From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.4) Gecko/20011019 Netscape6/6.2 Description of problem: Posting a newsgroup, such as "pl.test" results in a lisp error because rfc2047-encodable-p encounters a lisp symbol instead of a list. It probably occurs for any non-standard encoding setting, pl.test will make GNUS happen to use iso-8859-2. Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1.Post test posting to "pl.test" 2. 3. Actual Results: Lisp error. The posting actually goes out, but the posting buffer sits there and you have to kill it off manually. This confused my wife. Expected Results: The posting should have gone out without any lisp errors. Additional info: If we are going to ship a different version of GNUS that that contained in the stock emacs sources, we should at least test simple things like posting to non-english newsgroups. Here is the LISP backtrace btw: Signaling: (wrong-type-argument listp iso-8859-2) rfc2047-encodable-p() mail-encode-encoded-word-buffer() message-do-fcc() message-send(nil) message-send-and-exit(nil) * call-interactively(message-send-and-exit)
It has been tested with non-English newsgroups (Norwegian ones :). Can you check if emacs 21 (available from http://people.redhat.com/teg/emacs/) solves the problem?
This causes a complete hang of my wife's computer when GNUS is started. This was under X. We rebooted cleanly, and from a plain text console I tried again. Same thing, starting GNUS causes a full system hang. Wouldn't it be easier just to fix the code in the 20.7 SRPM so that the rfc2047-encodable-p elisp defun gets the correct types? ... Just an idea... When I traced the code it seems that the "newgroup regexp" --> "charset" mapping was just using incorrect types. System specs: Athlon 1700+ XP 512MB ram all kernel/glibc/etc. updates installed
>This causes a complete hang of my wife's computer when >GNUS is started. That sounds like a kernel bug to me. Or an X bug. Not an emacs bug...
And if I would keep reading, I would notice that it is definitely a kernel bug.
Still the case with modern kernels Dave ? and if so is it emacs running the box out of ram?
No, turned out to be dodgy ram in my wife's Athlon.