def test_other__check_thunderbird_mail_text_quote_signature(self):
text_quote_signature = '''
<p>Coucou<br>
</p>
<br>
<div class="moz-cite-prefix">Le 28/11/2017 à 11:22, Bidule a
écrit :<br>
</div>
<blockquote type="cite"
cite="mid:[email protected]">Réponse<br>
<br>
Le 28/11/2017 à 11:21, John Doe a écrit : <br>
<blockquote type="cite"> <br>
Test <br>
<br>
-- <br>
TEST DE signature <br>
</blockquote>
<br>
</blockquote>
<br>
<div class="moz-signature">-- <br>
TEST DE signature</div>
'''
mail = ParsedHTMLMail(text_quote_signature)
elements = mail.get_elements()
assert len(elements) == 3
assert elements[0].part_type == BodyMailPartType.Main
assert elements[1].part_type == BodyMailPartType.Quote
assert elements[2].part_type == BodyMailPartType.Signature
def test_other__check_outlook_com_mail_text_signature(self):
text_and_signature = '''
<div id="divtagdefaultwrapper"
style="font-size:12pt;color:#000000;
font-family:Calibri,Helvetica,sans-serif;"
dir="ltr">
<p style="margin-top:0;margin-bottom:0">Test<br>
</p>
<p style="margin-top:0;margin-bottom:0"><br>
</p>
<div id="Signature">
<div id="divtagdefaultwrapper" style="font-size: 12pt; color:
rgb(0, 0, 0); background-color: rgb(255, 255, 255);
font-family:
Calibri,Arial,Helvetica,sans-serif,"EmojiFont","Apple
Color Emoji","Segoe UI
Emoji",NotoColorEmoji,"Segoe UI
Symbol","Android Emoji",EmojiSymbols;">
Envoyé à partir de <a href="http://aka.ms/weboutlook"
id="LPNoLP">Outlook</a></div>
</div>
</div>
'''
mail = ParsedHTMLMail(text_and_signature)
elements = mail.get_elements()
assert len(elements) == 2
assert elements[0].part_type == BodyMailPartType.Main
assert elements[1].part_type == BodyMailPartType.Signature
def test_other__check_thunderbird_mail_text_quote_text_signature(self):
text_quote_text_sign = '''
<p>Avant<br>
</p>
<br>
<div class="moz-cite-prefix">Le 28/11/2017 à 11:19, Bidule a
écrit :<br>
</div>
<blockquote type="cite"
cite="mid:[email protected]">Coucou
<br><br>
</blockquote>
Aprés<br>
<br>
<div class="moz-signature">-- <br>
TEST DE signature</div>
'''
mail = ParsedHTMLMail(text_quote_text_sign)
elements = mail.get_elements()
assert len(elements) == 4
assert elements[0].part_type == BodyMailPartType.Main
assert elements[1].part_type == BodyMailPartType.Quote
assert elements[2].part_type == BodyMailPartType.Main
assert elements[3].part_type == BodyMailPartType.Signature
def test_other__check_thunderbird_mail_text_only(self):
text_only = '''Coucou<br><br><br>'''
mail = ParsedHTMLMail(text_only)
elements = mail.get_elements()
assert len(elements) == 1
assert elements[0].part_type == BodyMailPartType.Main
def test_other__check_gmail_mail_text_quote(self):
text_and_quote = '''
<div dir="ltr">Réponse<br>
<div class="gmail_extra"><br>
<div class="gmail_quote">Le 28 novembre 2017 à 10:29, John Doe <span
dir="ltr"><<a href="mailto:[email protected]"
target="_blank">[email protected]</a>></span>
a écrit :<br>
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">Voici ma réponse<br>
<br><br>
Le 28/11/2017 à 10:05, Foo Bar a écrit :<br>
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">
Voici le texte<span class="HOEnZb"><font color="#888888"><br>
