PdfSmartCopy generating bad PDFs

[Gylfi Ingvason]

The PDF file below, when modified and concatenated with a similar file, will
generate bad PDF output using PdfSmartCopy. This can be replicated doing the
following:

1. Cut and paste the PDF file below into a separate .pdf file. Use an editor
that does not molest line terminators and make sure no line wrapping is
taking place - all dictionary entries should be contained within a single
line and not broken up.

2. After you have verified that the file opens and displays in Acrobat (if
it asks to save changes when you close it, it means that line terminators
are not correct), make another copy of the file, and change "(Blue Content)"
to "(Red  Content)" and in the line above  change "0 0 1 rg" to "1 0 0 rg".
Make sure to add an extra space after "Red" to keep the byte count the same.

3. Concatenate the two files using the following code or something similar:

PdfReader reader = new PdfReader("File1.pdf");
Document document = new Document(reader.GetPageSizeWithRotation(1));
PdfSmartCopy writer = new PdfSmartCopy(document, new
FileStream("Concatenated.pdf", FileMode.Create, FileAccess.Write));
document.Open();

writer.AddPage(writer.GetImportedPage(reader, 1));
reader.Close();

reader = new PdfReader("File2.pdf");
writer.AddPage(writer.GetImportedPage(reader, 1));
reader.Close();
writer.Close();

5. The "Concatenated.pdf" will contain two pages that both have the "Blue
Content" displayed. Upon examining the generated file, you will find that
both pages reference the same content and the "Red  Content" is missing. A
slightly different composition of the input file will generate output where
an XObject Form will reference itself which causes the Foxit reader to
immediately kill itself and cause Acrobat to complain about a damaged file.

The pdf file below was hand created whittling down concatenated customer
files where single complex invoices and checks were being prepared for
printing. Personally, I find the re-use of named resources like this
somewhat questionable, but the PDF specification seems to indicate that this
is valid since the name re-use is within the scope of the Form only.

At any rate, I would like the group to know that using PdfSmartCopy can
produce erroneous, and under some circumstances, deadly (for the Reader)
results. I don't trust myself to debug the source for this, but perhaps
Paulo or somebody else with more knowledge of the code can take a closer
look and patch this.

Thanks - Gylfi

%PDF-1.4
%âãÏÓ
1 0 obj<</Pages 2 0 R/Type/Catalog>>
endobj
2 0 obj<</Kids[3 0 R]/Count 1/Type/Pages>>
endobj
3 0 obj<</Resources<</XObject<</Xf1 5 0 R>>/ProcSet [/PDF /Text]>>/Parent 2
0 R/MediaBox[0 0 612 792]/Contents 4 0 R/Type/Page>>
endobj
4 0 obj <</Length 26>>stream
q 1 0 0 1 0 0 cm /Xf1 Do Q
endstream
endobj
5 0 obj <</BBox[0 0 612 792]/Matrix[1 0 0 1 0 0]/FormType
1/Subtype/Form/Length 26/Resources<</XObject<</Xf1 6 0
R>>/ProcSet[/PDF/Text]>>/Type/XObject>>stream
q 1 0 0 1 0 0 cm /Xf1 Do Q
endstream
endobj
6 0 obj <</BBox[0 0 612 792]/Matrix[1 0 0 1 0 0]/FormType
1/Subtype/Form/Length 26/Resources<</XObject<</Xf1 7 0
R>>/ProcSet[/PDF/Text]>>/Type/XObject>>stream
q 1 0 0 1 0 0 cm /Xf1 Do Q
endstream
endobj
7 0 obj <</BBox[0 0 612 792]/Matrix[1 0 0 1 0 0]/FormType
1/Subtype/Form/Length 26/Resources<</XObject<</Xf1 8 0
R>>/ProcSet[/PDF/Text]>>/Type/XObject>>stream
q 1 0 0 1 0 0 cm /Xf1 Do Q
endstream
endobj
8 0 obj <</BBox[0 0 612 792]/Matrix[1 0 0 1 0 0]/FormType
1/Subtype/Form/Length 26/Resources<</XObject<</Xf1 9 0
R>>/ProcSet[/PDF/Text]>>/Type/XObject>>stream
q 1 0 0 1 0 0 cm /Xf1 Do Q
endstream
endobj
9 0 obj <</BBox[0 0 612 792]/Matrix[1 0 0 1 0 0]/FormType
1/Subtype/Form/Length 67/Resources<</Font<</F1 10 0
R>>/ProcSet[/PDF/Text]>>/Type/XObject>>stream
q
BT
1 0 0 1 10 700 Tm
/F1 15 Tf
0 0 1 rg
(Blue Content)Tj
0 g
ET
Q
endstream
endobj
10 0
obj<</Subtype/Type1/BaseFont/Courier/Encoding/WinAnsiEncoding/Type/Font>>
endobj
xref
0 11
0000000000 65535 f
0000000015 00000 n
0000000059 00000 n
0000000109 00000 n
0000000245 00000 n
0000000318 00000 n
0000000521 00000 n
0000000724 00000 n
0000000927 00000 n
0000001130 00000 n
0000001371 00000 n
trailer
<</Root 1 0 R/Size 11>>
startxref
1457
%%EOF



