ObjFW  History of src/OFSocket.m of 20122a7016c8d4c9

MorphOS has the header, but it's broken and including it makes the
compiler fail. file: [d36d90a400] check-in: [6235989999] user: js, branch: trunk, size: 27529 [annotate] [blame] [check-ins using] [diff]

For some strange reason, Windows does not use big endian here like
everyone else. file: [da4b96c1d5] check-in: [4571eada47] user: js, branch: trunk, size: 25851 [annotate] [blame] [check-ins using] [diff]

While it works for those two protocols, it's a bad precedent as other
protocols have different types for ports. file: [d0ea6d0ea9] check-in: [a54730b88f] user: js, branch: trunk, size: 22894 [annotate] [blame] [check-ins using] [diff]

This is more standard, albeit -[connectToNetwork:node:port:] sounding a
little bit more like you connect to a network rather than a node. file: [b9058d4f4d] check-in: [ccc6a77c55] user: js, branch: trunk, size: 22537 [annotate] [blame] [check-ins using] [diff]

This doesn't work yet (it crashes on recvfrom), so it's not enabled in
configure for now. file: [16116f3ca3] check-in: [a30a0ab261] user: js, branch: nintendo-switch, size: 22537 [annotate] [blame] [check-ins using] [diff]

There is a nasty bug in HP-UX: When using the proper socket APIs that
take a socklen_t (which is 64 bit in LP64 mode), it still calls into
methods that expect an int - so there is a type mismatch between the
headers and the compiled system libraries. This leads to all socklen_t
being interpreted as 0 (due to big endian). Defining
_HPUX_ALT_XOPEN_SOCKET_API makes it use the correct symbols in the
system libraries instead. file: [21df00d3c4] check-in: [1833194867] user: js, branch: trunk, size: 21022 [annotate] [blame] [check-ins using] [diff]

In our tests, we call objc_exit() using atexit(), so the runtime is
already gone by the time any destructors get called. file: [d886e134b6] check-in: [9cedc18bb3] user: js, branch: trunk, size: 21005 [annotate] [blame] [check-ins using] [diff]

errno is problematic for Amiga libraries and is also not thread-safe on
some systems, even though it should. file: [7f38594dbe] check-in: [5b37fbeb82] user: js, branch: trunk, size: 21102 [annotate] [blame] [check-ins using] [diff]

MorphOS allows sharing a single socket base between tasks, so having a
per-task socket base is unnecessary. file: [d03343d8fd] check-in: [b45a563f2b] user: js, branch: trunk, size: 21070 [annotate] [blame] [check-ins using] [diff]

The internal hash is still 32 bit in most places, but this way, it is at
least not baked into the API and ABI and can be upgraded later, should
that ever be necessary. file: [4343bc563f] check-in: [aeb403a1ed] user: js, branch: trunk, size: 20842 [annotate] [blame] [check-ins using] [diff]

This solves the old signed vs. unsigned problem and allows for more
bases than just 8, 10 and 16, as well as auto-detection of the base (if
base is 0). file: [0757bca949] check-in: [b6ee372b98] user: js, branch: trunk, size: 20789 [annotate] [blame] [check-ins using] [diff]

Instead, provide getters and setters for the various address parts. file: [b25bff7c1b] check-in: [2ff4218405] user: js, branch: trunk, size: 20756 [annotate] [blame] [check-ins using] [diff]

Supporting other styles of IPv4 is non-standard and is a BSD legacy. file: [41dfcc5d04] check-in: [66c1192d94] user: js, branch: trunk, size: 17463 [annotate] [blame] [check-ins using] [diff]

This allows the runtime to only link against the parts it needs, without
pulling in unnecessary parts like thread spawning, TLS and conditions. file: [2f5d16033d] check-in: [5358e9ea6a] user: js, branch: trunk, size: 14637 [annotate] [blame] [check-ins using] [diff]

This does not fully fix it yet, but at least the socket tests in the
test suite pass on 3DS now. file: [e7074a55dd] check-in: [ab89c47f42] user: js, branch: trunk, size: 14769 [annotate] [blame] [check-ins using] [diff]

