Differences From Artifact [86e7a28476]:
- File
src/OFDDPSocket.h
— part of check-in
[9c2f20e736]
at
2022-11-03 00:14:27
on branch trunk
— OFDDPSocket: Don't include the type with the data
This seems to be an oddity limited to OSes that have implemented DDP
exclusively for netatalk, while macOS and Windows don't include it with
the data.While on macOS it's possible to achieve the previous behavior via some
hacks, this is impossible on Windows, so the proper approach is to
handle it like everybody else: Specify the protocol type when binding
and only handle packets of the correct protocol type. (user: js, size: 3281) [annotate] [blame] [check-ins using] [more...]
To Artifact [0a178ce3d0]:
- File src/OFDDPSocket.h — part of check-in [b76008cee0] at 2022-12-04 11:11:59 on branch trunk — Fix several Doxygen warnings (user: js, size: 3303) [annotate] [blame] [check-ins using]
| ︙ | ︙ | |||
39 40 41 42 43 44 45 | * @ref OFSocketAddressAppleTalkNode to get the AppleTalk node and * @ref OFSocketAddressAppleTalkPort to get the port (sometimes also called * socket number). * * @note On some systems, packets received with the wrong protocol type just * get filtered by the kernel, however, on other systems, the packet is * queued up and will raise an @ref OFReadFailedException with the | | | 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 | * @ref OFSocketAddressAppleTalkNode to get the AppleTalk node and * @ref OFSocketAddressAppleTalkPort to get the port (sometimes also called * socket number). * * @note On some systems, packets received with the wrong protocol type just * get filtered by the kernel, however, on other systems, the packet is * queued up and will raise an @ref OFReadFailedException with the * @ref OFReadFailedException#errNo set to `ENOMSG` when being received. * * @warning Even though the OFCopying protocol is implemented, it does *not* * return an independent copy of the socket, but instead retains it. * This is so that the socket can be used as a key for a dictionary, * so context can be associated with a socket. Using a socket in more * than one thread at the same time is not thread-safe, even if copy * was called to create one "instance" for every thread! |
| ︙ | ︙ |