Repeating facts from different documents

[Timur Mirzaev]

Some facts are repeating when extracting several documents in a single call. I can see the reason (tag/concept change), but don’t know the solution. Is there any way to remove those duplicating values, but keep the two tags/concepts in a query?

Here is my API call with row 2 and 3 being similar:

XBRL.showData(""&fact&"&entity.cik="&CIK&fiscal&"3Q&concept.local-name="&"IncomeLossFromContinuingOperationsBeforeIncomeTaxesExtraordinaryItemsNoncontrollingInterest,IncomeLossFromContinuingOperationsBeforeIncomeTaxesMinorityInterestAndIncomeLossFromEquityMethodInvestments"&","&Legend!C6&"&fact.has-dimensions=false&fact.ultimus=true&report.document-index=1,2,3,4&fields=report.period-end,period.fiscal-year.sort(DESC),concept.local-name,fact.numerical-value,period.fiscal-period.sort(DESC)","","","1")

[Timur Mirzaev]

Just a little update: Missed part of my query that refers to a named range “CIK”.
In this very case it should stand for “0000789019” (for MSFT).

[David Tauriello]

Hi Timur – thanks for writing. I’m not sure what additional parameter you’re using from C6 on the Legend worksheet (?). If I remove that and update ‘fiscal’ to be the parameter ‘period.fiscal-period’ I don’t see any duplication:

https://api.xbrl.us/api/v1/fact/search?entity.cik=0000789019&period.fiscal-period=3Q&concept.local-name=IncomeLossFromContinuingOperationsBeforeIncomeTaxesExtraordinaryItemsNoncontrollingInterest,IncomeLossFromContinuingOperationsBeforeIncomeTaxesMinorityInterestAndIncomeLossFromEquityMethodInvestments&fact.has-dimensions=false&fact.ultimus=true&report.document-index=1,2,3,4&fields=report.period-end,period.fiscal-year.sort(DESC),concept.local-name,fact.numerical-value,period.fiscal-period.sort(DESC)

Drop the query in A1, and use =XBRL.showData(A1,””,””,”1″) in A2

Let me know if there’s something else I’m missing.

[Timur Mirzaev]

Hi, David!

Thank you very much for your reply. I forgot to fully correct my query for named ranges, hence the error.
But your correction is exactly what my query should have looked like. I have done exactly what you showed above and I DO get duplicating values – see D4 and D5 of the output (if xbrl.showData is in A2), which is $17,236,000,000 for 3Q of FY’2021.

So my question here is whether it is possible to introduce any parameter that would remove those duplicates (sort of a “unique” function maybe, but on the query level so that the per query limit is not effected that much).

Regards,
Timur

[David Tauriello]

Hi Timur – these are different facts appearing in different reports. Look at the full concept.local-name for rows 4 and 5. Further, if you add report.sec-url to the fields to be returned, you can browse to the EDGAR entries and confirm the values are different.

IncomeLossFromContinuingOperationsBeforeIncomeTaxesMinorityInterestAndIncomeLossFromEquityMethodInvestments – https://www.sec.gov/Archives/edgar/data/789019/000156459021020891/0001564590-21-020891-index.htm

IncomeLossFromContinuingOperationsBeforeIncomeTaxesExtraordinaryItemsNoncontrollingInterest – https://www.sec.gov/Archives/edgar/data/789019/000156459022015675/0001564590-22-015675-index.htm

When you review the face of the report these fact values do have the same label – https://api.xbrl.us/api/v1/label/search?dts.id=456408,583061&concept.id=27969523,31293140&fields=concept.id,concept.local-name,label.role-short,label.text – it may be that the company’s external reporting team reconsidered the definition it had been using, and decided to transition to a new concept.

The XBRL API does not currently normalize facts with a mechanism as you describe that would evaluate and eliminate facts with duplicated labels like this.

[Author]

Posts