|Anonymous | Login | Signup for a new account||2019-02-20 23:31 CET|
|Main | My View | View Issues | Change Log | Wiki | Tavern | News|
|Viewing Issue Simple Details|
|ID||Category||Severity||Reproducibility||Date Submitted||Last Update|
|0006548||[DCSS] Bug Report||minor||always||2013-01-18 10:16||2014-01-30 08:04|
|Status||resolved||Product Branch||0.11 ancient branch|
|Summary||0006548: There's no warning if attacking an helpless foo with portal projectile under TSO|
|Description||If you attack a monster who is helpless (confused, sleeping, etc.) shooting with portal projectile, there's no warning that action will put right under penance by TSO, while all other attack actions have one.|
|Attached Files||6548_patch_1.diff [^] (3,896 bytes) 2013-02-04 01:31 [Show Content]|
Crawl deliberately avoids "tracing" for portaled projectiles, just as it does for projectiles thrown while confused. The point of the tracing process is to bring up this prompt, and other prompts like it.
Does this mean that this is NOT a bug? I think it is still a bug, but a somewhat subtle one.
I think the intent was this: a portaled projectile teleports to its destination, so there's no need to check for stuff we don't want to hit along the way--the projectile can't hit anything else because it's not flying across the level. The bug is that it doesn't check the destination either.
I'm working on a patch. I won't finish it tonight though, I'm trying to figure out the right way to fix this.
I don't know if this is the right way to fix it or not: I added a call to stop_attack_prompt inside throw_it, just for portaled.
I tested to verify that:
-When you strike a sleeping monster with portaled projectile or a thrown projectile while worshipping TSO you are warned.
-If you strike anyway you go into penance.
-If you strike a non-helpless monster with portaled projectile or a thrown projectile while worshipping TSO you are NOT warned.
I was hoping for some feedback on part of the code:
+ // Kludgy. Ideally this would handled by the same code.
+ // Perhaps some notion of a zero length bolt, with the source and
+ // target both set to the target?
Anyway, here's a working patch which is more or less ready to go. I think.
edited on: 2014-01-30 08:04
Applied to trunk in: 0.14-a0-2343-gc2bc827
I added a test to avoid losing a turn when canceling the attack. And the next commit fixes a duplicate MSG_OK print, which was already happening if you canceled a normal throwing attack.
Sorry that this took so (very) long. Poking people in ##crawl-dev on FreeNode might help in cases where patches seem to be forgotten about.
|2013-01-18 10:16||nago||New Issue|
|2013-01-19 01:24||blackcustard||Status||new => acknowledged|
|2013-01-19 04:55||blackcustard||Note Added: 0020827|
|2013-02-04 01:31||blackcustard||File Added: 6548_patch_1.diff|
|2013-02-04 03:39||blackcustard||Note Added: 0020955|
|2013-02-04 04:12||blackcustard||Issue Monitored: blackcustard|
|2013-02-04 04:13||blackcustard||Status||acknowledged => feedback|
|2013-03-27 19:03||mumra||Relationship added||has duplicate 0006835|
|2013-04-20 12:51||Medar||Tag Attached: patch|
|2014-01-30 08:04||Medar||Note Added: 0025140|
|2014-01-30 08:04||Medar||Status||feedback => resolved|
|2014-01-30 08:04||Medar||Fixed in Branch||=> 0.14 development branch|
|2014-01-30 08:04||Medar||Resolution||open => done|
|2014-01-30 08:04||Medar||Assigned To||=> Medar|
|2014-01-30 08:04||Medar||Note Edited: 0025140|
|Mantis 1.1.8[^] Copyright © 2000 - 2009 Mantis Group|