|Age||Commit message (Collapse)||Author||Files||Lines|
Mainly fixes some mismatched indentation warnings with Ruby 2.6+
Some doc updates, too.
4 changes since v5.1.1 (2017-04-02)
quiet mismatched indentation warnings
t/hijack.ru: avoid shadow warning
t/test_isolate: various version updates
various documentation updates
Point to HTTPS to discourage sniffable traffic.
olddoc and pandoc are only necessary for packaging
gems/updating-the-website, not development.
Update some references to other projects.
Ruby trunk started warning about more mismatched indentations
starting around r62836.
This release fixes an incompatibility introduced with
the release of unicorn 5.3.0, yesterday :x.
Thanks to Claudio Poli for reporting the bug (and reminding
me this project exists).
Two changes since 5.1.0 (2017-01-12):
workaround for unicorn 5.3.0
tests: update isolate deps for unicorn and raindrops
Newer versions were just released, so we should test with
them to prevent mishaps like
from surfacing :x
unicorn 5.3.0 introduced a new Unicorn::TCPSrv and
Unicorn::TCPClient constants to distinguish TCP sockets
from Unix ones from its check_client_connection feature.
These classes screw up our direct inheritance of Kgio::Socket
from the Rainbows::Client class. Since we do not support
check_client_connection, we do not need these new classes in
unicorn. Removing Unicorn::TCPSrv and aliasing it as
Kgio::TCPServer is sufficient for now. In the future,
removing all kgio dependencies from Rainbows! will be prioritized.
Thanks to Claudio Poli for reporting the bug:
This release now supports rack 2.x along with existing rack 1.x
applications. Additionally, we now wait for deferred actions
for EventMachine users on graceful shutdown.
Thanks to Julia López Aladro and <email@example.com> for
bringing up these changes in the open-to-all
firstname.lastname@example.org mailing list:
There's also a few documentation and test fixes.
9 changes since Rainbows! 5.0.0:
doc: bump olddoc to ~> 1.2 for extra NNTP URLs
doc: move homepage and update URLs to HTTPS
doc: Static_Files: remove Future section
tests: use correct file:// URL for random_blob path
support rack 2.x
Revert "t/t0044-autopush.sh: remove test"
tests: re-enable EventMachine tests, again
eventmachine: wait for deferred actions to finish
doc: additional disclaimers
Honestly, these disclaimers should've been there all along;
and should apply to all servers handling multiple clients
within a single process.
Since EventMachine 1.0.0 in 2012, the EM.defers_finish? API
exists to check for the existence of deferred actions.
Support it if it exists in the running version of EM and
update the note in our SIGNALS document.
Thanks to <email@example.com> on the mailing list for bringing
Based on activity in <git://github.com/eventmachine/eventmachine>,
EventMachine seems to be maintained, again, so resume testing our
Stuck between a rock and a hard place..
This reverts commit 5c911ab2470a2ca9c140b3a2e20ae810bf758590
and restores autopush support when using kgio 2.11.0+
Explanations on the kgio mailing list below:
We can drop kgio when we drop Ruby <= 2.2 support in a few years.
This requires updating to unicorn 5.1+ to relax the rack
requirement. Unlike unicorn, Rainbows! is considerably
larger and cannot function without rack installed.
There were no code changes required to pass the test suite;
but some other optional gems needed to be updated for testing.
Support for async_sinatra is no longer tested as that is tied
to rack 1.x; but the remaining EventMachine-dependent parts
Thanks to Julia López Aladro for reminding us on the mailing list:
Apparently RFC1738 designates hostname support for file:// and
curl historically had this wrong, but my current curl version
(7.38.0-4+deb8u5 from Debian 8) is stricter. Anyways, this seems
to handle $PWD with spaces and possibly other strange
characters, so URL encoding here doesn't seem to be strictly
sendfile(2) is fewer syscalls and FDs than splice(2),
and an open-file-cache is unlikely to be worth the complexity.
Let's Encrypt is working well for us and having fewer domains
reduces subjectAltName bloat to speed up connection
HTTP will remain working indefinitely since some old systems
do not have modern TLS stacks.
