The world's easiest, most powerful edgar library

edgartools is one of the nicest looking EDGAR libraries out there. It is also powerful and easy to use. You can query, filter and select any filing since 1994 and view the filing's html, text, xml or structured data.

Install using pip

pip install edgartools

Import and start using

from edgar import *

# Tell the SEC who you are
set_identity("Michael Mccallum [email protected]")

filings = get_filings()

Depends on what you know

filing = find("0001065280-23-000273")

filings = Company("NFLX").get_filings(form="10-Q").latest(1)

filings = get_filings(form="10-Q")
filing = filings[0]

You can view it in the terminal or open it in the browser, get the filing as html, xml or text, and download attachments. You can extract data from the filing into a data object.

You can get the company's filings, facts and financials.

pip install edgartools

Before you can access the SEC Edgar API you need to set the identity that you will use to access Edgar. This is usually your name and email, or a company name and email.

Sample Company Name AdminContact@<sample company domain>.com

The user identity is sent in the User-Agent string and the Edgar API will refuse to respond to your request without it.

EdgarTools will look for an environment variable called EDGAR_IDENTITY and use that in each request. So, you need to set this environment variable before using it.

export EDGAR_IDENTITY="Michael Mccallum [email protected]"

 $Env:EDGAR_IDENTITY="Michael Mccallum [email protected]"

Alternatively, you can call set_identity which does the same thing.

from edgar import set_identity
set_identity("Michael Mccallum [email protected]")

For more detail see https://www.sec.gov/os/accessing-edgar-data

from edgar import *

# Get filings for the current year and quarter
filings = get_filings() 

# Get filings for 2022
filings = get_filings(2022)                 # OR filings = get_filings(year=2022)

# Get filings for 2022 quarter 4
filings = get_filings(2022, 4)              # OR filings = get_filings(year=2022, quarter=4)

# Get filings for 2020, 2021 and 2022
filings = get_filings([2020, 2021, 2022])   # OR filings = get_filings(year=range(2020, 2023))

# Get filings for 2020 quarters 1 and 2
filings = get_filings(2020, quarter=[1,2])

# Filter for form D
filings.filter(form="D")

# Filter for form 10-K and 10-Q 
filings.filter(form=["10-K", "10-Q"])

# When you filter by form e.g. "D" it includes amendments e.g. "D\A". You can omit amendments
filings.filter(form="D", amendments=False)

# Filter by filing_date. date and filing_date mean the same thing
# Get all filings on 2023-02-23
filings.filter(date="2023-02-23")                      
# OR
filings.filter(filing_date="2023-02-23")

# Filter to get all filings between 2023-01-23 and 2023-02-23     
filings.filter(date="2023-01-23:2023-02-23")

# Filter to get all filings since 2023-01-23   
filings.filter(date="2023-01-23")

# Filter to get all filings before 2023-02-23     
filings.filter(date=":2023-02-23")

get_filings(2022, form="D")

The filings data is stored in the Filings class as a pyarrow.Table. You can get the data as a pandas dataframe using to_pandas

df = filings.to_pandas()

The Filings object allows you to navigate through filings using filings.next() and filings.prev(). This shows you pages of the data - the page size is about 50.

# To see the next page of data
filings.next()

# To see the previous page
filings.prev()

# To see the current page
filings.current()

You can get the latest n filings by filing_date from a filings using filings.latest().

If you provide the parameter n it will return the latest n filings.

filing = filings.latest(n=5)
filing

If you omit this parameter, or set n=1 it will return a single `Filings object.

filing = filings.latest()
filing

You can filter the filings object using te filter() function. This allows you to filter by filing date, or by form.

To filter by filing date specify the filing date in YYYY-MM-DD format e.g. 2022-01-24 (Note the parameters date and filing_date are equivalent aliases for each other)

filings.filter(date="2021-01-24") # or filings.filter(filing_date="2021-01-24")

You can specify a filing date range using the colon

filings.filter(date="2021-01-12:2021-02-28") 

To filter by dates before a specified date use `:YYYY-MM-DD'

filings.filter(date=":2021-02-28") 

To filter by dates after a specified date use `YYYY-MM-DD:'

filings.filter(date="2021-02-28:") 

You can filter filings by form using the form parameter.

filings.filter(form="10-K") 

To filter by form e.g. 10-K and include form amendments use amendments = True.

filings.filter(form="10-K", amendments=True) 

You can get a single filing from the filings using the bracket operator [], specifying the index of the filing. The index is the value displayed in the leftmost position in the filings table. For example, to get the 10-Q for Costco in the table above use filings[3]

filing = filings[3]

You can view the filing homepage in the terminal using filing.homepage

This gives you access to the FilingHomepage class that you can use to list all the documents and datafiles on the filing.

filing.homepage

You can open the filing in your browser using filing.open(). This will work on environments with access to the browser, will probably not work on a remote server.

filing.open()

You can open the filing homepage in the browser using filing.homepage.open().

filing.homepage.open()

You can view the filing's HTML content as markdown in the console using view(). It works for all filing types but can be a little slow for filings with large HTML files

filing.view()

You can get the html content of the filing using.html()

filing.html()

You can get the html content as markdown using.markdown()

filing.markdown()

Every filing has a list of attachments. You can view the attachments using filing.attachments

# View the attachments
filing.attachments

You can access each attachment using the bracket operator [] and the index of the attachment.

