This page is for XBRL software vendors. It provides information about:
- How to enable DQR rule processing in software;
- The upcoming DQR certification process;
- The upcoming DQR governance function; and
- Timing of DQR rule deployment.
Executing DQR Rules
The DQR Rules are documented on this site, and are free to run. For each rule there is a “rule description” giving a high-level overview of the rule, and a “rule specification” that gives a formal definition of the rule.
A conformance suite for the rules is currently under development. This will provide minimal pass and fail cases for each rule. The conformance suite also contains the expected results of running the rules against all filings in the filings.xbrl.org repository, which can be used to verify consistency of rule implementations.
In addition, a reference implementation of the rules is also available. The reference implementation makes use of XULE, an open rules language developed by XBRL US.
There are a number of options for software vendors implementing DQR checks in software:
- Use the rule specifications and conformance suite tests to develop an implementation of the DQR ruleset;
- Embed an existing XULE processor and use the reference implementation of the rules; or
- Implement your own XULE processor and use the reference implementation of the rules.
These are described in a bit more detail below.
Develop your own rule implementations
If you use an XBRL Processor that provides API access to an XBRL model then implementing the DQR rules against that API may be relatively straightforward. The conformance suite and expected rule results can be used to verify the implementation, and the reference implementation may also be of assistance in understanding the required behaviour.
One advantage of this approach is that you may be able to provide a richer, more interactive user interface for investigating any errors than would be possible using a separate rules processor. This option may also provide better performance as it can be run as part of an existing validation process, rather than using a separate processor for DQR evaluation.
The primary disadvantage of this approach is that as well as the upfront development effort, an ongoing effort will be required in order to incorporate updates to the DQR rules. DQR rules often consist of rule specifying expected validation behaviour, and a list of taxonomy elements for which the rule is applied or excluded. If pursuing this approach, it is strongly recommended that you do not hard-code these element lists.
Embed an existing XULE processor
We are aware of at least two implementations of XULE in existing processors (Altova and Arelle – please let us know if there are others).
By using an existing XULE processor, you will be able to execute the reference implementation of the DQR rules directly, and will be able to incorporate any updates to the rules with no additional development work.
The downside of this approach is that if you are not currently using a processor with XULE support there may be a performance overhead associated with running validation in a second processor, and a development overhead associated with supporting additional 3rd party software in your product.
Implement your own XULE processor
If you have your own XBRL processor, then it is possible to add support for processing XULE rules, so that the reference implementation can be run directly. Implementing a XULE processor is not trivial, but this approach allows you to leverage your existing processor development effort, and will allow you to incorporate future updates to the rules with no additional development work.
If you are considering this approach, please see below for more information about XULE.
More about XULE
XULE is an open rules language that has been developed by XBRL US in order to support DQC rule execution. It is not currently an XBRL International standard. XULE has a number of features that are specifically designed to support DQC/DQR rule execution which are not currently available in the XBRL International Formula Rules language. In particular, XULE makes it easier to write rules that traverse the relationships in XBRL taxonomies, and supports simultaneous access to both base and extension taxonomies, facilitating the checking of extension taxonomies. Details about the XULE language can be found here. At present there are no firm plans to create an XBRL International Standard based on XULE, but this topic is under review by the XBRL Standards Board.
Upcoming Certification for DQR
XBRL International will offer DQR Certification, targeted for Q3 2022. This certification framework will test that DQR evaluation is accurate and, like other XBRL International certification arrangements, will use an objective set of test mechanisms. Vendors will be able to incorporate DQR testing into their re-certification arrangements, or apply to carry out DQR certification separately. Pricing and testing arrangements will be announced in due course.
Upcoming Governance Arrangements for DQR
The initial DQR tests have been drawn from the well-tested set of rules that relate to IFRS filings that have been submitted to the SEC. These same rules form part of XBRL US’ DQC arrangements. XBRL International will continue to work with XBRL US in this area but will create a DQR-specific governance arrangement to ensure that vendors, service providers, data providers, standards setters, regulators, analysts and other interested stakeholders can collaborate in the creation and testing of relevant and helpful data quality tests in this field. Organisations that are interested in participating can get in touch via email here.
Throughout 2022, the DQR tests will be labelled as “Beta”, to provide time for vendors, service providers, and issuers to get used to the idea of voluntary testing arrangements that will improve data quality. Vendors are strongly encouraged to ensure that they offer DQR support as soon as possible, to help ensure the consistency and quality of XBRL filings and certainly by the end of 2022.