I forgot to add any NNTP servers to our .olddoc.yml :x
Not advertising additional HTTP-based URLs just yet. They could
contain images/frames/JS/CSS and add unnecessary clutter to the
footer. NNTP puts the client in control of UI.
This release syncs with unicorn 5 and drops some old compatibility
cruft from old releases. Performance should be roughly unchanged
for Ruby 2.2 users while older Rubies (1.9.3 - 2.1) will see
minor, probably unnoticeable performance regressions.
* The horrible, proprietary (:P) "Status:" response header is
finally gone, saving at least 16 precious bytes in every HTTP
response. This should make it easier to write custom HTTP clients
which are compatible across all HTTP servers. It will hopefully
make migrating between different Rack servers easier for new
* Ruby 1.8 support removed. Ruby 1.9.3 is currently the earliest
supported version. However, expect minor, likely-unnoticeable
performance regressions if you use Ruby 2.1 or earlier. Going
forward, Rainbows! will favor the latest version (currently 2.2) of
the mainline Ruby implementation, potentially sacrificing
performance on older Rubies.
* sd_listen_fds(3) emulation added for systemd compatibility.
You may now stop using PID files and other process monitoring
software when using systemd.
* Newly-set TCP socket options are now applied to inherited sockets.
* Dynamic changes in the application to Rack::Utils::HTTP_STATUS
hash is now supported; allowing users to set custom status lines
in Rack to be reflected in unicorn. This feature causes a minor
performance regression, but is made up for Ruby 2.2 users with
* The monotonic clock is used under Ruby 2.1+, making the
timeout feature immune to system clock changes.
As Rainbows! may be used anonymously without registration, the
project is committed to supporting anonymous and pseudonymous
help requests, contributions and feedback via plain-text mail to:
The mail submission port (587) is open to those behind firewalls
and allows access via Tor and anonymous remailers.
Archives are accessible via:
and mirrored to various other places, so you do not even need
to use a valid address when posting.
18 changes since Rainbows! 4.7.0
README: remove Zbatery references
http_parser: handle keepalive_requests internally
kill the moronic Status: header
reflect changes in Rack::Utils::HTTP_STATUS_CODES
reduce constant lookup dependencies
http_parser: workaround hijack changes in unicorn 5
http_server: add master_pid attribute
stream_response_epoll: remove hijack_prepare call
bump to unicorn 5.0.1, use monotonic clock
add .gitattributes for Ruby method detection
response: avoid garbage string entirely
tiny bytecode reductions for cold paths
Ruby 1.9.3+-only cleanups
revactor: remove fcntl dependency
response: simplify regexp
t0105: fix test reliability
fix Rainbows.now definition for old Rubies
fix broken constant lookups in unmaintained bits
In case Revactor starts being maintained again...
(heck, it was probably the reason I started Rainbows! in the first
Our ReverseProxy stuff was never complete nor marketed;
and probably not worth keeping around at all.
Oops, blind substitution :x
It's possible curl will be sending to us and detect an error in the
send before it has a chance to read our (premature) 413 response.
Of course, we cannot afford to read an entire request when returning
a premature 413 response because we risk wasting bandwidth that way.
Redundant \z statements are ugly and wastes 4 bytes on x86-64
according to ObjectSpace.memsize_of
In the unlikely case somebody runs revactor, they won't need to
load the extra fcntl.so library into their process anymore.
In retrospect, we could've alway used IO#close_on_exec= since
it appeared in 1.9.1 and (IIRC) revactor always required 1.9.1+
unicorn 5 will only support Ruby 1.9.3 and later, so remove
some checks for Hash#compare_by_identity and IO.copy_stream
which we know exist in Ruby 1.9.
Favor &:sym proc dispatch to avoid unnecessary captures and
bytecode size increases, too.
Finally, ensure we fail fast by converting some literal
hashes to use non-arrow syntax for symbolic keys.
Less code in cold paths can improve speed for hot paths.
Single-byte strings for String#split is optimized in mainline Ruby,
so it's not actually a performance loss for sendfile_range in
Regexps are at least 400 bytes each, so prefer non-Regexps
if possible, especially for cold sites where performance does
Even with frozen string literals enabled in Ruby 2.3.0dev,
dstrings still create garbage as the optimizer is
not yet smart enough to optimize it despite the limited
choice of internals being known.
