ebola in Gnus 5.6.44

ebola in Gnus 5.6.44

[G. Evermann]

When downloding articles for the agent I get lots of the following
messages on the console:

Comparison between integer and character is constant nil (?2 and 0)
   [nntp-retrieve-articles, gnus-retrieve-articles, gnus-agent-fetch-articles, gnus-agent-fetch-group-1, <compiled-function of unknown origin>, call-interactively]

This is XEmacs 21.2-beta3 but I have also seen it in 21.0 (also on a
different news-server).

If anybody wants to reproduce this, remember to compile with -debug to
get the ebola notices.


Gunnar

Re: ebola in Gnus 5.6.44

[Lars Magne Ingebrigtsen]

"G. Evermann" <ge204@eng.cam.ac.uk> writes:

> When downloding articles for the agent I get lots of the following
> messages on the console:
> 
> Comparison between integer and character is constant nil (?2 and 0)
>    [nntp-retrieve-articles, gnus-retrieve-articles, gnus-agent-fetch-articles, gnus-agent-fetch-group-1, <compiled-function of unknown origin>, call-interactively]

Let's see...  The only places ?2 appears is in

(eq (following-char) ?2)

and

(memq (char-after (point)) '(?2 ?3 ?4 ?5))

in nntp.el.

Is this verboten?

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: ebola in Gnus 5.6.44

[Hrvoje Niksic]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> Let's see...  The only places ?2 appears is in
> 
> (eq (following-char) ?2)
[...]
> Is this verboten?

This is technically legal in XEmacs, but confuses the hell out of
ebola-warning code because (following-char) returns 0 (not ?\0 -- the
two are distinct in XEmacs 20) so the code thinks that you might be
comparing integers and characters heedlessly, which you are not.

It's better style to replace (following-char) with (char-after), which 
returns nil at eob.

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
We must all confront and destroy our own inner geek.

Re: ebola in Gnus 5.6.44

[Lars Magne Ingebrigtsen]

Hrvoje Niksic <hniksic@srce.hr> writes:

> It's better style to replace (following-char) with (char-after), which 
> returns nil at eob.

Okidoke.  I've now done this thoughout 0.37 (and changed a bunch of
='s to eq's).

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: ebola in Gnus 5.6.44

[Hrvoje Niksic]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> Hrvoje Niksic <hniksic@srce.hr> writes:
> 
> > It's better style to replace (following-char) with (char-after), which 
> > returns nil at eob.
> 
> Okidoke.  I've now done this thoughout 0.37

Cool, thanks.  XEmacs will be happy.  XEmacs will eat someone else.

> (and changed a bunch of ='s to eq's).

Note that, under XEmacs 20, (eq ?A 65) is nil, while (= ?A 65) remains 
t.  So for some comparisons where characters need to be coerced to
integers, `=' makes sense.

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
Oh lord won't you buy me a color TV...