------------------------------------------------------------------------------
Crystal Reports - New Free Runtime and 30 Day Trial
Check out the new simplified licensing option that enables unlimited
royalty-free distribution of the report engine for externally facing
server and web deployment.
http://p.sf.net/sfu/businessobjects
_______________________________________________
itextsharp-questions mailing list
itextsharp-questions@...
https://lists.sourceforge.net/lists/listinfo/itextsharp-questions

[Paulo Soares-3]

Fixed in the SVN. This month will come out a new release if you want to wait for it.

Paulo

> -----Original Message-----
> From: Gylfi Ingvason [mailto:Gylfi.Ingvason@...]
> Sent: Thursday, June 18, 2009 6:11 PM
> To: itextsharp-questions@...
> Subject: [itextsharp-questions] PdfSmartCopy generating bad PDFs
>
>
> The PDF file below, when modified and concatenated with a
> similar file, will
> generate bad PDF output using PdfSmartCopy. This can be
> replicated doing the
> following:
>
> 1. Cut and paste the PDF file below into a separate .pdf
> file. Use an editor
> that does not molest line terminators and make sure no line
> wrapping is
> taking place - all dictionary entries should be contained
> within a single
> line and not broken up.
>
> 2. After you have verified that the file opens and displays
> in Acrobat (if
> it asks to save changes when you close it, it means that line
> terminators
> are not correct), make another copy of the file, and change
> "(Blue Content)"
> to "(Red  Content)" and in the line above  change "0 0 1 rg"
> to "1 0 0 rg".
> Make sure to add an extra space after "Red" to keep the byte
> count the same.
>
> 3. Concatenate the two files using the following code or
> something similar:
>
> PdfReader reader = new PdfReader("File1.pdf");
> Document document = new Document(reader.GetPageSizeWithRotation(1));
> PdfSmartCopy writer = new PdfSmartCopy(document, new
> FileStream("Concatenated.pdf", FileMode.Create, FileAccess.Write));
> document.Open();
>
> writer.AddPage(writer.GetImportedPage(reader, 1));
> reader.Close();
>
> reader = new PdfReader("File2.pdf");
> writer.AddPage(writer.GetImportedPage(reader, 1));
> reader.Close();
> writer.Close();
>
> 5. The "Concatenated.pdf" will contain two pages that both
> have the "Blue
> Content" displayed. Upon examining the generated file, you
> will find that
> both pages reference the same content and the "Red  Content"
> is missing. A
> slightly different composition of the input file will
> generate output where
> an XObject Form will reference itself which causes the Foxit reader to
> immediately kill itself and cause Acrobat to complain about a
> damaged file.
>
> The pdf file below was hand created whittling down
> concatenated customer
> files where single complex invoices and checks were being prepared for
> printing. Personally, I find the re-use of named resources like this
> somewhat questionable, but the PDF specification seems to
> indicate that this
> is valid since the name re-use is within the scope of the Form only.
>
> At any rate, I would like the group to know that using
