Geoffrey Hutchison wrote:
> - While v2.3 will obviously have new API, I think we should make 
 > every effort not to break ABI -- the binary compatibility. I have
 > not yet tested this, but the main restrictions are not to add
 > overloaded parameters to function calls, new virtual functions, etc.
> http://techbase.kde.org/Policies/Binary_Compatibility_Issues_With_C++

I wonder whether binary compatibility is attainable for version 2.3. 
As I understand it, it also means no new non-static data members. I 
see OBAtom already has a new unsigned long _id, which would be a 
tragedy to remove again.

Wouldn't source compatibility, with external programs needing to be 
re-compiled and linked but not re-written, be a more practical goal, 
even if it is not as desirable or prestigious? Leave the more 
difficult and restrictive ABI to the most minor releases. Doing it at 
higher levels would have been more feasible if the code had been 
originally structured with this goal in mind.

Chris



------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
OpenBabel-Devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/openbabel-devel

Reply via email to