</font></span></blockquote>
<span class="HOEnZb"><font color="#888888">
<br>
-- <br>
TEST DE signature<br>
</font></span></blockquote>
</div><br></div></div>
'''
mail = ParsedHTMLMail(text_and_quote)
elements = mail.get_elements()
assert len(elements) == 2
assert elements[0].part_type == BodyMailPartType.Main
assert elements[1].part_type == BodyMailPartType.Quote
def test_other__check_thunderbird_mail_text_signature(self):
text_and_signature = '''
<p>Test<br>
</p>
<div class="moz-signature">-- <br>
TEST DE signature</div>
'''
mail = ParsedHTMLMail(text_and_signature)
elements = mail.get_elements()
assert len(elements) == 2
assert elements[0].part_type == BodyMailPartType.Main
assert elements[1].part_type == BodyMailPartType.Signature
def test_other__check_outlook_com_mail_text_quote(self):
text_and_quote = '''
<div id="divtagdefaultwrapper"
style="font-size:12pt;color:#000000;font-family:Calibri,Helvetica,sans-serif;"
dir="ltr">
<p style="margin-top:0;margin-bottom:0">Salut !<br>
</p>
</div>
<hr style="display:inline-block;width:98%" tabindex="-1">
<div id="divRplyFwdMsg" dir="ltr"><font style="font-size:11pt"
color="#000000" face="Calibri, sans-serif"><b>De :</b> John Doe<br>
<b>Envoyé :</b> mardi 28 novembre 2017 12:44:59<br>
<b>À :</b> [email protected]<br>
<b>Objet :</b> voila</font>
<div> </div>
</div>
<style type="text/css" style="display:none">
<!--
p
	{margin-top:0;
	margin-bottom:0}
-->
</style>
<div dir="ltr">
<div id="x_divtagdefaultwrapper" dir="ltr" style="font-size:12pt;
color:#000000; font-family:Calibri,Helvetica,sans-serif">
Contenu
<p style="margin-top:0; margin-bottom:0"><br>
</p>
<div id="x_Signature">
<div id="x_divtagdefaultwrapper" dir="ltr"
style="font-size:12pt; color:rgb(0,0,0);
background-color:rgb(255,255,255);
font-family:Calibri,Arial,Helvetica,sans-serif,"EmojiFont","Apple
Color Emoji","Segoe UI
Emoji",NotoColorEmoji,"Segoe UI
Symbol","Android Emoji",EmojiSymbols">
DLMQDNLQNDMLQS<br>
qs<br>
dqsd<br>
d<br>
qsd<br>
</div>
</div>
</div>
</div>
'''
mail = ParsedHTMLMail(text_and_quote)
elements = mail.get_elements()
assert len(elements) == 2
assert elements[0].part_type == BodyMailPartType.Main
assert elements[1].part_type == BodyMailPartType.Quote
def test_other__check_gmail_mail_text_quote_signature(self):
text_quote_signature = '''
<div dir="ltr">Hey !<br>
</div>
<div class="gmail_extra"><br>
<div class="gmail_quote">Le 28 novembre 2017 à 10:29,
John Doe <span
dir="ltr"><<a href="mailto:[email protected]"
target="_blank">[email protected]</a>></span>
a écrit :<br>
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">Voici ma
réponse<br>
<br>
<br>
Le 28/11/2017 à 10:05, Foo Bar a écrit :<br>
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">
Voici le texte<span class="HOEnZb"><font color="#888888"><br>
</font></span></blockquote>
<span class="HOEnZb"><font color="#888888">
<br>
-- <br>
TEST DE signature<br>
</font></span></blockquote>
</div>
<br>
<br clear="all">
<br>
-- <br>
<div class="gmail_signature" data-smartmail="gmail_signature">
<div dir="ltr">Voici Ma signature. En HTML <br>
<ol>
<li>Plop</li>
<li>Plip</li>
<li>Plop<br>
</li>
</ol>
</div>
</div>
</div>
'''
