Skip to content

Scraper

Scraping functions.

ScraperObject

Bases: object

Work In Progress. Do Not Use

A scraper that will parse sections and information from the retrieved files.

Attributes:

Name Type Description
headers

Request headers to avoid bot detection for scraping
Source code in sec_web_scraper/Scraper.py 
274
275
276
277
278
279
280
281
282
283
284
285
286
287
class ScraperObject(object):
    """Work In Progress. Do Not Use

    A scraper that will parse sections and information from the retrieved files.

    Attributes:
        headers: Request headers to avoid bot detection for scraping
    """

    def __init__(self):
        self.headers = {
            "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:94.0) Gecko/20100101 Firefox",
            "Accept": "application/json, text/javascript, */*; q=0.01",
        }

bs4_scraping_text(string_inp)

A BeautifulSoup wrapper function for processing the text document we retrieved to utilize the lxml parser.

Work in Progress.

Parameters:

Name Type Description Default
string_inp str

An HTML txt file (the output of get_document_given_link)

 required

Returns:

Type Description
BeautifulSoup

A BeautifulSoup object
Source code in sec_web_scraper/Scraper.py 
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
def bs4_scraping_text(string_inp: str) -> BeautifulSoup:
    """A BeautifulSoup wrapper function for processing the text document we retrieved
    to utilize the lxml parser.

    Work in Progress.

    Args:
        string_inp: An HTML txt file (the output of get_document_given_link)

    Returns:
        A BeautifulSoup object
    """

    text = BeautifulSoup(string_inp, "lxml")
    return text

create_selenium_browser_headless(sec_link='https://www.sec.gov/edgar/search/')

Creates a Selenium Headless Web Browser for Full Text Search.

The goal of this method is to perform full text search queries by using SEC EDGAR's full text page. There used to be a public API for this but it has been removed.

Parameters:

Name Type Description Default
sec_link str

The initial link for the Selenium Web Browser

 'https://www.sec.gov/edgar/search/'

Returns:

Type Description
webdriver.chrome.webdriver.WebDriver

None

Raises:

Type Description
ConnectionError

requests couldn't get your link so Selenium browser not created
Source code in sec_web_scraper/Scraper.py 
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
def create_selenium_browser_headless(
    sec_link: str = 'https://www.sec.gov/edgar/search/',
) -> webdriver.chrome.webdriver.WebDriver:
    """Creates a Selenium Headless Web Browser for Full Text Search.

    The goal of this method is to perform full text search queries by using SEC EDGAR's full text page.
    There used to be a public API for this but it has been removed.

    Args:
        sec_link: The initial link for the Selenium Web Browser

    Returns:
        None

    Raises:
        ConnectionError: requests couldn\'t get your link so Selenium browser not created
    """
    r = requests.get(sec_link, headers)
    if r.ok:
        print("Good")
        chrome_options = Options()
        chrome_options.add_argument("--headless=new")
        service = Service(ChromeDriverManager().install())

        driver = webdriver.Chrome(service=service, options=chrome_options)
        # driver = webdriver.Chrome(service=service)
        return driver
    else:
        raise ConnectionError('requests couldn\'t get your link so Selenium browser not created')

get_company_filings_given_cik(cik)

Find a company submission history given CIK.

This method will look at the SEC EDGARs official submission history for a company based on the provided CIK.It will then get the JSON document containing this information.

Parameters:

Name Type Description Default
cik str

A 10 digit unique string representing each public company.Can be retrieved using Downloader get_company_info.

 required

Returns:

Type Description
dict

A dict representing all the submission history for a partiular company (cik).
dict

The dict contains keys such as entityType,tickers,exchanges,stateofIncoporation,addresses, etc
dict

Empty dict is returned in the case that the CIK is invalid.

Raises:

Type Description
AssertionError

Length of CIKs must be 10.
Please see this link for more information

https://data.sec.gov/submissions/CIK0000320193.json
Source code in sec_web_scraper/Scraper.py 
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
def get_company_filings_given_cik(cik: str) -> dict:
    """Find a company submission history given CIK.

