Summary of the Structure of PDF files

Deeper Insight into the Complex Structure of PDF Files and Their Key Components.

PDF file format structure can be looked upon as a combination of different file types presented in a single container. The reason for this is that a PDF file contains Text, vector art, images, fonts and other file format can be embedded – even the native files that were used to create the PDF in the first place.

The complex structure of PDF files consists of objects where items can be connected directly or indirectly to each other. Often the indirection is because an object might is used multiple times as would be the case for a logo, font, color.
pdf structure

The objects within a PDF file can be divided into the following types:

Dictionaries

A group containing direct or references to indirect objects. Dictionaries can be seen as the glue holding together the elements in a PDF files. The example below shows the structure of a typical page dictionary:

pdf page structure
The Contents stream has an attributes dictionary that contains a filter name and the length of the stream
The CropBox array contains the coordinates of the rectangle that defines the area that is visible on the page.
The MediaBox array contains the coordinates of the rectangle that defines the media size. This will typically match a standard media size such as Letter or A4 and will allow the PDF page to be reliably printed on a device that contains these standard media sizes.
The Resources dictionary contains references and information for elements that are needed to reliably output the visual elements of the page such as colors, fonts and Images.

Streams

The collection of operators outputting information onto the page. Normally the stream will also require elements of the page resources dictionary such as colors and fonts. Streams are either stored as a single element or in an array.
 
				
					q
567.48 61.011 -540 720 re
W* n
q
/GS0 gs
0 720 -541.1399536 0 567.4799194 61.0105438 cm
/Im0 Do
Q
Q
/CS0 cs 0.302 0.302 0.302  scn
1 i 
/GS1 gs
56.7 286.911 m
56.7 295.191 56.7 303.471 56.7 311.751 c
59.1 311.751 61.5 311.751 63.9 311.751 c
63.9 306.831 63.9 301.911 63.9 296.991 c
65.88 296.991 67.8 296.991 69.72 296.991 c
69.72 301.191 69.72 305.391 69.72 309.591 c
72 309.591 74.22 309.591 76.5 309.591 c
76.5 305.391 76.5 301.191 76.5 296.991 c
81.06 296.991 85.62 296.991 90.18 296.991 c
90.18 293.631 90.18 290.271 90.18 286.911 c
79.02 286.911 67.86 286.911 56.7 286.911 c
f*
You can see that there are several references to items in the page resources dictionary:
GS0 is a reference to a graphics state and gs is the operator that sets it.
Im0 is an XObject image and the Do operator draws the image.
CS0 is a reference to a color dictionary and the scn operator assigns it to strokes.
 
You can also see usage of several path operators re – rectangle, m – moveto, c – curve f* – fill.
 

Text strings

These can either be ANSI (single byte characters) or Unicode (multi-byte). The example here is the representation of the last date modified in the catalog dictionary.These can either be ANSI (single byte characters) or Unicode (multi-byte). The example here is the representation of the last date modified in the catalog dictionary.

PDF text

Images

Images are normally held within the page resources and the stream will also have an associated Attributes dictionary that will describe the attributes of the data within the stream. BitsPerComponent size of the data that is used to define a single pixel (dot) within the image. The ColorSpace dictionary describes the colour model that is used to define the colors within the image.

PDF image

Names

Used normally to provide a name that can be used to refer to a dictionary or dictionary item. For example, the pages dictionary has a name “Type” with the value “Pages” and a single page has a name of “Type” with a value of “Page”.

PDF names

Arrays

Fixed length data holding types and/or references to other elements. For an example see the Real Numbers example below.

Real numbers

Decimal numbers. In this example they are being used to define the rectangle of the page media box:

PDF real numbers

Integers

Whole numbers. For example to show the total number of  pages in the PDF file.

PDF integer
For further details on pdf file format structure see the PDF Specification at https://www.adobe.com/devnet/pdf/pdf_reference.html
 
Contact:
 
Michael Peters

Associated Links:

Share the Post:

Related Posts

Join Our Newsletter

Summary of the Structure of PDF files

Deeper Insight into the Complex Structure of PDF Files and Their Key Components.

