Open sobolevnrm opened 4 years ago
The most likely modernization route here is just to delete the library.
This issue is being developed in branch remove-maloc-dev
@sobolevnrm In the branch mentioned above, I've removed maloc as a dependency. I haven't changed too much, just copied five header/source pairs into another subdirectory of src
. Would you like to iteratively replace the remnants of maloc as well or are we content with this solution?
CC @intendo
I don't think we need to rewrite MALOC. If native functionality is available to replace MALOC functionality, then that's all we need.
So would you like to leave a few components of maloc in the repo like in remove-maloc-dev
branch? If we go the native functionality route, we'd probably want to refactor pieces of apbs source into C++. Biggest MALOC component used is Vio/socket stuff I think, we could replace all that with std::string. APBS/MALOC sources seem compile-able in C++ already anyways, but you know better than I if it would be too invasive to convert to c++.
I think it's best to replace whatever we can with native functionality and then see what's left?
On Sat, Oct 10, 2020 at 10:00 AM Asher Mancinelli notifications@github.com wrote:
So would you like to leave a few components of maloc in the repo like in remove-maloc-dev branch? If we go the native functionality route, we'd probably want to refactor pieces of apbs source into C++. Biggest MALOC component used is Vio/socket stuff I think, we could replace all that with std::string. APBS/MALOC sources seem compile-able in C++ already anyways.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/Electrostatics/apbs/issues/23#issuecomment-706579761, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAOX7WFFQJXAAQWGRVEHB33SKCHJZANCNFSM4OQBZVXQ .
It's no longer needed and takes extra effort to maintain.