This avoids the entire problem of keeping the buffer alive until the
send finished. file: [b9d1cba50d] check-in: [0f3ae90eef] user: js, branch: trunk, size: 14664 [annotate] [blame] [check-ins using] [diff]

Instead, OFDNSResolver is used for everything now. file: [00380c738c] check-in: [6e9ee122eb] user: js, branch: trunk, size: 14578 [annotate] [blame] [check-ins using] [diff]

This avoids unnecessary string construction and parsing and removes
duplicate code to convert an address into a string. file: [62a31a2fc1] check-in: [75d2d3ba01] user: js, branch: trunk, size: 14194 [annotate] [blame] [check-ins using] [diff]

Instead of containing a struct sockaddr_storage - which does not exist
on all supported platforms - it now contains a union of all struct
sockaddr_* types.

Additionally, if struct sockaddr_in6 does not exist, it is defined so
that an IPv6 can be stored even if the system does not support IPv6. file: [23b837d278] check-in: [8f810ecb7e] user: js, branch: trunk, size: 14200 [annotate] [blame] [check-ins using] [diff]

Not all OSes supported by ObjFW provide them. file: [2650e52f57] check-in: [ab9810ef69] user: js, branch: trunk, size: 13826 [annotate] [blame] [check-ins using] [diff]

This has always been wrong. file: [93c697cb3c] check-in: [e36455441d] user: js, branch: trunk, size: 9786 [annotate] [blame] [check-ins using] [diff]

This avoids shadowing sin(). file: [fac7286c11] check-in: [2f73172fd6] user: js, branch: trunk, size: 9757 [annotate] [blame] [check-ins using] [diff]

This replaces of_address_to_string_and_port(). file: [9e65efeb36] check-in: [8aeee6680a] user: js, branch: trunk, size: 9710 [annotate] [blame] [check-ins using] [diff]

This should make the init behavior the same for --enable-threads and
--disable-threads. file: [d3a737b015] check-in: [885d84a1a7] user: js, branch: trunk, size: 4180 [annotate] [blame] [check-ins using] [diff]

Casts are now written like types in variable declarations. file: [c5f416940a] check-in: [4af49a13c3] user: js, branch: trunk, size: 4104 [annotate] [blame] [check-ins using] [diff]

Forgot to add 2017, even though I already did quite some changes in
2017. file: [798d29602d] check-in: [44f45c2e35] user: js, branch: trunk, size: 4097 [annotate] [blame] [check-ins using] [diff]

This removes the port registry that existed for the Wii and instead
tries binding to random ports in a loop until it succeeds or errno is
not EADDRINUSE. This is done for the Wii and Nintendo 3DS now.

This also includes several other socket-related cleanups. file: [1832892042] check-in: [12b8a7552e] user: js, branch: trunk, size: 4219 [annotate] [blame] [check-ins using] [diff]

While at it, also update the mail address. file: [ac5bad5b02] check-in: [cec0f072f8] user: js, branch: 0.8, size: 3612 [annotate] [blame] [check-ins using] [diff]

While at it, also update the mail address. file: [deff42ec22] check-in: [2a27cf3000] user: js, branch: trunk, size: 5419 [annotate] [blame] [check-ins using] [diff]

This is necessary as the Wii does not allow picking a random free port,
and thus we need to track which ports are used. file: [ebf56cad93] check-in: [ed4549ddd3] user: js, branch: trunk, size: 5411 [annotate] [blame] [check-ins using] [diff]

of_getsockname() calls getsockname() after acquiring a mutex. This is
necessary as some implementations of getsockname() are not thread-safe. file: [a024642fe6] check-in: [f18615e057] user: js, branch: trunk, size: 3600 [annotate] [blame] [check-ins using] [diff]

The old behaviour where the exception would access errno directly on
creation of the exception was very fragile. The two main problems with
it were that sometimes it would pick up an errno even though none had
been set and in other cases that when the exception was created errno
had already been overridden.

This also greatly increases errno handling on Win32, especially in
conjunction with sockets. It can still be improved further, though. file: [f110c9fc59] check-in: [62e2de30b9] user: js, branch: trunk, size: 3012 [annotate] [blame] [check-ins using] [diff]