Fix for bug C340-38 (bouldersym set and / command used)
Ray Chason
don't want to know how promising. (Recover worked, but the save file
refused to load and I accidentally deleted it.)
But I will have my revenge.
src/pager.c, lines 648-650:
/* handle optional boulder symbol as a special case */
if (iflags.bouldersym && sym == iflags.bouldersym)
found += append_str(out_str, "boulder");
should read:
/* handle optional boulder symbol as a special case */
if (iflags.bouldersym && sym == iflags.bouldersym) {
if (!found) {
Sprintf(out_str, "%s a boulder", symstr);
firstmatch = "boulder";
found++;
} else {
found += append_str(out_str, "a boulder");
}
}
The crash occurs because firstmatch has not been set when the program
tries to print the description.
The Unicode patch, when released, will contain this fix. (I had boulder
set to U+2182, "Roman Numeral Ten Thousand"; it looks like a pair of
back-to-back Ds with a smaller pair of back-to-back Ds drawn inside.)
--
--------------===============<[ Ray Chason ]>===============--------------
PGP public key at http://www.smart.net/~rchason/pubkey.asc
Delenda est Windoze
David Damerell
>I just lost a *wickedly* promising Archeologist to this bug. You really
>don't want to know how promising. (Recover worked, but the save file
>refused to load and I accidentally deleted it.)
I trust you've sent the fix to the Dev Team, as well?
--
David Damerell <dame...@chiark.greenend.org.uk> flcl?
Dylan O'Donnell
> Ray Chason <johnn...@southland.smart.net.SPAMMEN.VERBOTEN> wrote:
> >I just lost a *wickedly* promising Archeologist to this bug. You really
> >don't want to know how promising. (Recover worked, but the save file
> >refused to load and I accidentally deleted it.)
>
> I trust you've sent the fix to the Dev Team, as well?
Since C340-38 is listed as "fixed", there should be no need to do so.
(Unless you have particular reason to think that your fix is going to
be supremely more elegant than whatever they might have come up with.)
--
: Dylan O'Donnell http://www.spod-central.org/~psmith/ :
: "The Foo service relies upon the Bar system service, which is :
: currently experiencing problems in dimensional stability, :
: believed to be caused by a passing walrus." -- pdp :
David Damerell
>David Damerell <dame...@chiark.greenend.org.uk> writes:
>>Ray Chason <johnn...@southland.smart.net.SPAMMEN.VERBOTEN> wrote:
>>>I just lost a *wickedly* promising Archeologist to this bug. You really
>>>don't want to know how promising. (Recover worked, but the save file
>>>refused to load and I accidentally deleted it.)
>>I trust you've sent the fix to the Dev Team, as well?
>Since C340-38 is listed as "fixed", there should be no need to do so.
I would normally if I wrote a fix for something myself. Perhaps it will be
more elegant; and presumably if rgrn keep up with each other's patches, at
most one of us will write a fix for any given bug.
[But no, I didn't look at the known bugs list. Ooops. :-]
Ray Chason
>Dylan O'Donnell <psm...@spod-central.org> wrote:
>>David Damerell <dame...@chiark.greenend.org.uk> writes:
>>>Ray Chason <johnn...@southland.smart.net.SPAMMEN.VERBOTEN> wrote:
>>>>I just lost a *wickedly* promising Archeologist to this bug. You really
>>>>don't want to know how promising. (Recover worked, but the save file
>>>>refused to load and I accidentally deleted it.)
>>>I trust you've sent the fix to the Dev Team, as well?
>>Since C340-38 is listed as "fixed", there should be no need to do so.
>
>I would normally if I wrote a fix for something myself. Perhaps it will be
>more elegant; and presumably if rgrn keep up with each other's patches, at
>most one of us will write a fix for any given bug.
Reasonable, I suppose, but certain Devteamers do read RGRN. Furthermore,
the fix I posted merely cuts, pastes, and tweaks a bit of code from
elsewhere in the relevant function, so that unique boulder symbols are
handled the same way as any other symbol. My patch probably differs
trivially, if at all, from the Devteam's.