open
https://gitlab.synchro.net/main/sbbs/-/work_items/1361
`MsgBase.get_msg_parts()` (new in v3.22) reports an attachment's
`name` from the `name=` parameter of Content-Type or the
`filename=` parameter of Content-Disposition (src/smblib/smbtxt.c, `mime_param()` called with "name" and "filename").
Many mail clients send a non-ASCII file name only in the RFC 2231
form:
```
Content-Disposition: attachment;
filename*=UTF-8''R%C3%A9sum%C3%A9.pdf
```
or split into continuations (`filename*0*=`, `filename*1*=`). Those
parameters are not recognized, so the part comes back with an empty
name and a script has nothing to show but a generic one.
Wanted: recognize `filename*=` / `name*=` (charset, optional
language, percent-encoded value) and the numbered continuation form,
and return the decoded name; say in the method's documentation which
charset the returned name is in. RFC 2047 encoded words inside a
quoted `filename="..."` are common too and are currently returned as
stored, which is fine as long as it is documented.
-- *Authored by Claude (Claude Code), on behalf of @rswindell*
--- SBBSecho 3.38-Linux
* Origin: Vertrauen - [vert/cvs/bbs].synchro.net (1:103/705)