PDF file format structure can be looked upon as a combination of different file types presented in a single container. The reason for this is that a PDF file contains Text, vector art, images, fonts and other file format can be embedded – even the native files that were used to create the PDF in the first place.

The complex structure of PDF files consists of objects where items can be connected directly or indirectly to each other. Often the indirection is because an object might is used multiple times as would be the case for a logo, font, color.
pdf structure

The objects within a PDF file can be divided into the following types:

Dictionaries

A group containing direct or references to indirect objects. Dictionaries can be seen as the glue holding together the elements in a PDF files. The example below shows the structure of a typical page dictionary:

pdf page structure
The Contents stream has an attributes dictionary that contains a filter name and the length of the stream
The CropBox array contains the coordinates of the rectangle that defines the area that is visible on the page.
The MediaBox array contains the coordinates of the rectangle that defines the media size. This will typically match a standard media size such as Letter or A4 and will allow the PDF page to be reliably printed on a device that contains these standard media sizes.
The Resources dictionary contains references and information for elements that are needed to reliably output the visual elements of the page such as colors, fonts and Images.

Streams

The collection of operators outputting information onto the page. Normally the stream will also require elements of the page resources dictionary such as colors and fonts. Streams are either stored as a single element or in an array.
 
				
					q
567.48 61.011 -540 720 re
W* n
q
/GS0 gs
0 720 -541.1399536 0 567.4799194 61.0105438 cm
/Im0 Do
Q
Q
/CS0 cs 0.302 0.302 0.302  scn
1 i 
/GS1 gs
56.7 286.911 m
56.7 295.191 56.7 303.471 56.7 311.751 c
59.1 311.751 61.5 311.751 63.9 311.751 c
63.9 306.831 63.9 301.911 63.9 296.991 c
65.88 296.991 67.8 296.991 69.72 296.991 c
69.72 301.191 69.72 305.391 69.72 309.591 c
72 309.591 74.22 309.591 76.5 309.591 c
76.5 305.391 76.5 301.191 76.5 296.991 c
81.06 296.991 85.62 296.991 90.18 296.991 c
90.18 293.631 90.18 290.271 90.18 286.911 c
79.02 286.911 67.86 286.911 56.7 286.911 c
f*
You can see that there are several references to items in the page resources dictionary:
GS0 is a reference to a graphics state and gs is the operator that sets it.
Im0 is an XObject image and the Do operator draws the image.
CS0 is a reference to a color dictionary and the scn operator assigns it to strokes.
 
You can also see usage of several path operators re – rectangle, m – moveto, c – curve f* – fill.
 

Text strings

These can either be ANSI (single byte characters) or Unicode (multi-byte). The example here is the representation of the last date modified in the catalog dictionary.These can either be ANSI (single byte characters) or Unicode (multi-byte). The example here is the representation of the last date modified in the catalog dictionary.

PDF text

Images

Images are normally held within the page resources and the stream will also have an associated Attributes dictionary that will describe the attributes of the data within the stream. BitsPerComponent size of the data that is used to define a single pixel (dot) within the image. The ColorSpace dictionary describes the colour model that is used to define the colors within the image.

PDF image

Names

Used normally to provide a name that can be used to refer to a dictionary or dictionary item. For example, the pages dictionary has a name “Type” with the value “Pages” and a single page has a name of “Type” with a value of “Page”.

PDF names

Arrays

Fixed length data holding types and/or references to other elements. For an example see the Real Numbers example below.

Real numbers

Decimal numbers. In this example they are being used to define the rectangle of the page media box:

PDF real numbers

Integers

Whole numbers. For example to show the total number of  pages in the PDF file.

PDF integer
For further details on pdf file format structure see the PDF Specification at https://www.adobe.com/devnet/pdf/pdf_reference.html
 
Contact:
 
Michael Peters

Associated Links:

Share the Post:

Related Posts

Join Our Newsletter