Fix wrong assert. #2547
Fix wrong assert. #2547
Conversation
can trigger an abort when compiling with debug=yes, and using 7men TB. The assert should check that less than 8 pieces are in the key for each side. Also take explicitly care of a 'v' character in material strings. Fixes an issue reported in the forum: https://groups.google.com/d/msg/fishcooking/yoVC7etIpz0/7mS7ntZMBAAJ No functional change.
|
Can't we just throw away the "v" character and leave the assert as it was? Less than 8 pieces for each side means assert will be true for |
|
I don't think so, the original assert is limiting the total number of pieces to 7, while there is no reason for that as far as I can tell. It would trigger as soon as the first 8men TB files become available (or if we have 8men specialized endgame routines). |
|
@vondele However, 8-man TB files seem too far away in the future to care about imho. I would eventually add a note in the comment, but only deal with what we currently have (max. 7-man tables/endgames). |
|
@joergoster I agree that complete 8men TB are not so likely, but one could generate a few. More importantly, the asserts now reflect the assumptions of the implementation. So, I'll go ahead and merge this version. |
can trigger an abort when compiling with debug=yes, and using 7men TB. The assert should check that less than 8 pieces are in the key for each side, matching the assumption that underlies the FEN string construction. Also take explicitly care of a 'v' character in material strings. Fixes an issue reported in the forum: https://groups.google.com/d/msg/fishcooking/yoVC7etIpz0/7mS7ntZMBAAJ closes official-stockfish#2547 No functional change.
can trigger an abort when compiling with debug=yes, and using 7men TB.
The assert should check that less than 8 pieces are in the key for each
side. Also take explicitly care of a 'v' character in material strings.
Fixes an issue reported in the forum:
https://groups.google.com/d/msg/fishcooking/yoVC7etIpz0/7mS7ntZMBAAJ
No functional change.