Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
_mem
#14
steve said what he said in the spirit of "basic does it all for me because otherwise i hate programming in c and doing it all by myself".  like what was said about "close" and variable-length strings.  many basic programmers don't think about those two.  but because the veteran has to think about _mem, he avoids it.  i was like this for a long time.  in part because there used to be a bug in qb64 (before phoenix edition) that just disallowed me to use _memcopy or otherwise use a simple addition between an integer and an _offset.

but there cannot be a "garbage collector" for everything.  such a thing tends to make the programmer lazy and affects program performance.  it's a penalty one is willing to pay for using strings in basic all the time.  but hopefully the _mem gang would never be used even more than vls's.

watch out: might also require whatever was created with redim to be matched with erase before the program ends.

the qb64 help set a good precedent, in all keywords _mem and related ones.  saying the _mem variables should be freed with _memfree.
Reply


Messages In This Thread
_mem - by Jack - 12-31-2024, 09:52 PM
RE: _mem - by DSMan195276 - 01-01-2025, 07:11 AM
RE: _mem - by Jack - 01-01-2025, 09:08 AM
RE: _mem - by Jack - 01-01-2025, 10:01 AM
RE: _mem - by SMcNeill - 01-01-2025, 10:18 AM
RE: _mem - by SMcNeill - 01-01-2025, 10:10 AM
RE: _mem - by Jack - 01-01-2025, 10:27 AM
RE: _mem - by SMcNeill - 01-01-2025, 10:36 AM
RE: _mem - by Jack - 01-01-2025, 10:40 AM
RE: _mem - by SMcNeill - 01-01-2025, 10:45 AM
RE: _mem - by Kernelpanic - 01-01-2025, 11:55 PM
RE: _mem - by DSMan195276 - 01-01-2025, 11:13 PM
RE: _mem - by DSMan195276 - 01-02-2025, 12:27 AM
RE: _mem - by hsiangch_ong - Today, 01:50 AM



Users browsing this thread: 1 Guest(s)