Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add rudimentary keyword search for CCPA policies using HTML file #5

Open
wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Add rudimentary keyword search for CCPA policies using HTML file #5

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
6b14803
Rudimentary working
objorkman Apr 19, 2022
8fcbd7d
Address comments, add rest of categories, csv output
objorkman Apr 27, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
17 changes: 17 additions & 0 deletions polipy/constants.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,3 +2,20 @@

UTC_DATE = datetime.datetime.utcnow().strftime('%Y%m%d')
CWD = os.getcwd()

KEYWORDS = {
'identifiers': [
'real name', 'alias', 'postal address', 'address', 'unique personal identifier',
'online identifier', 'IP address', 'email address', 'email', 'account name',
'social security number', 'driver license number', 'passport number'
],
'customer records information': [
'name', 'signature', 'social security number', 'ssn',
'physical characteristics', 'address', 'telephone number', 'phone number',
'passport number', 'drivers license',
'state identification card number', 'insurance policy number',
'education', 'employment', 'employment history', 'bank account number',
'credit card number', 'debit card number', 'financial information',
'medical information', 'health insurance information'
]
}
26 changes: 24 additions & 2 deletions polipy/extractors.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,7 @@
from bs4 import BeautifulSoup
from io import BytesIO, StringIO
from pdfminer.high_level import extract_text as parse_pdf
from .constants import KEYWORDS

extractors = [
'text'
Expand All @@ -11,8 +12,10 @@ def extract(extractor, **kwargs):
content = extract_text(**kwargs)
return content

def extract_text(url_type, url=None, dynamic_source=None, static_source=None, **kwargs):
if url_type is None or url_type in ['html', 'other']:
def extract_text(url_type, url=None, dynamic_source=None, static_source=None, html_file=None, **kwargs):
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good! I would do it this way so you can save the extracted keywords separately, as a separate extractor:

def extract(extractor, **kwargs):
    content = None
    if extractor == 'text':
        content = extract_text(**kwargs)
    elif extractor == 'keywords' and 'html_file' in kwargs and kwargs['html_file'] is not None:
        content = extract_ccpa_info(kwargs['html_file'])
    return content

def extract_text(url_type, url=None, dynamic_source=None, static_source=None, **kwargs):
    if url_type is None or url_type in ['html', 'other']:
        content = extract_html(dynamic_source, url)
    elif url_type == 'pdf':
        content = extract_pdf(static_source)
    elif url_type == 'plain':
        content = dynamic_source
    else:
        content = dynamic_source
    return content

Maybe take a look and see if that achieves the same functionality? (apart from not saving it under the "text" extractor).

if html_file is not None:
content = extract_ccpa_info(html_file)
elif url_type is None or url_type in ['html', 'other']:
content = extract_html(dynamic_source, url)
elif url_type == 'pdf':
content = extract_pdf(static_source)
Expand Down Expand Up @@ -40,6 +43,25 @@ def extract_google_docs(source):
text = ''.join(chunks).strip()
return text

def extract_ccpa_info(html_file):
with open(html_file) as fp:
soup = BeautifulSoup(fp, features="html.parser")
text = soup.get_text().lower()
result = ''
for category, values in KEYWORDS.items():
substring = category + ': '
for v in values:
if v in text:
substring += v + ','
# words = v.split()
# if len(words) > 1:
# for w in words:
# if w in text:
# substring += w + ','
result += substring + '\n'
print(result)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Don't forget to remove the print statements at some later point.

return result

def extract_other(source):
"""
From https://stackoverflow.com/questions/328356/extracting-text-from-html-file-using-python
Expand Down
11 changes: 8 additions & 3 deletions polipy/polipy.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,7 @@
from .constants import UTC_DATE, CWD
from .exceptions import NetworkIOException, ParserException
from .logger import get_logger
from bs4 import BeautifulSoup
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is this import needed here?


import os
import json
Expand All @@ -27,7 +28,7 @@ class Policy:

url, source, content = {}, {}, {}

def __init__(self, url):
def __init__(self, url, html_file=None):
"""
Constructor method. Populates the `Policy.url` attribute.

Expand All @@ -39,6 +40,7 @@ def __init__(self, url):
self.url['url'] = url
self.url = self.url | parse_url(url)
self.url['domain'] = self.url['domain'].strip().strip('.').strip('/')
self.html_file = html_file
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yeah, I think this is fine for now. Maybe ultimately we could merge the self.htm_file better into the Polipy object, as that's basically the exact same attribute as the self.source['dynamic_html'] after the scraping part. But otherwise it's good to have an option of either providing an URL (and then scraping) or the already-scraped HTML to the constructor.


# Generate the hash to avoid collisions in output file names.
self.url['hash'] = hashlib.md5(url.encode()).hexdigest()[:10]
Expand Down Expand Up @@ -107,6 +109,7 @@ def extract(self, extractors=['text']):
'url_type': self.url['type'],
'static_source': self.source['static_html'],
'dynamic_source': self.source['dynamic_html'],
'html_file': self.html_file
}
for extractor in extractors:
content = extract(extractor, **vargs)
Expand Down Expand Up @@ -186,7 +189,7 @@ def __repr__(self):
return '{}({})'.format(self.__class__, self.to_dict())

# Public module methods.
def get_policy(url, screenshot=False, timeout=30, extractors=['text'], **kwargs):
def get_policy(url, html_file=None, screenshot=False, timeout=30, extractors=['text'], **kwargs):
"""
Helper method that returns a `polipy.Policy` object containing
information about the policy, scraped and processed from the given URL.
Expand All @@ -195,6 +198,8 @@ def get_policy(url, screenshot=False, timeout=30, extractors=['text'], **kwargs)
----------
url : str
The URL of the privacy policy.
html_file : str
Relative location of HTML file to process.
screenshot : bool, optional
Flag that indicates whether to capture and save the screenshot of the privacy policy page (default is `False`).
timeout : int, optional
Expand All @@ -218,7 +223,7 @@ def get_policy(url, screenshot=False, timeout=30, extractors=['text'], **kwargs)
ParserException
Raised if an error occured while extracting text from page source.
"""
policy = Policy(url)
policy = Policy(url, html_file)
policy.scrape(screenshot=screenshot, timeout=timeout)
policy.extract(extractors=extractors)
return policy
Expand Down