# INFO - G.M - 2017-11-28 -
# Now Quote + Signature block in Gmail is considered as one Quote
# Block.
mail = ParsedHTMLMail(text_quote_signature)
elements = mail.get_elements()
assert len(elements) == 2
assert elements[0].part_type == BodyMailPartType.Main
assert elements[1].part_type == BodyMailPartType.Quote
def test_other__check_outlook_com_mail_text_only(self):
text_only = '''
<div id="divtagdefaultwrapper"
style="font-size:12pt;color:#000000;
font-family:Calibri,Helvetica,sans-serif;"
dir="ltr">
<p style="margin-top:0;margin-bottom:0">message<br>
</p>
</div>
'''
mail = ParsedHTMLMail(text_only)
elements = mail.get_elements()
assert len(elements) == 1
assert elements[0].part_type == BodyMailPartType.Main
def test_other__check_gmail_mail_text_signature(self):
text_and_signature = '''
<div dir="ltr">POF<br clear="all"><div><br>-- <br>
<div class="gmail_signature" data-smartmail="gmail_signature">
<div dir="ltr">Voici Ma signature. En HTML <br><ol>
<li>Plop</li>
<li>Plip</li>
<li>Plop<br>
</li></ol></div></div></div></div>
'''
mail = ParsedHTMLMail(text_and_signature)
elements = mail.get_elements()
assert len(elements) == 2
assert elements[0].part_type == BodyMailPartType.Main
assert elements[1].part_type == BodyMailPartType.Signature
def test_other__check_thunderbird_mail_text_quote_text(self):
text_quote_text = '''
<p>Pof<br>
</p>
<br>
<div class="moz-cite-prefix">Le 28/11/2017 à 11:54,
Bidule a
écrit :<br>
</div>
<blockquote type="cite"
cite="mid:[email protected]">
<meta http-equiv="Context-Type"
content="text/html; charset=utf-8">
<p>Pof<br>
</p>
<br>
<div class="moz-cite-prefix">Le 28/11/2017 à 11:21, John Doe a
écrit :<br>
</div>
<blockquote type="cite"
cite="mid:[email protected]">
<p>Test<br>
</p>
<div class="moz-signature">-- <br>
TEST DE signature</div>
</blockquote>
<br>
</blockquote>
Pif<br>
'''
mail = ParsedHTMLMail(text_quote_text)
elements = mail.get_elements()
assert len(elements) == 3
assert elements[0].part_type == BodyMailPartType.Main
assert elements[1].part_type == BodyMailPartType.Quote
assert elements[2].part_type == BodyMailPartType.Main
def get_body(
self,
use_html_parsing=True,
use_txt_parsing=True,
) -> typing.Optional[str]:
body_part = self._get_mime_body_message()
body = None
if body_part:
charset = body_part.get_content_charset('iso-8859-1')
content_type = body_part.get_content_type()
if content_type == CONTENT_TYPE_TEXT_PLAIN:
txt_body = body_part.get_payload(decode=True).decode(charset)
if use_txt_parsing:
txt_body = EmailReplyParser.parse_reply(txt_body)
html_body = markdown.markdown(txt_body)
body = HtmlSanitizer.sanitize(html_body)
elif content_type == CONTENT_TYPE_TEXT_HTML:
html_body = body_part.get_payload(decode=True).decode(charset)
if use_html_parsing:
html_body = str(ParsedHTMLMail(html_body))
body = HtmlSanitizer.sanitize(html_body)
return body
def test_other__check_outlook_com_mail_text_signature_quote(self):
text_signature_quote = '''
<div id="divtagdefaultwrapper"
style="font-size:12pt;color:#000000;font-family:Calibri,Helvetica,sans-serif;"
dir="ltr">
<p style="margin-top:0;margin-bottom:0">Salut !<br>
</p>
<p style="margin-top:0;margin-bottom:0"><br>
</p>
<div id="Signature">
<div id="divtagdefaultwrapper" dir="ltr" style="font-size: 12pt;
color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);
font-family:
Calibri,Arial,Helvetica,sans-serif,"EmojiFont","Apple
Color Emoji","Segoe UI
Emoji",NotoColorEmoji,"Segoe UI
Symbol","Android Emoji",EmojiSymbols;">
Envoyée depuis Outlook<br>
</div>
</div>
</div>
<hr style="display:inline-block;width:98%" tabindex="-1">
<div id="divRplyFwdMsg" dir="ltr"><font style="font-size:11pt"
color="#000000" face="Calibri, sans-serif"><b>De :</b> John Doe
<[email protected]><br>
<b>Envoyé :</b> mardi 28 novembre 2017 12:51:42<br>
<b>À :</b> John Doe<br>
<b>Objet :</b> Re: Test</font>
<div> </div>
</div>
<div style="background-color:#FFFFFF">
<p>Coucou<br>
</p>
<br>
<div class="x_moz-cite-prefix">Le 28/11/2017 à 12:39, John Doe a
écrit :<br>
</div>
<blockquote type="cite">
<div id="x_divtagdefaultwrapper" dir="ltr">
<p>Test<br>
</p>
<p><br>
</p>
<div id="x_Signature">
<div id="x_divtagdefaultwrapper">Envoyé à partir de <a
href="http://aka.ms/weboutlook" id="LPNoLP">
Outlook</a></div>
</div>
</div>
</blockquote>
<br>
<div class="x_moz-signature">-- <br>
TEST DE signature</div>
</div>
'''
mail = ParsedHTMLMail(text_signature_quote)
elements = mail.get_elements()
assert len(elements) == 3
assert elements[0].part_type == BodyMailPartType.Main
assert elements[1].part_type == BodyMailPartType.Signature
assert elements[2].part_type == BodyMailPartType.Quote
def test_other__check_gmail_mail_text_only(self):
text_only = '''<div dir="ltr">Voici le texte<br></div>'''
mail = ParsedHTMLMail(text_only)
elements = mail.get_elements()
assert len(elements) == 1
assert elements[0].part_type == BodyMailPartType.Main