Is it possible to parse the GAAP standards taxonomy?

[scottcarlson]

Hi @manusimidt. Thanks for such a great library. I've really enjoyed digging into the code.

I'm new to XBRL, but reading through your answers to #21 and #109 I was quickly able to print out a hierarchy from the presentation linkbases when I provided an iXBRL schema URL for a company.

However, what I'm really after is creating a presentation linkbase hierarchy from the GAAP standards themselves, unrelated to a specific company. Which I believe would essentially be parsing a https://xbrl.fasb.org/us-gaap/2022/elts/us-gaap-2022.xsd schema(?)

I'm trying to create a similar https://xbrlview.fasb.org schema output which could be used to understand the parent relationship of a child concept. For example, how OilAndGasRevenue is nested under the RevenuesAbstract as shown here:

Ultimately I'm trying to create something similar to http://www.xbrlsite.com/2018/Prototype/ReportingStylesUSGAAP/COMID-BSU-CF2-IS5-IEMIB-OILN/mapping-definition.html which could be used to reference which fundamental abstract concepts a child GAAP concept belongs to.

I am currently able to create a hierarchy of the presentation linkbases with py-xbrl on company specific documents but it does not seem to provide the entire tree for most concepts. Ex: RevenueFromContractWithCustomerExcludingAssessedTax not showing as a child of RevenuesAbstract:

The full hierarchy tree shown from the https://xbrlview.fasb.org/ screenshot above is what I'm trying to achieve.

After reading through #103, I'm unsure if this is the proper library for such a task, but wanted to start a discussion to understand if this is something that would be easy to accomplish with py-xbrl, or if it's even possible with this library?

[scottcarlson]

I was finally able to find time to return to this project and I realized I completely missed the TaxonomySchema functionality. With that I was able to create what I needed.

Here's a sample code snippet in case anyone else comes across this:

from xbrl.cache import HttpCache
from xbrl.taxonomy import TaxonomySchema, parse_taxonomy_url
from xbrl.linkbase import PresentationArc

cache: HttpCache = HttpCache('./.cache')

schema_paths = [
  # TODO: Dynamically fetch all root XSD files from `stm` and `dis` directories
  # at https://xbrl.fasb.org/us-gaap/2022/ and other years of choice
  "https://xbrl.fasb.org/us-gaap/2022/stm/us-gaap-stm-soc-2022.xsd",
  "https://xbrl.fasb.org/us-gaap/2022/dis/us-gaap-dis-rcc-2022.xsd",
  # ...
]

def print_presentation_arc(level: int, arc: PresentationArc):
  print(f"{'  ' * level}{arc.to_locator.concept_id}")
  for child_arc in arc.to_locator.children:
    print_presentation_arc(level + 1, child_arc)

for schema_path in schema_paths:
  taxonomy: TaxonomySchema = parse_taxonomy_url(schema_path, cache)

  for pre_linkbase in taxonomy.pre_linkbases:
      for elr in pre_linkbase.extended_links:
        print(f"\n\n")
        print(elr.elr_id.split('#')[1])
        # if the elr is empty, skip it
        if len(elr.root_locators) == 0: continue
        # print abstract
        print(f"  {elr.root_locators[0].concept_id}")
        for root_locator in elr.root_locators:
          for pre_arc in root_locator.children:
            print_presentation_arc(2, pre_arc)

Obviously change your output style/file as needed, this is just an example.

Thanks again for the great library!

[manusimidt]

Hey @scottcarlson,

sorry for my late answer, for some reason I did not get a notification for the question you asked a week ago.
But yes, the code you provided should work. In theory, you should also be able to parse the entire taxonomy using the endpoint us-gaap-entryPoint-all-2022.xsd by recursively looping over all taxonomies and printing out their presentation linkbases using the code you provided.

Thanks for the feedback, always nice to hear that py-xbrl is helping other people :)