Open Bug 1585288 Opened 6 years ago Updated 3 years ago

When saving mail messages, RFC 2047 header fields are not decoded but stored 1:1, like in To: =?UTF-8?B?Ss ...

Categories

(MailNews Core :: General, defect, P4)

defect

Tracking

(Not tracked)

People

(Reporter: jorgk-bmo, Unassigned)

Details

In a message saved as HTML I see:

<td><b>To: </b>=?UTF-8?B?Ss [snip]=?= &lt;jorgk@jorgk.com&gt;</td>

Also, the HTML document we produce doesn't have any charset information. So we need to add that and then decode RFC 2047 headers as UTF-8.

I wonder what happens if the original mail was in windows-1252 or any other non-UTF-8 charset and is exported.

OK, to answer that: I tried windows-1252. It's exported in the original character set and in absence of charset information, the browser needs to guess the charset. Of course we will run into trouble if a header is in UTF-8 and the body in non-UTF-8.

Looks like a can of worms.

Severity: normal → S3
You need to log in before you can comment on or make changes to this bug.