UK filing iXBRL

Jan 2, 2012 at 7:11 AM

It is now mandatory to prepare & submit accounts in this format in the UK as the folloinwg item from MIKE2 reports.

"Inline XBRL has already been adopted in the UK. It is being used as a central part of the HM. Revenue & Customs (HMRC) mandate that requires UK companies to provide statutory accounts in iXBRL format, replacing the paper arrangements that have existed to date. Nearly 2 million companies will file their accounts to the UK's tax regulator in this format starting in April 2011. Companies will be able to file their accounts with the companies registrar (Companies House) from the summer of 2010. A range of other regulators are closely examining the iXBRL specifications and many are expected to follow suit."

Does Gepsio support iXBRL?

Jan 2, 2012 at 2:11 PM
Edited Jan 2, 2012 at 2:12 PM

Hello HughMeares,

I have not (yet) seen an iXBRL document, but I have tried to design Gepsio with iXBRL in mind. Specifically, an XbrlDocument contains a collection of XbrlFragment objects. There will be one XbrlFragment object for every XBRL markup item in the document you loaded. This allows for support of an XML document with multiple XBRL fragments in it, which is possible with an iXBRL document. The general pattern would be:

var doc = XbrlDocument.Load("iXBRLDoc.xml");
foreach(var fragment in doc.XbrlFragments)
{
    // process iXBRL fragment
}

If you find problems, please let me know. If you find problems with your iXBRL document, and would be able to share the document, please ZIP up the document and all supporting files (like schemas and linkbases) and post it so I can take a look. I would like to ensure that Gepsio does indeed support iXBRL.

Thanks for trying Gepsio!

Jeff Ferguson

Jan 2, 2012 at 5:41 PM

Dear Jeff

Thanks for coming back so quickly. I am ashamed to say that I did not even know about XBRL until the evening of 31 Dec when I was trying to file accounts with HMRC and bumped into this for the first time. HMRC declined to accept my pdf and I now have to pay a fine for late filing!

It might help if I explain that I trained as an accountant (with PW) but ended up in the software industry and am competent in VBA and tolerably competent in C and C#.

My initial feel is that iXBRL is something of a camel (on the lines that a camel has been described as a horse designed by a committee). The whole point of submitted accounts is that presentation is king and for that you want something as precise as pdf so that the presentation is independent of the rendering agent. The US concept of an XBRL structured data representation makes perfect sense as a complement to a properly designed layout presentation. I

n contrast, the iXBRL documents I have seen result in nasty looking accounts which is hardly surprising as xml/html is not what you would choose for precision layout. So why did the iXBRL folk not just work out a way of including the XBRL data content as a separate section at the end of the presentation content in a way that the rendering agent would treat this as metadata not for display.

Preparation of such documents would then have been quite easy instead of the significant problem that we now have as we have to convert our word and excel documents into inline XBRL in a way that meets the HMRC standard. There are commercial solutions which will do this for you but they all cost a ridiculous amount of money and add another layer of complication. Products like CorelFiling are now tied into the accounting majors and accountancy software firms.

Perhaps I am wrong and separating out all the XBRL stuff into one section of the xml file could be valid solution acceptable to the standard and HMRC? Or are my first thoughts right and the inline XBRL has to be adjacent to the xml fragments which contain the matching natural language content.

I attach some sample files from the HMRC Corporation Tax site for your perusal. When you file the CT return, HMRC require you to submit accounts in iXBRL format.

I have not had time yet to download the gepsio material from the codeplex site but it would be useful to know what other operations one could perform using powershell script along the lines of the example you give.

Many thanks and I shall follow this with interest

Hugh Meares

From: jeffreyferguson [email removed]
Sent: 02 January 2012 14:12
To: hugh_meares@msn.com
Subject: Re: UK filing iXBRL [gepsio:284709]

From: jeffreyferguson

Hello HughMeares,

Although I have not (yet) seen an iXBRL document, but I have tried to design Gepsio with iXBRL in mind. Specifically, an XbrlDocument contains a collection of XbrlFragment objects. There will be one XbrlFragment object for every XBRL markup item in the document you loaded. This allows for support of an XML document with multiple XBRL fragments in it, which is possible with an iXBRL document. The general pattern would be:

var doc = XbrlDocument.Load("iXBRLDoc.xml");
foreach(var fragment in doc.XbrlFragments)
{
    // process iXBRL fragment
}
 

If you find problems, please let me know. If you find problems with your iXBRL document, and would be able to share the document, please ZIP up the document and all supporting files (like schemas and linkbases) and post it so I can take a look. I would like to ensure that Gepsio does indeed support iXBRL.

Thanks for trying Gepsio!

Jeff Ferguson

Jan 4, 2012 at 1:31 PM

Hello again, Hugh,

I emailed you privately, but hadn't heard back, so I thought that I would repost my email here.

Hello again, Hugh,

In a post on Gepsio's Codeplex site, you mentioned that you had attached some iXBRL samples from the HMRC Corporation Tax Site. The attachments were somehow lost in the posting to the Codeplex site, so I thought that I would email you directly and ask if you would be so kind as to resend them here, through direct email.

I'd be happy to author some additional Powershell samples for you. Powershell is a nice expedient front-end for Gepsio when quick perusal of XBRL data is required. Do you have any use cases in mind? If you do, let me know what you think a use case might be ("find all facts with a value over 1,000") and I would be happy to post additional samples.

Thanks for taking a look at Gepsio,

Jeff Ferguson

Jan 6, 2012 at 12:37 AM

RESEND

From: Hugh Meares [email removed]
Sent: 06 January 2012 00:34
To: 'gepsio@discussions.codeplex.com'; 'notifications@codeplex.com'
Subject: UK filing iXBRL [gepsio:284709]

3 no files should be attached

Jan 6, 2012 at 1:27 AM
Hello Hugh,
The attachments aren't coming through; however, if you could email them directly to me at jeffrey.ferguson@gmail.com, I would be happy to take a look.
Thanks,
Jeff Ferguson
Nov 21, 2014 at 2:56 PM
Edited Nov 21, 2014 at 2:57 PM
Hi Jeff

Just downloaded Gepsio to process some XBRL files and it worked perfectly. Thanks for that.

My next task is to read the XBLR in an "Inline XBRL" document (a .HTML file) along the lines discussed above. I have some sample files from a UK Gov site which I can send you as good valid samples. But, when I load these files into Gepsio, no exceptions are reported, but also there are no fragments either, indicating that there is no XBRL in the document. Looking at the files in notepad I can see the Inline XBRL, and the documents look good to the eye.
  • Does Gepsio currently support Inline XBRL?
  • Do you have plans to support it?
  • Would you like my sample files to test against? (I can email a few, they are not large).
Thanks in anticipation.
Ian
Nov 21, 2014 at 3:16 PM
I don't believe that Gepsio currently supports iXBRL, but I would love to add that support. I believe the difference between XBRL and iXBRL is how to discover the XBRL data islands in an iXBRL instance. If you could send some iXBRL samples to gepsio@outlook.com, I'd love to take a look. Thank you!
Nov 21, 2014 at 4:14 PM
Thanks for the quick response.

Email sent with sample documents.

Regards
Ian