ObjFW  History of src/OFMutableDictionary.h of bcff1dfd70c5e7d9

This is only migrating some places, others will be migrated as they are
touched. file: [d0bf1761a1] check-in: [a0fd103a0b] user: js, branch: trunk, size: 3658 [annotate] [blame] [check-ins using] [diff]

Mostly because Doxygen fails to find references to things that *do*
exist. file: [ceac8e2f46] check-in: [240eccca97] user: js, branch: trunk, size: 3755 [annotate] [blame] [check-ins using] [diff]

This allows ABI stability with the fragile ABI. file: [e6550bad92] check-in: [496695d778] user: js, branch: trunk, size: 3788 [annotate] [blame] [check-ins using] [diff]

-[setObject:forSubscriptedKey:] is supposed to support removing a key by
setting it to nil. file: [4bdbc2aa53] check-in: [bb8de03577] user: js, branch: trunk, size: 3565 [annotate] [blame] [check-ins using] [diff]

Instead, explicitly declare them, as OF_ASSUME_NONNULL_{BEGIN,END} does
not apply to implicit return types. This means that after this commit,
all init methods have a nonnull return type, as they should have. file: [432341193d] check-in: [2f4e0df8be] user: js, branch: trunk, size: 3222 [annotate] [blame] [check-ins using] [diff]

Casts are now written like types in variable declarations. file: [50af542eae] check-in: [4af49a13c3] user: js, branch: trunk, size: 3002 [annotate] [blame] [check-ins using] [diff]

Forgot to add 2017, even though I already did quite some changes in
2017. file: [98bf6cee07] check-in: [44f45c2e35] user: js, branch: trunk, size: 3004 [annotate] [blame] [check-ins using] [diff]

If the key starts with an @, the @ is stripped and the super method is
called. Otherwise, this is equivalent to -[objectForKey:] /
-[setValue:forKey:]. file: [549070ec8e] check-in: [cbacea7ca3] user: js, branch: trunk, size: 3374 [annotate] [blame] [check-ins using] [diff]

While at it, also update the mail address. file: [df79bdb41b] check-in: [cec0f072f8] user: js, branch: 0.8, size: 3000 [annotate] [blame] [check-ins using] [diff]

While at it, also update the mail address. file: [d87ccc345a] check-in: [2a27cf3000] user: js, branch: trunk, size: 2998 [annotate] [blame] [check-ins using] [diff]

Now that Clang switched to use _Nullable and _Nonnull instead of
__nullable / __nonnull, there is no longer a conflict with glibc, which
means we can just define _Nullable / _Nonnull to nothing if they are not
understood by the compiler (which did not work with __nullable /
__nonnull due to this conflict).

This also defines _Null_unspecified to nothing if unavailable. file: [cf03a53ce5] check-in: [d526d938d7] user: js, branch: trunk, size: 2994 [annotate] [blame] [check-ins using] [diff]

Using __nonnull directly doesn't work on systems using glibc, as glibc
defines __nonnull as a parameterized define. While this does not fix the
problem of Clang introducing __nonnull even though it conflicts with
glibc, this at least means it's possible again to compile things with
versions of Clang that don't support __nonnull on systems with glibc. file: [20d6c7fa26] check-in: [b7097a67b6] user: js, branch: trunk, size: 2996 [annotate] [blame] [check-ins using] [diff]

This also removes "bool *stop" from some blocks where it does not make
sense. file: [478ab9b477] check-in: [e8473b9db8] user: js, branch: trunk, size: 2465 [annotate] [blame] [check-ins using] [diff]

The only places where BOOL is left are those where they are required by
the ABI. file: [87d84a54c5] check-in: [c5ef582958] user: js, branch: trunk, size: 2250 [annotate] [blame] [check-ins using] [diff]

This was already implemented in OFDictionary_hashtable for internal
usage and is now publicly available. file: [2e2922b971] check-in: [4fb0f4bf66] user: js, branch: trunk, size: 2253 [annotate] [blame] [check-ins using] [diff]

We assume now that every object understands retain, release, etc. so
that we can weaken the type from OFObject* to id. This makes it
possible to use different root object classes. file: [fdc610a5c4] check-in: [e2b06423e0] user: js, branch: trunk, size: 1166 [annotate] [blame] [check-ins using] [diff]

The new implementation is easier to use as it does automatic resizing,
but therefore it's not realtime-capable anymore. The new implementation
should also be a little bit faster.

I decided to change the implementation as only very few need a
realtime-capable dictionary and those few will most likely write their
own implementation for their specific case anyway.

As the new implementation no longer uses OFList, this also made it
possible to remove a hack from OFList. file: [21cc3f3819] check-in: [bbf1f79b8f] user: js, branch: trunk, size: 831 [annotate] [blame] [check-ins using] [diff]