressRelease/WUP2014_PressRelease.pdf
> 
> 
> 
> After I load it, it is empty, and the call document.getNumberOfPages()
> returns 0.
> 
> However when I download the file manually and then load it into PdfBox, then
> everything is fine.
> 
> 
> 
> Any idea what could be happening? I am currently using PdfBox 1.8.5.
> 
> 
> 
> Thanks and Best Regards
> 
> Walter
> 
> 
> 
> 
> 
> 
>