[Freeciv-Dev] Re: (PR#12559) [Patch] is_ocean() -> can_unit_exist_at_tile()

["Jason Short" <jdorje@xxxxxxxxxxxxxxxxxxxxx>]

<URL: http://bugs.freeciv.org/Ticket/Display.html?id=12559 >

Marko Lindqvist wrote:

>>       if (ptile->city
>>           || TEST_BIT(target->ai.hunted, pplayer->player_no)
>>-          || (!is_ocean(ptile->terrain) && is_sailing_unit(punit))
>>-          || (is_ocean(ptile->terrain) && is_ground_unit(punit))
>>+       || !can_unit_exist_at_tile(punit, ptile)
>>           || (!is_sailing_unit(target) && is_sailing_unit(punit))
>>           || (is_sailing_unit(target) && !is_sailing_unit(punit))
>>           || !goto_is_sane(punit, target->tile, TRUE)) {
>>         /* Can't hunt this one. */
>>+     /* FIXME: this prevents airplanes from hunting sailing units. */
>>         continue;
> 
>   No, it doesn't. punit is hunter, not target. Air unit can exist at any 
> tile.

The "problem" isn't in your change but in the is_sailing_unit checks.

>>--- ai/aiunit.c       21 Mar 2005 12:28:00 -0000      1.347
>>+++ ai/aiunit.c       22 Mar 2005 04:23:13 -0000
>>@@ -1745,7 +1745,7 @@
>>     struct tile *ftile;
>> 
>>     if ((pc = dist_nearest_city(pplayer, punit->tile, FALSE, TRUE))) {
>>-      if (!is_ocean(map_get_terrain(punit->tile))) {
>>+      if (can_unit_exist_at_tile(punit, punit->tile)) {
>>         UNIT_LOG(LOG_DEBUG, punit, "Barbarian marching to conquer %s", 
>> pc->name);
>>         (void) ai_gothere(pplayer, punit, pc->tile);
>>       } else {
> 
> 
>   It might be possible that SEA_MOVING unit sometimes ends here. But 
> then again, calling ai_gothere() for them (in these situations) might be 
> ok...

I don't understand this.

-jason