mirrored totem can't properly see glyph

[MattWeiner]

In seed 335089435, floor 19, a mirrored totem was constructed around the corner (and across some brimstone) from its associated glyphs, so it failed to beckon the player and I could walk right across the glyphs without any special measures.

This can be seen on turn 12253 of brogue.roguelikelike.com/#viewRecording/BROGUECEV120-64a3644e4fbd8e1fdd29d484. The totem is visible from the glyphs directly horizontal from the door it is guarding, but (because of the new staff bolt pathing?) it doesn't beckon the player. It would not be visible from many of the other glyphs.

[MattWeiner]

Updated issue with the seed and turn number.

[tmewett]

I think this is the same as #265 ? The other way around to the specific example though

[nstoddard]

I think it's sort of the opposite of #497, actually; FOV isn't relevant here. If the bolt were computed the other way around (from the player to the totem) it would be able to find a path that doesn't hit a wall. This is yet another example of an issue caused by asymmetric bolt paths.

Would it be worth it to make a meta-issue for improving FOV/bolt paths? I think most if not all of these sorts of issues could be fixed by using the same algorithm for both FOV and bolt paths (or some other way to make them consistent) and/or making FOV/bolt paths symmetric.

[MattWeiner]

For what it's worth, the way that the floor was constructed, some of the glyphs would be blocked from the totem no matter what the bolt targeting algorithm is--the ones north of the door are on the other side of a wall. But it's the ones next to the door that count.