> PdfSmartCopy can
> produce erroneous, and under some circumstances, deadly (for
> the Reader)
> results. I don't trust myself to debug the source for this,
> but perhaps
> Paulo or somebody else with more knowledge of the code can
> take a closer
> look and patch this.
>
> Thanks - Gylfi
>
> %PDF-1.4
> %âãÏÓ
> 1 0 obj<</Pages 2 0 R/Type/Catalog>>
> endobj
> 2 0 obj<</Kids[3 0 R]/Count 1/Type/Pages>>
> endobj
> 3 0 obj<</Resources<</XObject<</Xf1 5 0 R>>/ProcSet [/PDF
> /Text]>>/Parent 2
> 0 R/MediaBox[0 0 612 792]/Contents 4 0 R/Type/Page>>
> endobj
> 4 0 obj <</Length 26>>stream
> q 1 0 0 1 0 0 cm /Xf1 Do Q
> endstream
> endobj
> 5 0 obj <</BBox[0 0 612 792]/Matrix[1 0 0 1 0 0]/FormType
> 1/Subtype/Form/Length 26/Resources<</XObject<</Xf1 6 0
> R>>/ProcSet[/PDF/Text]>>/Type/XObject>>stream
> q 1 0 0 1 0 0 cm /Xf1 Do Q
> endstream
> endobj
> 6 0 obj <</BBox[0 0 612 792]/Matrix[1 0 0 1 0 0]/FormType
> 1/Subtype/Form/Length 26/Resources<</XObject<</Xf1 7 0
> R>>/ProcSet[/PDF/Text]>>/Type/XObject>>stream
> q 1 0 0 1 0 0 cm /Xf1 Do Q
> endstream
> endobj
> 7 0 obj <</BBox[0 0 612 792]/Matrix[1 0 0 1 0 0]/FormType
> 1/Subtype/Form/Length 26/Resources<</XObject<</Xf1 8 0
> R>>/ProcSet[/PDF/Text]>>/Type/XObject>>stream
> q 1 0 0 1 0 0 cm /Xf1 Do Q
> endstream
> endobj
> 8 0 obj <</BBox[0 0 612 792]/Matrix[1 0 0 1 0 0]/FormType
> 1/Subtype/Form/Length 26/Resources<</XObject<</Xf1 9 0
> R>>/ProcSet[/PDF/Text]>>/Type/XObject>>stream
> q 1 0 0 1 0 0 cm /Xf1 Do Q
> endstream
> endobj
> 9 0 obj <</BBox[0 0 612 792]/Matrix[1 0 0 1 0 0]/FormType
> 1/Subtype/Form/Length 67/Resources<</Font<</F1 10 0
> R>>/ProcSet[/PDF/Text]>>/Type/XObject>>stream
> q
> BT
> 1 0 0 1 10 700 Tm
> /F1 15 Tf
> 0 0 1 rg
> (Blue Content)Tj
> 0 g
> ET
> Q
> endstream
> endobj
> 10 0
> obj<</Subtype/Type1/BaseFont/Courier/Encoding/WinAnsiEncoding/
> Type/Font>>
> endobj
> xref
> 0 11
> 0000000000 65535 f
> 0000000015 00000 n
> 0000000059 00000 n
> 0000000109 00000 n
> 0000000245 00000 n
> 0000000318 00000 n
> 0000000521 00000 n
> 0000000724 00000 n
> 0000000927 00000 n
> 0000001130 00000 n
> 0000001371 00000 n
> trailer
> <</Root 1 0 R/Size 11>>
> startxref
> 1457
> %%EOF

Aviso Legal:
Esta mensagem é destinada exclusivamente ao destinatário. Pode conter informação confidencial ou legalmente protegida. A incorrecta transmissão desta mensagem não significa a perca de confidencialidade. Se esta mensagem for recebida por engano, por favor envie-a de volta para o remetente e apague-a do seu sistema de imediato. É proibido a qualquer pessoa que não o destinatário de usar, revelar ou distribuir qualquer parte desta mensagem.

Disclaimer:
This message is destined exclusively to the intended receiver. It may contain confidential or legally protected information. The incorrect transmission of this message does not mean the loss of its confidentiality. If this message is received by mistake, please send it back to the sender and delete it from your system immediately. It is forbidden to any person who is not the intended receiver to use, distribute or copy any part of this message.

