VM

vm-8.0.13 mapvector ; autoloads failing

Bug #492117 reported by Matthew Lovell
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
VM
Status tracked in Trunk
8.0.x
Fix Released
High
Uday Reddy
Trunk
Fix Released
High
Uday Reddy

Bug Description

vm-8.0.13 introduced a defun for mapvector in vm-folder. This definition appears to stomp upon the function built into xemacs already:

  `mapvector' is a built-in function
  (mapvector FUNCTION SEQUENCE)

This became evident trying to invoke latex-mode on a buffer after having upgraded from vm-8.0.12 to vm-8.0.13. This resulted in an error saying that a mapvector usage had an incorrect argument.

Changing vm-folder.el such that it defines and uses vm-mapvector appears to have fixed this behavior.

I was attempting to revert this change to obtain a back trace of the error, but I am currently unable to reproduce the exact one I am experiencing. Nevertheless, it would seem unwise to defun a duplicate of a built-in function.

Related branches

Revision history for this message
Matthew Lovell (lovell-hp) wrote :

Backtrace obtained when trying to enter latex-mode with original vm-8.0.13 installed:

Debugger entered--Lisp error: (wrong-type-argument arrayp (?\`))
  mapvector(identity (?\`))
  edmacro-parse-keys("`")
  read-kbd-macro("`")
  LaTeX-math-abbrev-prefix()
  byte-code("..." [map make-sparse-keymap define-key LaTeX-math-abbrev-prefix self-insert-command] 4)
  (defvar LaTeX-math-keymap (byte-code "Á Âà Ä#ˆ)‡" [map make-sparse-keymap define-key LaTeX-math-abbrev-prefix self-insert-command] 4) ("/home/lovell/.xemacs/xemacs-packages/lisp/auctex/latex.elc" . 99787))
  load-internal("latex" nil nil nil binary)
  load("latex" nil nil nil)
  command-execute(latex-mode t)
  execute-extended-command(nil)
  call-interactively(execute-extended-command)

Uday Reddy (reddyuday)
Changed in vm:
status: New → Confirmed
assignee: nobody → Uday Reddy (reddyuday)
Revision history for this message
Uday Reddy (reddyuday) wrote :

Thanks very much, Matthew. Sigh... I have to do another release in quick succession and withdraw 8.0.13. Embarassing...

Can you help us about by testing pre-releases in XEmacs?

Changed in vm:
importance: Undecided → High
milestone: none → 8.0.14
Revision history for this message
Matthew Lovell (lovell-hp) wrote : [Bug 492117] Re: vm-8.0.13 mapvector possibly collides with xemacs built-in function

Hi Uday,

Uday Reddy writes:
 > Thanks very much, Matthew. Sigh... I have to do another release in
 > quick succession and withdraw 8.0.13. Embarassing...
 >
 > Can you help us about by testing pre-releases in XEmacs?

Certainly.

I find I'm also still having problems with auto-autoloads.el and

  vm-decode-postponed-mime-message

my solution currently is to copy auto-autoloads.el over to
vm-autoloads.el and add

  (provide 'vm-autoload)

at the tail.

Thanks,
  Matt

 > ** Changed in: vm
 > Importance: Undecided => High
 >
 > ** Changed in: vm
 > Milestone: None => 8.0.14
 >

--
Matthew Lovell
Hewlett-Packard

Revision history for this message
Uday Reddy (reddyuday) wrote : Re: vm-8.0.13 mapvector possibly collides with xemacs built-in function

Hi Matt, I just built 8.0.13 in xemacs and find that the auto-autoloads.el has the right autoload for vm-pine including vm-decode-postponed-mime-message. (see below)

Can you elaborate what you mean by "still having problems"?

Cheers,
Uday

;;;### (autoloads (vm-mail-to-fcc vm-mail-select-folder vm-mail-get-header-contents vm-mail-auto-fcc vm-mail-fcc vm-mail-priority vm-mail-notice-requested-upon-delivery-to vm-mail-return-receipt-to vm-continue-what-message-other-frame vm-continue-what-message-other-window vm-continue-what-message vm-postpone-message vm-decode-postponed-mime-message vm-reply-by-continue-postponed-message vm-continue-postponed-message vm-summary-function-p vm-summary-function-f) "vm-pine" "lisp/vm-pine.el")

Uday Reddy (reddyuday)
Changed in vm:
status: Confirmed → Fix Committed
Revision history for this message
Uday Reddy (reddyuday) wrote :

Matt, I have put the release of 8.0.14 on the Launchpad project page. Can you please verify that it works? -- Uday

Uday Reddy (reddyuday)
Changed in viewmail:
assignee: nobody → Uday Reddy (reddyuday)
status: New → In Progress
importance: Undecided → High
Uday Reddy (reddyuday)
Changed in viewmail:
status: In Progress → Fix Committed
Uday Reddy (reddyuday)
Changed in viewmail:
status: Fix Committed → Fix Released
Revision history for this message
Uday Reddy (reddyuday) wrote :

Hi Matthew, I am waiting for your confirmation that the release 8.0.14 is ok before I announce it. So, can you check it please? I am also curious whether there is still a problem with the build process regarding auto-autoloads.el etc.

Cheers,
Uday

Revision history for this message
Uday Reddy (reddyuday) wrote : [Bug 492117] Re: vm-8.0.13 mapvector possibly collides with xemacs built-in function

Matthew Lovell writes:

> I find I'm also still having problems with auto-autoloads.el and
>
> vm-decode-postponed-mime-message
>
> my solution currently is to copy auto-autoloads.el over to
> vm-autoloads.el and add
>
> (provide 'vm-autoload)
>
> at the tail.

Hi Matthew,

You probably have (provide 'vm-autoloads) in your init.el file. Can
you change it to

  (load-library "/path/to/vm/auto-autoloads")

and see if it solves the problem?

Cheers,
Uday

summary: - vm-8.0.13 mapvector possibly collides with xemacs built-in function
+ vm-8.0.13 mapvector ; autoloads failing
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.