     This method will look at the SEC EDGARs official submission history for a company
     based on the provided CIK.It will then get the JSON document containing this information.

    Args:
        cik: A 10 digit unique string representing each public company.Can be retrieved using Downloader
            get_company_info.

    Returns:
        A dict representing all the submission history for a partiular company (cik).
        The dict contains keys such as entityType,tickers,exchanges,stateofIncoporation,addresses, etc

        Empty dict is returned in the case that the CIK is invalid.
    Raises:
        AssertionError: Length of CIKs must be 10.

        Please see this link for more information: https://data.sec.gov/submissions/CIK0000320193.json
    """
    assert len(cik) == 10
    link = f"https://data.sec.gov/submissions/CIK{cik}.json"
    r = requests.get(link, headers=headers)
    if r.ok:
        company_filling_cik = json.loads(r.text)
        print(company_filling_cik["sicDescription"])

        return company_filling_cik
    else:
        return {}

Retrieve a document given it's URL

Get the raw text of a document provided its URL

Parameters:

Name Type Description Default
link str

A url string for .txt file found on a documents's index page. For example, https://www.sec.gov/Archives/edgar/data/20/0000893220-96-000500.txt that can be retrieved from the index page: https://www.sec.gov/Archives/edgar/data/20/0000893220-96-000500-index.html

 required

Returns:

Type Description
str

None if document doesn't exist or a text str contaning the text
Source code in sec_web_scraper/Scraper.py 
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
def get_document_given_link(link: str) -> str:
    """Retrieve a document given it's URL

    Get the raw text of a document provided its URL

    Args:
        link: A url string for .txt file found on a documents's index page.
            For example, https://www.sec.gov/Archives/edgar/data/20/0000893220-96-000500.txt that can be retrieved
            from the index page: https://www.sec.gov/Archives/edgar/data/20/0000893220-96-000500-index.html
    Returns:
        None if document doesn't exist or a text str contaning the text
    """

    print(link)
    print(headers)
    r = requests.get(link, headers=headers)
    if r.ok:
        print("ok")
        return r.text
    else:
        return None

get_document_tags(txt)

Find all document tags inside of a document.

The document tags in a document are very helfpul for identifying change of sections in a document.

Parameters:

Name Type Description Default
txt str

An HTML txt file (the output of get_document_given_link)

 required

Returns:

Type Description
list[tuple]

A list of tuples of the form : (start_tag_index,end_tag_index,tag_name).
list[tuple]

One example: (1385 , 176135 , 10-K) means the 10-K section start at index 1385 and
list[tuple]

ended at 176135.
Source code in sec_web_scraper/Scraper.py 
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
def get_document_tags(txt: str) -> list[tuple]:
    """Find all document tags inside of a document.

    The document tags in a document are very helfpul for identifying change of sections
    in a document.
    Args:
        txt: An HTML txt file (the output of get_document_given_link)

    Returns:
        A list of tuples of the form : (start_tag_index,end_tag_index,tag_name).
        One example: (1385 , 176135 , <TYPE>10-K) means the 10-K section start at index 1385 and
        ended at 176135.
    """
    try:
        doc_start = re.compile(r"<DOCUMENT>")
        doc_end = re.compile(r"</DOCUMENT>")
        doc_type = re.compile(r"<TYPE>[^\n]+")

        beg_seq = []
        for y in doc_start.finditer(txt):
            beg_seq.append(y.end())
        end_seq = []
        for y in doc_end.finditer(txt):
            end_seq.append(y.start())

        type_list = []
        for y in doc_type.findall(txt):
            type_list.append(y)

        results = []
        for (
            x,
            y,
            z,
        ) in zip(beg_seq, end_seq, type_list):
            results.append((x, y, z))
            print(f'This is x, y, z: {x} , {y} , {z}')
        return results
    except TypeError as t:
        print(f'Error : {t}')
        return None

get_filings_by_query(query, driver)

Find the first 100 submissions that contain the specific query This method will look at the SEC EDGARs official database and perform a full-text-search on the provided query. #WARNING: sometimes it may return an empty DataFrame

