I am trying to debug a problem with Supercite's attribution line
when used with the VM mailreader, and I'm not sure which of the two
programs is to blame. Supercite is called by VM in the standard way,
i.e. via mail-citation-hook and function sc-cite-original.
The problem occurs when replying to an e-mail message with a
QP-encoded "From" header, which will have a broken attribution line
like the following:
>>>>> On Fri, 30 Oct 2009, =?ISO-8859-1?Q?Ulrich_M=FCller?= wrote:
Now the Supercite Info file says in section 10 "Hints to MUA Authors":
| [...] when setting up a reply or forward buffer, your MUA should
| follow these steps:
|
| 1. Insert the original message, including the mail headers into the
| reply buffer. At this point you should not modify the raw text in
| any way, and you should place all the original headers into the
| body of the reply. [...]
The wording ("not modify the raw text in any way", "original headers")
seems to indicate that the MUA shouldn't do any decoding here.
However, Supercite also doesn't do any QP decoding, which leads to the
above broken attribution line.
So my question is, who should decode the headers? The MUA (then the
documentation of mail-citation-hook in sc.texi should be changed or
clarified), or Supercite (then it's a bug there)?
Ulrich