------------------------------------------------------------------------------
Crystal Reports - New Free Runtime and 30 Day Trial
Check out the new simplified licensing option that enables unlimited
royalty-free distribution of the report engine for externally facing
server and web deployment.
http://p.sf.net/sfu/businessobjects
_______________________________________________
itextsharp-questions mailing list
itextsharp-questions@...
https://lists.sourceforge.net/lists/listinfo/itextsharp-questions

[Gylfi Ingvason]

Thanks Paulo. I will grab the SVN copy and test with it. Again, many thanks
for the quick repair.

Gylfi

-----Original Message-----
From: Paulo Soares [mailto:psoares@...]
Sent: Friday, June 19, 2009 6:56 AM
To: Post all your questions about iTextSharp here
Subject: Re: [itextsharp-questions] PdfSmartCopy generating bad PDFs

Fixed in the SVN. This month will come out a new release if you want to wait
for it.

Paulo

> -----Original Message-----
> From: Gylfi Ingvason [mailto:Gylfi.Ingvason@...]
> Sent: Thursday, June 18, 2009 6:11 PM
> To: itextsharp-questions@...
> Subject: [itextsharp-questions] PdfSmartCopy generating bad PDFs
>
>
> The PDF file below, when modified and concatenated with a similar
> file, will generate bad PDF output using PdfSmartCopy. This can be
> replicated doing the
> following:
>
> 1. Cut and paste the PDF file below into a separate .pdf file. Use an
> editor that does not molest line terminators and make sure no line
> wrapping is taking place - all dictionary entries should be contained
> within a single line and not broken up.
>
> 2. After you have verified that the file opens and displays in Acrobat
> (if it asks to save changes when you close it, it means that line
> terminators are not correct), make another copy of the file, and
> change "(Blue Content)"
> to "(Red  Content)" and in the line above  change "0 0 1 rg"
> to "1 0 0 rg".
> Make sure to add an extra space after "Red" to keep the byte count the
> same.
>
> 3. Concatenate the two files using the following code or something
> similar:
>
> PdfReader reader = new PdfReader("File1.pdf"); Document document = new
> Document(reader.GetPageSizeWithRotation(1));
> PdfSmartCopy writer = new PdfSmartCopy(document, new
> FileStream("Concatenated.pdf", FileMode.Create, FileAccess.Write));
> document.Open();
>
> writer.AddPage(writer.GetImportedPage(reader, 1)); reader.Close();
>
> reader = new PdfReader("File2.pdf");
> writer.AddPage(writer.GetImportedPage(reader, 1)); reader.Close();
> writer.Close();
>
> 5. The "Concatenated.pdf" will contain two pages that both have the
> "Blue Content" displayed. Upon examining the generated file, you will
> find that both pages reference the same content and the "Red  Content"
> is missing. A
> slightly different composition of the input file will generate output
> where an XObject Form will reference itself which causes the Foxit
> reader to immediately kill itself and cause Acrobat to complain about
> a damaged file.
>
> The pdf file below was hand created whittling down concatenated
> customer files where single complex invoices and checks were being
> prepared for printing. Personally, I find the re-use of named
> resources like this somewhat questionable, but the PDF specification
> seems to indicate that this is valid since the name re-use is within
> the scope of the Form only.
>
> At any rate, I would like the group to know that using PdfSmartCopy
> can produce erroneous, and under some circumstances, deadly (for the
> Reader) results. I don't trust myself to debug the source for this,
> but perhaps Paulo or somebody else with more knowledge of the code can
> take a closer look and patch this.
>
> Thanks - Gylfi
>
> %PDF-1.4
> %âãÏÓ
> 1 0 obj<</Pages 2 0 R/Type/Catalog>>
> endobj
> 2 0 obj<</Kids[3 0 R]/Count 1/Type/Pages>> endobj
> 3 0 obj<</Resources<</XObject<</Xf1 5 0 R>>/ProcSet [/PDF
> /Text]>>/Parent 2 0 R/MediaBox[0 0 612 792]/Contents 4 0 R/Type/Page>>
> endobj
> 4 0 obj <</Length 26>>stream
> q 1 0 0 1 0 0 cm /Xf1 Do Q
> endstream
> endobj
> 5 0 obj <</BBox[0 0 612 792]/Matrix[1 0 0 1 0 0]/FormType
> 1/Subtype/Form/Length 26/Resources<</XObject<</Xf1 6 0
> R>>/ProcSet[/PDF/Text]>>/Type/XObject>>stream
> q 1 0 0 1 0 0 cm /Xf1 Do Q
> endstream
> endobj
> 6 0 obj <</BBox[0 0 612 792]/Matrix[1 0 0 1 0 0]/FormType
> 1/Subtype/Form/Length 26/Resources<</XObject<</Xf1 7 0
> R>>/ProcSet[/PDF/Text]>>/Type/XObject>>stream
> q 1 0 0 1 0 0 cm /Xf1 Do Q
> endstream
> endobj
> 7 0 obj <</BBox[0 0 612 792]/Matrix[1 0 0 1 0 0]/FormType
> 1/Subtype/Form/Length 26/Resources<</XObject<</Xf1 8 0
> R>>/ProcSet[/PDF/Text]>>/Type/XObject>>stream
> q 1 0 0 1 0 0 cm /Xf1 Do Q
> endstream
> endobj
> 8 0 obj <</BBox[0 0 612 792]/Matrix[1 0 0 1 0 0]/FormType
> 1/Subtype/Form/Length 26/Resources<</XObject<</Xf1 9 0
> R>>/ProcSet[/PDF/Text]>>/Type/XObject>>stream
> q 1 0 0 1 0 0 cm /Xf1 Do Q
> endstream
> endobj
> 9 0 obj <</BBox[0 0 612 792]/Matrix[1 0 0 1 0 0]/FormType
> 1/Subtype/Form/Length 67/Resources<</Font<</F1 10 0
> R>>/ProcSet[/PDF/Text]>>/Type/XObject>>stream
> q
> BT
> 1 0 0 1 10 700 Tm
> /F1 15 Tf
> 0 0 1 rg
> (Blue Content)Tj
> 0 g
> ET
> Q
> endstream
> endobj
> 10 0
> obj<</Subtype/Type1/BaseFont/Courier/Encoding/WinAnsiEncoding/
> Type/Font>>
> endobj
> xref
> 0 11
> 0000000000 65535 f
> 0000000015 00000 n
> 0000000059 00000 n
> 0000000109 00000 n
> 0000000245 00000 n
> 0000000318 00000 n
> 0000000521 00000 n
> 0000000724 00000 n
> 0000000927 00000 n
> 0000001130 00000 n
> 0000001371 00000 n
> trailer
> <</Root 1 0 R/Size 11>>
> startxref
> 1457
> %%EOF

