A small library to access files from SEC's edgar.

pip install edgar

To get a company's latest 5 10-Ks, run

from edgar import Company
company = Company("Oracle Corp", "0001341439")
tree = company.get_all_filings(filing_type = "10-K")
docs = Company.get_documents(tree, no_of_documents=5)

or

from edgar import Company, TXTML

company = Company("INTERNATIONAL BUSINESS MACHINES CORP", "0000051143")
doc = company.get_10K()
text = TXTML.parse_full_10K(doc)

To get all companies and find a specific one, run

from edgar import Edgar
edgar = Edgar()
possible_companies = edgar.find_company_name("Cisco System")

To get XBRL data, run

from edgar import Company, XBRL, XBRLElement

company = Company("Oracle Corp", "0001341439")
results = company.get_data_files_from_10K("EX-101.INS", isxml=True)
xbrl = XBRL(results[0])
XBRLElement(xbrl.relevant_children_parsed[15]).to_dict() // returns a dictionary of name, value, and schemaRef

The Company class has two fields:

• name (company name)
• cik (company CIK number)
• timeout (optional) (default: 10)

Returns a url to fetch filings data

• Input
  • filing_type: The type of document you want. i.e. 10-K, S-8, 8-K. If not specified, it'll return all documents
  • prior_to: Time prior which documents are to be retrieved. If not specified, it'll return all documents
  • ownership: defaults to include. Options are include, exclude, only.
  • no_of_entries: defaults to 100. Returns the number of entries to be returned. Maximum is 100.

Returns the HTML in the form of lxml.html

• Input
  • filing_type: The type of document you want. i.e. 10-K, S-8, 8-K. If not specified, it'll return all documents
  • prior_to: Time prior which documents are to be retrieved. If not specified, it'll return all documents
  • ownership: defaults to include. Options are include, exclude, only.
  • no_of_entries: defaults to 100. Returns the number of entries to be returned. Maximum is 100.

Returns the HTML in the form of lxml.html of concatenation of all the documents in the 10-K

• Input
  • no_of_documents (default: 1): numer of documents to be retrieved

Returns the HTML in the form of lxml.html of the document within 10-K

• Input
  • document_type: Tye type of document you want, i.e. 10-K, EX-3.2
  • no_of_documents (default: 1): numer of documents to be retrieved

Returns the HTML in the form of lxml.html of the data file within 10-K

• Input
  • document_type: Tye type of document you want, i.e. EX-101.INS
  • no_of_documents (default: 1): numer of documents to be retrieved
  • isxml (default: False): by default, things aren't case sensitive and is parsed with html in lxml. If this is True, then it is parsed with etree` which is case sensitive

Returns a list of strings, each string contains the body of the specified document from input

• Input
  • tree: lxml.html form that is returned from Company.getAllFilings
  • no_of_documents: number of document returned. If it is 1, the returned result is just one string, instead of a list of strings. Defaults to 1.
  • debug (default: False): if True, displays the URL and form

Gets all companies from EDGAR

• Input
  • name: name of the company

• Input
  • cik: cik of the company

• Input
  • words: input words to search the company

Parses data from XBRL

• relevant_children
  • get children that are not context
• relevant_children_parsed
  • get children that are not context, unit, schemaRef
  • cleans tags