Dwarf Fortress Bug Tracker - Dwarf Fortress
View Issue Details
0005853Dwarf FortressDwarf Mode -- Buildings, Generalpublic2012-04-30 10:172012-04-30 10:21
Quietust 
 
normalminorhave not tried
newopen 
0.34.07 
 
0005853: Bridge Support logic incorrect for left corners
If a retracting bridge is not supported by a floor tile at any of its edges, it will deconstruct as soon as it realizes something is wrong (e.g. when digging nearby or trying to retract it).

However, the logic for ignoring the corner tiles is broken - the upper-left and lower-left corners provide support, but the 3rd-from-left tiles along the top and bottom do not provide support.

Intended "support" tiles for a 5x5 bridge ("B" = bridge, "+" = good, "x" = bad):

x+++++x
+BBBBB+
+BBBBB+
+BBBBB+
+BBBBB+
+BBBBB+
x+++++x


Actual "support" tiles for a 5x5 bridge ("B" = bridge, "+" = good, "x" = bad):

++x+++x
+BBBBB+
+BBBBB+
+BBBBB+
+BBBBB+
+BBBBB+
++x+++x
This was originally discovered by angavrilov while examining parts of the code for constructing buildings (in order to figure out how to construct buildings within DFHack) - apparently, the code is using "x+1" instead of "x-1" for the upper-left and lower-left tiles.
No tags attached.
related to 0009946new  Retracting bridges don't require support for construction, harmlessly deconstruct when checking support later 
Issue History
2012-04-30 10:17QuietustNew Issue
2012-04-30 10:21QuietustNote Added: 0022403
2016-07-31 07:22LociRelationship addedrelated to 0009946

Notes
(0022403)
Quietust   
2012-04-30 10:21   
I have a save file which individually tests every possible support tile for a 5x5 bridge - 24 bridges were built and linked to single lever (in case digging out their support tiles didn't trigger a collapse), and all appropriate tiles were designated for channeling.

If necessary, I will post it on DFFD.