Aviso Legal:
Esta mensagem é destinada exclusivamente ao destinatário. Pode conter
informação confidencial ou legalmente protegida. A incorrecta transmissão
desta mensagem não significa a perca de confidencialidade. Se esta mensagem
for recebida por engano, por favor envie-a de volta para o remetente e
apague-a do seu sistema de imediato. É proibido a qualquer pessoa que não o
destinatário de usar, revelar ou distribuir qualquer parte desta mensagem.

Disclaimer:
This message is destined exclusively to the intended receiver. It may
contain confidential or legally protected information. The incorrect
transmission of this message does not mean the loss of its confidentiality.
If this message is received by mistake, please send it back to the sender
and delete it from your system immediately. It is forbidden to any person
who is not the intended receiver to use, distribute or copy any part of this
message.

----------------------------------------------------------------------------
--
Crystal Reports - New Free Runtime and 30 Day Trial Check out the new
simplified licensing option that enables unlimited royalty-free distribution
of the report engine for externally facing server and web deployment.
http://p.sf.net/sfu/businessobjects
_______________________________________________
itextsharp-questions mailing list
itextsharp-questions@...
https://lists.sourceforge.net/lists/listinfo/itextsharp-questions



------------------------------------------------------------------------------
Crystal Reports - New Free Runtime and 30 Day Trial
Check out the new simplified licensing option that enables unlimited
royalty-free distribution of the report engine for externally facing
server and web deployment.
http://p.sf.net/sfu/businessobjects
_______________________________________________
itextsharp-questions mailing list
itextsharp-questions@...
https://lists.sourceforge.net/lists/listinfo/itextsharp-questions