That's certainly my goal. The current patches on the KI page will eliminate all min/max macro defines in header files. There are still a few min/max macros local to .cpp files, but that's certainly less egregious (though still bad.)
I was planning to post a patch for the removal of the last of the clamp macros once I get the SDK update fully working (linking on Linux). At 2006/08/26 05:08 PM, Aaron Schiff wrote: >-- >[ Picked text/plain from multipart/alternative ] >Shouldn't we just rather remove the defines from the code all together? > >On 8/26/06, [EMAIL PROTECTED] <[EMAIL PROTECTED]> >wrote: >> >> Thanks, I appreciate the sentiment. One thing I learned long ago doing >> software development is that doing something right takes a lot less time in >> the long run than trying to ignore the problem, or covering it up with a >> quick hack. >> >> Ie, when a bug is found in the open-source side of the SDK, the best thing >> to do is to root cause it and solicit community feedback on the fix. The >> alternative is to float about in a soup of bugs - you never know where >> problems are coming from. (Unfortunately, the most time-consuming issues >> I've hit are with the closed-source side of things, such as the infamous >> vphysics bugs, and the recent missing .lib symbols on Linux, but that's an >> unrelated issue.) >> >> Specifically in regards to asserts, I glanced at a post on this list >> recently where someone advised ignoring asserts, and that's >> unfortunate. Asserts, used correctly, are a powerful tool. I get nervous >> when I see a page of code without any asserts. Valve's software is light on >> asserts as it is, so it's best not to ignore the ones that are there. >> >> -bk >> >> At 2006/08/26 02:48 PM, Nick wrote: >> >Valve should pay you, I can't imagine how long you must have spent >> >trying to fix some of these bugs. Or why anyone would ever spend so >> >much time fixing such bugs? >> > >> >On 8/26/06, [EMAIL PROTECTED] >> ><[EMAIL PROTECTED]> wrote: >> >>In debug mode after the SDK update, and the HL2 server asserted before >> it even loaded the map: >> >> >> >>ai_activity.cpp (41) : Assertion Failed: actID >= LAST_SHARED_ACTIVITY >> || actID == lastActID + 1 || actID == ACT_INVALID >> >> >> >>Looks like the "portal" things are missing from ai_activity.cpp so I >> added this KI which fixed the issue: >> >> >> >> >> http://developer.valvesoftware.com/wiki/SDK_Known_Issues_List#AI_Activity.cpp_assert_on_startup >> >> >> >> >> >>Confirmation from Valve that this is the right fix for the bug would be >> appreciated. >> >> _______________________________________________ >> To unsubscribe, edit your list preferences, or view the list archives, >> please visit: >> http://list.valvesoftware.com/mailman/listinfo/hlcoders >> >> > > >-- >ts2do >-- > >_______________________________________________ >To unsubscribe, edit your list preferences, or view the list archives, please >visit: >http://list.valvesoftware.com/mailman/listinfo/hlcoders _______________________________________________ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlcoders
