paradizelost/paperless-ngx
1
0
Fork 0
mirror of https://github.com/paperless-ngx/paperless-ngx.git synced 2025-04-02 13:45:10 -05:00
Code Issues Packages Projects Releases Wiki Activity

use html.escape instead of some self build functions

Browse Source
This commit is contained in:
phail 2022-11-20 12:36:49 +01:00
parent e3c1bde793
commit 073c3c8fed
2 changed files with 5 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
src/paperless_mail/parsers.py
View File

@ -1,5 +1,6 @@
import os
import re
from html import escape
from io import BytesIO
from io import StringIO
@ -198,12 +199,7 @@ class MailDocumentParser(DocumentParser):
text = "\n".join([str(e) for e in text])
if type(text) != str:
text = str(text)
text = text.replace("&", "&")
text = text.replace("<", "&lt;")
text = text.replace(">", "&gt;")
text = text.replace(" ", " &nbsp;")
text = text.replace("'", "&apos;")
text = text.replace('"', "&quot;")
text = escape(text)
text = clean(text)
text = linkify(text, parse_email=True)
text = text.replace("\n", "<br>")

4
src/paperless_mail/tests/test_parsers.py
View File

@ -364,11 +364,13 @@ class TestParser(TestCase):
def test_mail_to_html(self):
mail = self.parser.get_parsed(os.path.join(self.SAMPLE_FILES, "html.eml"))
html_handle = self.parser.mail_to_html(mail)
html_received = html_handle.read()
with open(
os.path.join(self.SAMPLE_FILES, "html.eml.html"),
) as html_expected_handle:
self.assertHTMLEqual(html_expected_handle.read(), html_handle.read())
html_expected = html_expected_handle.read()
self.assertHTMLEqual(html_expected, html_received)
@mock.patch("paperless_mail.parsers.requests.post")
@mock.patch("paperless_mail.parsers.MailDocumentParser.mail_to_html")