[new release] eio_main, eio_luv, eio_linux and eio (0.7)

[talex5]

Effect-based direct-style IO mainloop for OCaml

• Project page: https://github.com/ocaml-multicore/eio
• Documentation: https://ocaml-multicore.github.io/eio/

CHANGES:

API changes:

• Unify IO errors as Eio.Io (@talex5 Unify IO errors as Eio.Io ocaml-multicore/eio#378).
  This makes it easy to catch and log all IO errors if desired.
  The exception payload gives the type and can be used for matching specific errors.
  It also allows attaching extra information to exceptions, and various functions were updated to do this.

• Add Time.Mono for monotonic clocks (@bikallem @talex5 Add Time.Mono for monotonic clocks ocaml-multicore/eio#338).
  Using the system clock for timeouts, etc can fail if the system time is changed during the wait.

• Allow datagram sockets to be created without a source address (@bikallem @haesbaert Allow datagram_socket to be created without address ocaml-multicore/eio#360).
  The kernel will allocate an address in this case.
  You can also now control the reuse_addr and reuse_port options.

• Add File.stat and improve Path.load (@haesbaert @talex5 Expose File.stat and improve Path.load ocaml-multicore/eio#339).
  Path.load now uses the file size as the initial buffer size.

• Add Eio_unix.pipe (@patricoferris Add pipe to eio_unix ocaml-multicore/eio#350).
  This replaces Eio_linux.pipe.

• Avoid short reads from getrandom(2) (@haesbaert Fix getrandom(2) ocaml-multicore/eio#344).
  Guards against buggy user code that might not handle this correctly.

• Rename Flow.read to Flow.single_read (@talex5 Rename Flow.read to Flow.single_read ocaml-multicore/eio#353).
  This is a low-level function and it is easy to use it incorrectly by ignoring the possibility of short reads.

Bug fixes:

• Eio_luv: Fix non-tail-recursive continue (@talex5 Unify IO errors as Eio.Io ocaml-multicore/eio#378).
  Affects the Socket_of_fd and Socketpair effects.

• Eio_linux: UDP sockets were not created close-on-exec (@talex5 Allow datagram_socket to be created without address ocaml-multicore/eio#360).

• Eio_linux: work around io_uring non-blocking bug (@haesbaert Fix pipe hang (issue #319) ocaml-multicore/eio#327 Add Eio_unix.pipe workaround, zap Eio_linux.pipe ocaml-multicore/eio#355).
  The proper fix should be in Linux 6.1.

• Eio_mock.Backend: preserve backtraces from main (@talex5 Eio_mock.Backend: preserve backtraces from main ocaml-multicore/eio#349).

• Don't lose backtrace in Switch.run_internal (@talex5 Don't lose backtrace in Switch.run_internal ocaml-multicore/eio#369).

Documentation:

• Use a proper HTTP response in the README example (@talex5 Use a proper HTTP response in the README example ocaml-multicore/eio#377).

• Document that read_dir excludes "." and ".." (@talex5 Document that read_dir excludes '.' and '..' ocaml-multicore/eio#379).

• Warn about both operations succeeding in Fiber.first (@talex5 Warn about both operations succeeding in Fiber.first ocaml-multicore/eio#358, reported by @iitalics).

• Update README for OCaml 5.0.0~beta2 (@talex5 Update README for OCaml 5.0.0~beta2 ocaml-multicore/eio#375).

Backend-specific changes:

• Eio_luv: add low-level process support (@patricoferris eio_luv: add low-level process support ocaml-multicore/eio#359).
  A future release will add Eio_linux support and a cross-platform API for this.

• Expose Eio_luv.Low_level.Stream.write (@patricoferris eio_luv: add low-level process support ocaml-multicore/eio#359).

• Expose Eio_luv.Low_level.get_loop (@talex5 Expose Eio_luv.Low_level.get_loop ocaml-multicore/eio#371).
  This is needed if you want to create resources directly and then use them with Eio_luv.

• Eio_linux.Low_level.openfile is gone (@talex5 Unify IO errors as Eio.Io ocaml-multicore/eio#378).
  It was just left-over test code.

[talex5]

The mac builds are failing with:

fatal: Unable to create '/private/var/lib/ocluster-worker/git/opam-repository.git-b322f7b18d276db68487afc3dfc5a5befdd45818/.git/index.lock': File exists.

However, the tests will likely fail there anyway due to the sandboxing.

I also added an upper-bound for cohttp-eio, which then failed with the old version for an unrelated reason ("ncat: not found").

[kit-ty-kate]

Thanks