| 2016-11-23 21:28:07 |
Mark Sapiro |
description |
This may not occur in a normal installation, but on a test list I had an accumulated digest.mbox with non-ascii characters in some message subjects. I then changed the list's preferred_language from English to Japanese and during some subsequent testing, a digest was triggered. The creation of the digest would throw UnicodeError in attempting to create the MIME toc part and in attempting to set the plain (RFC 1133) digest payload.
This may be due only to the change to a language with a different character set or it may be due to the fact that Mailman's character set for Japanese is euc-jp, but the email modules output characterset for euc-jp is iso-2022-jp. The exception thrown was due to unicodes nt representable in iso-2022-jp. |
This may not occur in a normal installation, but on a test list I had an accumulated digest.mbox with non-ascii characters in some message subjects. I then changed the list's preferred_language from English to Japanese and during some subsequent testing, a digest was triggered. The creation of the digest would throw UnicodeError in attempting to create the MIME toc part and in attempting to set the plain (RFC 1133) digest payload.
This may be due only to the change to a language with a different character set or it may be due to the fact that Mailman's character set for Japanese is euc-jp, but the email modules output characterset for euc-jp is iso-2022-jp. The exception thrown was due to unicodes not representable in iso-2022-jp. |
|