# Get the first attachment
attachment = filing.attachments[0]

You can download the attachment using attachment.download(). This will download the attachment to string or bytes in memory.

Now the reason you may want to download attachments is to get information contained in data files. For example, 13F-HR filings have attached infotable.xml files containing data from the holding report for that filing.

Fortunately, the library handles this for you. If you call filing.obj() it will automatically download and parse the data files into a data object, for several different form types. Currently, the following forms are supported:

For example, to get the data object for a 13F-HR filing you can do the following:

filings = get_filings(form="13F-HR")
filing = filings[0]
thirteenf = filing.obj()

If you call obj() on a filing that does not have a data file, then it will return None.

Some filings are in XBRL (eXtensible Business Markup Language) format. These are mainly the newer filings, as the SEC has started requiring this for newer filings.

If a filing is in XBRL format then it opens up a lot more ways to get structured data about that specific filing and also about the company referred to in that filing.

The Filing class has an xbrl function that will download, parse and structure the filing's XBRL document if one exists. If it does not exist, then filing.xbrl() will return None.

The function filing.xbrl() returns a FilingXbrl instance, which wraps the data, and provides convenient ways of working with the xbrl data.

filing_xbrl = filing.xbrl()

With the company API you find a company using the cik or ticker. From the company you can access all their historical filings, and a dataset of the company facts. The SEC's company API also supplies a lot more details about a company including industry, the SEC filer type, the mailing and business address and much more.

The cik is the id that uniquely identifies a company at the SEC. It is a number, but is sometimes shown in SEC Edgar resources as a string padded with leading zero's. For the edgar client API, just use the numbers and omit the leading zeroes.

company = Company(1324424)

You can get a company using a ticker e.g. SNOW. This will do a lookup for the company cik using the ticker, then load the company using the cik. This makes it two calls versus one for the cik company lookup, but is sometimes more convenient since tickers are easier to remember that ciks.

Note that some companies have multiple tickers, so you technically cannot get SEC filings for a ticker. You instead get the SEC filings for the company to which the ticker belongs.

The ticker is case-insensitive so you can use Company("snow") or Company("SNOW")

snow = Company("snow")

Company(1832950)

To get the company's filings use get_filings(). This gets all the company's filings that are available from the Edgar submissions endpoint.

company.get_filings()

You can filter the company filings using a number of different parameters.

class CompanyFilings:
    
    ...
    
    def get_filings(self,
                    *,
                    form: str | List = None,
                    accession_number: str | List = None,
                    file_number: str | List = None,
                    is_xbrl: bool = None,
                    is_inline_xbrl: bool = None
                    ):
        """
        Get the company's filings and optionally filter by multiple criteria
        :param form: The form as a string e.g. '10-K' or List of strings ['10-Q', '10-K']
        :param accession_number: The accession number that uniquely identifies an SEC filing e.g. 0001640147-22-000100
        :param file_number: The file number e.g. 001-39504
        :param is_xbrl: Whether the filing is xbrl
        :param is_inline_xbrl: Whether the filing is inline_xbrl
        :return: The CompanyFiling instance with the filings that match the filters
        """

The result of get_filings() is a CompanyFilings class. This class contains a pyarrow table with the filings and provides convenient functions for working with filings. You can access the underlying pyarrow Table using the .data property

filings = company.get_filings()

# Get the underlying Table
data: pa.Table = filings.data

To access a filing in the CompanyFilings use the bracket [] notation e.g. filings[2]

filings[2]

The CompanyFilings class has a latest function that will return the latest Filing. So, to get the latest 10-Q filing, you do the following

# Latest filing makes sense if you filter by form  type e.g. 10-Q
snow_10Qs = snow.get_filings(form='10-Q')
latest_10Q = snow_10Qs.latest()

# Or chain the function calls
snow.get_filings(form='10-Q').latest()

Facts are an interesting and important dataset about a company accumlated from data the company provides to the SEC. Company facts are available for a company on the Company Factsf"https://data.sec.gov/api/xbrl/companyfacts/CIK{cik:010}.json" It is a JSON endpoint and edgartools parses the JSON into a structured dataset - a pyarrow.Table.

To get company facts, first get the company, then call company.get_facts()

company = Company("SNOW")
company_facts = company.get_facts()

The result is a CompanyFacts object which wraps the underlying facts and provides convenient ways of working with the facts data. To get access to the underyling data use the facts property.

You can get the facts as a pandas dataframe by calling to_pandas

df = company_facts.to_pandas()

Facts differ among companies. To see what facts are available you can use the facts_meta property.

Contributions are welcome! We would love to hear your thoughts on how this library could be better at working with SEC Edgar.

We use GitHub issues to track public bugs. Report a bug by opening a new issue; it's that easy!

• Fork the repo and create your branch from master.
• If you've added code that should be tested, add tests.
• If you've changed APIs, update the documentation.
• Ensure the test suite passes.
• Make sure your code lints.
• Issue that pull request!

edgartools is distributed under the terms of the MIT license.

LinkedIn

2 3 4