Hi folks,
We have a number of .NET applications which use Crystal Reports for VS for transactional reporting. The primary export format is PDF, although we do some Excel exports as well.
We've recently started checking the accessibility of our reports. For the PDF exports we're using PDF Accessiblity Checker:
http://www.access-for-all.ch/en/pdf-lab/pdf-accessibility-checker-pac.html
It's identified a number of issues in our exported PDFs. I did some Googling to try and find out how other people have handled these in Crystal, but I was only able to come up with people asking about accessibility on these forums. Is there any guidance on how to make PDFs exported from Crystal accessible? Here are the top three issues that our QA department has found in our exported PDFs:
- Tagged PDF documents
Tags are the foundation of accessible PDF documents. They deliver the structural information, which enables assistive technology, such as screen readers, to
- Properly recognise different text elements, such as headings, lists, graphics and tables.
- Read the document following the correct reading order.
Many authoring systems are capable of producing tagged PDF documents. Unfortunately, however, the quality of those automatically generated tagged documents is often insufficient, which makes additional visual as well as technical checks inevitable.
3.1 Check 1: Document is marked as tagged
This check evaluates if the document is marked as containing tags.
This is necessary because some viewers and screen readers don't recognise tags within a PDF document unless it specifically contains this flag.WCAG 2.0 | Principle 1: Perceivable |
ISO 32000-1:2008 | Table 321, Key 'Marked' |
There is a defined set of PDF standard tags in order to mark content in a PDF file. The naming and usage is described in the PDF specification. Typical standard tags are
- <H1> for heading level 1,
- <P> for paragraph,
- <L> for list or
- <Table> for table.
The name of a PDF standard tag is equivalent to its role. A role mapping is not necessary.
User-defined PDF Tags
The usage of user-defined tags is also possible. But you have to map the role of such tags to PDF standard tags.
Limited range of structure types
Unfortunately, the range of PDF standard tags is limited. You cannot mark every piece of content with an appropriate structure type, yet. For example, there is no tag for a title or for elements beside the main content flow like infoboxes or sidenotes.
2. Natural language of text object cannot be determined
What is the issue?
PDF/UA > Basic requirements > Natural language > Natural language of text object > Natural language of text object cannot be determined
What does it mean?
There is no document language determined.
3. Metadata and Settings
What is the issue?
PDF/UA > Metadata and Settings > Metadata > XMP Metadata entry in Catalog > XMP metadata stream missing in document
What does it mean?
There are no document metadata at all available in your document.
Thanks!
Cory