[FIX] mail: mime type alias

Some printers incorrectly specify the content-type of PDFs as "pdf" instead of
the correct "application/pdf." To handle this situation,
an alias has been introduced, treating content-type "pdf" as "application/pdf."

[Reproducing original bug]
1. run odoo with documents module
2. prepare email file
2.1 create email with a pdf attached
2.2 export email to .eml
2.3 modify content-type of the pdf from "application/pdf" to "pdf"
3. send prepared email to odoo to inbox-financial alias (the default one
   creating documents)
4. Go to documents, observe that the created PDF is blank!

[Motivation for fix]
Since:
1. Odoo uses the Python built-in module for parsing emails*
    (the actual issue lies outside Odoo code).
2. "pdf" is not an accepted MIME type according to current norms
    (https://datatracker.ietf.org/doc/html/rfc6838#section-4.2).
It's not an actual Odoo bug.
However, redirecting this issue to the appropriate printer manufacturer and
waiting for a fix may be time-consuming, hence the proposed fix.

*function parsing email with the Python built-in module
https://github.com/odoo/odoo/blob/407ea60796a2b18eb02f5569e1cfab9f1163a572/addons/mail/models/mail_thread.py#L1282

opw-3462260

closes odoo/odoo#147363

X-original-commit: e95a19341536e442239a7e940089d7f66ab9f534
Signed-off-by: Thibault Delavallee (tde) <tde@openerp.com>

addons/mail/models/mail_thread.py

@@ -1523,6 +1523,10 @@ class MailThread(models.AbstractModel):
                     part.set_charset('utf-8')
                 encoding = part.get_content_charset()  # None if attachment
 
+                # Correcting MIME type for PDF files
+                if part.get('Content-Type', '').startswith('pdf;'):
+                    part.replace_header('Content-Type', 'application/pdf' + part.get('Content-Type', '')[3:])
+
                 content = part.get_content()
                 info = {'encoding': encoding}
                 # 0) Inline Attachments -> attachments, with a third part in the tuple to match cid / attachment

addons/test_mail/data/test_mail_data.py

@@ -1292,3 +1292,23 @@ To: lucie@petitebedaine.fr
 From: "Bruce Wayne" <bruce@wayneenterprises.com>
 
 SSBhbSB0aGUgQmF0TWFuCg=="""
+
+MAIL_PDF_MIME_TEMPLATE = """\
+To: {to}
+From: {email_from}
+Subject: {subject}
+MIME-Version: 1.0
+Content-Type: multipart/alternative;
+    boundary="----=_Part_4200734_24778174.1344608186754"
+Date: Fri, 10 Aug 2012 14:16:26 +0000
+
+------=_Part_4200734_24778174.1344608186754
+Content-Type: {pdf_mime}; name="scan_soraya.lernout_1691652648.pdf"
+Content-Transfer-Encoding: base64
+
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
+
+------=_Part_4200734_24778174.1344608186754--
+"""
+
+PDF_PARSED = b'''%PDF-1.1\n%\xc2\xa5\xc2\xb1\xc3\xab\n\n1 0 obj\n  << /Type /Catalog\n     /Pages 2 0 R\n  >>\nendobj\n\n2 0 obj\n  << /Type /Pages\n     /Kids [3 0 R]\n     /Count 1\n     /MediaBox [0 0 300 144]\n  >>\nendobj\n\n3 0 obj\n  <<  /Type /Page\n      /Parent 2 0 R\n      /Resources\n       << /Font\n           << /F1\n               << /Type /Font\n                  /Subtype /Type1\n                  /BaseFont /Times-Roman\n               >>\n           >>\n       >>\n      /Contents 4 0 R\n  >>\nendobj\n\n4 0 obj\n  << /Length 55 >>\nstream\n  BT\n    /F1 18 Tf\n    0 0 Td\n    (Hello World) Tj\n  ET\nendstream\nendobj\n\nxref\n0 5\n0000000000 65535 f \n0000000018 00000 n \n0000000077 00000 n \n0000000178 00000 n \n0000000457 00000 n \ntrailer\n  <<  /Root 1 0 R\n      /Size 5\n  >>\nstartxref\n565\n%%EOF\n'''

addons/test_mail/tests/test_mail_gateway.py

@@ -83,6 +83,19 @@ class TestEmailParsing(MailCommon):
         res = self.env['mail.thread'].message_parse(self.from_string(test_mail_data.MAIL_MULTIPART_WEIRD_FILENAME))
         self.assertEqual(res['attachments'][0][0], '62_@;,][)=.(ÇÀÉ.txt')
 
+    def test_message_parse_attachment_pdf_nonstandard_mime(self):
+        # This test checks if aliasing content-type (mime type) of "pdf" with "application/pdf" works correctly. (i.e. Treat "pdf" as "application/pdf")
+
+        # Baseline check. Parsing mail with "application/pdf"
+        mail_with_standard_mime = self.format(test_mail_data.MAIL_PDF_MIME_TEMPLATE, pdf_mime="application/pdf")
+        res_std = self.env['mail.thread'].message_parse(self.from_string(mail_with_standard_mime))
+        self.assertEqual(res_std['attachments'][0].content, test_mail_data.PDF_PARSED, "Attachment with Content-Type: application/pdf must parse without error")
+
+        # Parsing the same email, but with content-type set to "pdf"
+        mail_with_aliased_mime = self.format(test_mail_data.MAIL_PDF_MIME_TEMPLATE, pdf_mime="pdf")
+        res_alias = self.env['mail.thread'].message_parse(self.from_string(mail_with_aliased_mime))
+        self.assertEqual(res_alias['attachments'][0].content, test_mail_data.PDF_PARSED, "Attachment with aliased Content-Type: pdf must parse without error")
+
     def test_message_parse_bugs(self):
         """ Various corner cases or message parsing """
         # message without Final-Recipient