Comments by "" (@jboss1073) on "Brodie Robertson" channel.

@isodoubIet ""It is physically impossible" It has nothing to do with physics." This is proof that isodoublet is a "word thinker". The phrase "physically impossible" has existed since Ancient Greek to mean "naturally impossible" but for isodoublet the only proper meaning of "physics" is the academic discipline taught in universities!
11
@isodoubIet You seem offended that Rust "lied" to you. Why did you believe them? You should always doubt people know what they're talking about. Instead it seems you believed the "Rust people who first said leaks are unsafe" and then now you're offended or heart-broken that they reneged on their position. Couldn't they simply have been wrong? Why put so much weight on the fact they thought and said this before? So what?
4
@ "Then how do destructors work? Do they overwrite the object in memory with null" The point is not whether they overwrite the object in memory with null - they may or may not. The point is destructors are invoked by explicit frees. It just so happens that C++ has a rule where it will invoke the destructors for stack allocated items at the end of their scope, since this is trivial to know because when the stack unwinds one stack frame, it takes away all stack allocated items in that frame. That does not mean that "destructors are a thing that runs at the end of the scope". That's just what C++ decided to do with stack items. It does not alter the nature of destructors, which are originally triggered only by explicit frees.
3
@isodoubIet "destructors must eventually run when references become unreachable." Are they called "unreachablors" or are they called "destructors"? I'm pretty sure a DESTRUCTor is meant to be run upon DESTRUCTion, not upon UNREACHableness. Talk about moving goalposts.
2
@adiaphoros6842 "So destroying an object is the same as making it unreachable." THINK about what you just said. If something is unreachable, how can it be destroyed? It's just beyond reach, existing undestroyed.
2
No, we pity him.
2
Redis and most in-memory cache systems are nothing but memory leaks. After all, they're not tracking whether you still hold a key to a given value in its in-memory hashtable inside your application code. So memory leaks are a safe operation because they are sometimes necessary: for example, for building in-memory caches. If memory leaks were unsafe, no one would be able to build in-memory caching in that language, as a service to other processes.
2
Cheney on the MTA.
1
How can you code an in-memory Redis-like cache without leaking memory?
1