(stock ticker -> CIK) dictionary using SEC EDGAR

cik_dict.py

import re
from cPickle import dump
from requests import get

DEFAULT_TICKERS = ['goog', 'aapl']
URL = 'http://www.sec.gov/cgi-bin/browse-edgar?CIK={}&Find=Search&owner=exclude&action=getcompany'
CIK_RE = re.compile(r'.*CIK=(\d{10}).*')

cik_dict = {}
for ticker in DEFAULT_TICKERS:
    results = CIK_RE.findall(get(URL.format(ticker)).content)
    if len(results):
        cik_dict[str(ticker).lower()] = str(results[0])
f = open('cik_dict', 'w')
dump(cik_dict, f)
f.close()

Solid stuff Doug. I have been trying to figure out the best way to attack the trifecta: cik with ticker with company name. Thanks.

Doug your code just saved me a bunch of time! Thanks man

Curious if anyone has tried to port this into Python3? Regardless, thank you Doug!

It works in python 3, just with a tiny change:
f = requests.get(URL.format(ticker), stream = True);
results = CIK_RE.findall(f.text)

Some tickers don't work tho: e.g., BRK.A (Buffett's company), anyone has a solution?

I want to convert this into a dataframe and it keeps spitting out the following

{'goog': '0001652044', 'aapl': '0000320193'}
0
goog 0001652044
aapl 0000320193

How do I rename my columns?
I tried df = df.rename(columns = {"0": "CIK"}) but that didn't work.

Here is the version of the code that I used:

`import re
from _pickle import dump
import requests
import pandas as pd

DEFAULT_TICKERS = ['goog', 'aapl']
URL = 'http://www.sec.gov/cgi-bin/browse-edgar?CIK={}&Find=Search&owner=exclude&action=getcompany'
CIK_RE = re.compile(r'.CIK=(\d{10}).')

cik_dict = {}
for ticker in DEFAULT_TICKERS:
f = requests.get(URL.format(ticker), stream=True);
results = CIK_RE.findall(f.text)
if len(results):
cik_dict[str(ticker).lower()] = str(results[0])

print(cik_dict)

df = pd.DataFrame.from_dict(cik_dict, orient = 'index')
df = df.rename(columns = {"0": "CIK"})
print(df)`

Thank you everyone, your comments and code was a huge help.
for 3.6:

import re, requests

def getCIKs(TICKERS):
    URL = 'http://www.sec.gov/cgi-bin/browse-edgar?CIK={}&Find=Search&owner=exclude&action=getcompany'
    CIK_RE = re.compile(r'.*CIK=(\d{10}).*')    
    cik_dict = {}
    for ticker in TICKERS:
        f = requests.get(URL.format(ticker), stream = True)
        results = CIK_RE.findall(f.text)
        if len(results):
            results[0] = int(re.sub('\.[0]*', '.', results[0]))
            cik_dict[str(ticker).upper()] = str(results[0])
    f = open('cik_dict', 'w')   
    print(cik_dict)
    f.close()

getCIKs(['wmt','amzn','nflx'])
# returns:
# {'WMT': '104169', 'AMZN': '1018724', 'NFLX': '1065280'}

@Micah81 instead of print(cik_dict), return(cik_dict) worked better for me

Today you have to set a user agent:

import re, requests
headers = {"user-agent": "Safari"}

def getCIKs(TICKERS):
    URL = 'http://www.sec.gov/cgi-bin/browse-edgar?CIK={}&Find=Search&owner=exclude&action=getcompany'
    CIK_RE = re.compile(r'.*CIK=(\d{10}).*')    
    cik_dict = {}
    for ticker in TICKERS:
        f = requests.get(URL.format(ticker),headers= headers, stream = True)
        results = CIK_RE.findall(f.text)
        if len(results):
            results[0] = int(re.sub('\.[0]*', '.', results[0]))
            cik_dict[str(ticker).upper()] = str(results[0])
    f = open('cik_dict', 'w')
    f.close()
    return(cik_dict)

getCIKs(['wmt','amzn','nflx'])
# returns:
# {'WMT': '104169', 'AMZN': '1018724', 'NFLX': '1065280'}

It's failed today because the SEC has already changed the website