Parameters:

Name Type Description Default
query str

A string query/keyword that you are looking for.

 required
driver

A selenium web driver created by create_selenium_browser_headless.

 required

Returns:

Type Description
pd.DataFrame

A pandas DataFrame containing the columns: ['File Type','CIK','Filename','Date Filed', 'Company Name','File Link']
pd.DataFrame

See pandas.DataFrame to learn more about Pandas DataFrames
Source code in sec_web_scraper/Scraper.py 
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
def get_filings_by_query(query: str, driver) -> pd.DataFrame:
    """Find the first 100 submissions that contain the specific query
     This method will look at the SEC EDGARs official database and perform a full-text-search on the provided query.
     #WARNING: sometimes it may return an empty DataFrame

    Args:
        query:A string query/keyword that you are looking for.
        driver: A selenium web driver created by create_selenium_browser_headless.

    Returns:
        A pandas DataFrame containing the columns:
           ['File Type','CIK','Filename','Date Filed', 'Company Name','File Link']
        See [pandas.DataFrame][] to learn more about Pandas DataFrames

    Raises:
        An Exception if DataFrame can't be generated after 10 iterations of selenium get.

    """
    # service = Service(ChromeDriverManager().install())
    # chrome_options = Options()
    # chrome_options.add_argument("--headless=new")
    # driver = webdriver.Chrome(service=service,options=chrome_options)
    list_return = []
    found = False
    i = 0
    while found is False and i != 15:
        driver.get(f'https://www.sec.gov/edgar/search/#/q={query}')
        soup = BeautifulSoup(driver.page_source, 'html.parser')
        # search = soup.find_all('tr')
        search = soup.find(id='hits')
        new_search = search.find_all('tr')[1:]
        link_ = 'https://www.sec.gov/Archives/edgar/data/'
        column_names = ['File Type', 'CIK', 'Filename', 'Date Filed', 'Company Name', 'File Link']
        for ele in new_search:
            ele_attr = ele.find_all("td")
            # File name :
            file_type = ele_attr[0].text
            submission_num = ele_attr[0].a['data-adsh'].replace('-', '')
            file_name = ele_attr[0].a['data-file-name']
            cik = ele_attr[4].text.split()[1]
            file_link = f'{link_}/{cik}/{submission_num}/{file_name}'
            company_name = ele_attr[3].text
            date_filed = ele_attr[1].text
            list_return.append([file_type, cik, file_name, date_filed, company_name, file_link])
            # keep only file_name, file_type, cik, file_link,name
        driver.implicitly_wait(5)
        found = len(list_return) != 0
        i += 1

    # if i == 10:
    #    raise Exception("Tried 10 times, could not generate DataFrame")
    return pd.DataFrame(list_return, columns=column_names)

iterate_over_filings(filings)

Extract number of filings for a company.

This function will iterate over the filings dictionary and try to extract relevant information for the user. As of now, it only returns the total count of filings.

Parameters:

Name Type Description Default
filings dict

A dictionary retrieved from get_company_filings_given_cik. get_company_filings_given_cik will return a nested dictionary. One can pass in the "filings" key of the dictionary returned by get_company_filings_given_cik.

 required

Returns:

Type Description
int

A count of all the filings recorded for this company
Source code in sec_web_scraper/Scraper.py 
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
def iterate_over_filings(filings: dict) -> int:
    """Extract number of filings for a company.

    This function will iterate over the filings dictionary and try to extract
    relevant information for the user. As of now, it only returns the total count
    of filings.

    Args:
        filings: A dictionary retrieved from get_company_filings_given_cik.
            get_company_filings_given_cik will return a nested dictionary.
            One can pass in the "filings" key of the dictionary returned by get_company_filings_given_cik.

    Returns:
        A count of all the filings recorded for this company
    """
    print(filings.keys())
    for k, v in filings["recent"].items():
        print(f"This is the key {k} and the item length: {len(v)} and type : {type(v)}")
    for j in filings["files"]:
        print(j)
        print(type(j))
    print("------")
    return filings["files"][0]["filingCount"]