Maybe in the future Ruby will be smart enough, but not yet...
The "diff" function detection for C does not map well to
Ruby files, take advantage of gitattributes(5) to improve
method name detection in generated patches as well as
making "git diff -W" output more useful.
The timeout (mis)feature in unicorn uses the monotonic clock
if available. We must follow suit to avoid having our timeout
functionality completely broken.
unicorn 5.0.0 assumes Rack hijack is always available if
the application tries to use it, so the wrapper method
We'll need this for unicorn 5, as they dropped this publically
accessible attribute while retaining the ivar. Eventually we
may not have to check this attribute at all, instead detecting
parent death via worker pipe.
unicorn lost the hijack_setup method in version 5,
so we must recreate it ourselves.
Unicorn 5 removes some constants we were using, and constant
lookups + inline caching are waste of time anyways on newer
Rubies with the opt_str_freeze bytecode instruction.
This may reduce performance for folks on older Rubies (probably
not noticeable); but improves performance for folks on newer
Applications may want to alter the message associated with HTTP
status codes in Rack::Utils::HTTP_STATUS_CODES. Avoid memoizing
status lines ahead-of-time
Note: this introduces a minor performance regression, but ought to
be unnoticeable unless you're running "Hello world"-type apps.
Whatever compatibility reasons which existed in 2009 likely do not exist
now. Other servers (e.g. thin, puma) seem to work alright without it,
so there's no reason to waste precious bytes.
unicorn 5 dropped support for this, essentially allowing unlimited
persistent connections if we used the parser as-is.
Since most of our concurrency models cannot handle infinite
persistent connections without being vulnerable to DoS,
we must support keepalive_requests like nginx does.
yahns runs as a single process just fine.
This release includes fixes for upcoming changes in Ruby 2.3.0
(due December 2015). Use of Rainbows! for new projects is not
recommended, try other servers instead.
* update dependencies for Ruby 2.2.0dev
* switch docs + website to olddoc
* gemspec: fix bad reference to rdoc_options
* README: reference yahns
* build: fix quoting issue with double parens
* response: avoid unnecessary args to IO.copy_stream
* t/close-has-env.ru: ensure close is idempotent
* sync_close: This fix breakage from Ruby-trunk r50118
* t/t0044-autopush.sh: remove test
* t/test_isolate.rb: updates for various gem versions
* response: convert source arg to path before IO.copy_stream
* speed up QUIT for users of the unicorn worker_loop
* gemspec: use SPDX-compatible license for GPL-2.0+
Current versions of RubyGems still complains about the '+'
is valid according to: http://spdx.org/licenses
WriterThreadPool, WriterThreadSpawn, Base, and potentially
other rarely-used concurrency options experienced slow shutdowns
due to the destruction of a listener socket failing to wake up
a thread in IO.select.
Send ourselves a signal to interrupt the IO.select call in
the main thread to force the wakeup.
This will allow us use the sendfile syscall under Linux on
Ruby which favor #read/#readpartial methods for non-IO objects.
This also allows us to revert changes made in
("sync_close: This fix breakage from Ruby-trunk r50118")
kgio no longer does anything with autopush, so this test is
broken. Autopush was overkill and badly done (MSG_MORE is
better on Linux, FreeBSD should copy it :P).
By providing read+readpartial methods
IO.copy_stream behaves slightly differently when operating on
non-IO-subclassed objects nowadays.
> * io.c (copy_stream_body): use the arguments without conversion if
> having read, readpartial, and write methods, than conversion by
> to_path method. [ruby-core:68676] [Bug #11015]
Ruby 2.3.0 will have idempotent IO#close, so closing an IO
twice will be idempotent and not raise an IOError exception.
Ensure we do not rely on the IOError exception to catch
our own errors.
Not all responses are seekable, so do not attempt to pass seek
arguments to them since Ruby may attempt to seek (and fail!).
Apparently GNU make parses that strangely and mangles it for the
Because nobody has time to read about all the options Rainbows!
provides. yahns is basically XEpollThreadPool, with minor
improvements which weren't easily supportable with other
olddoc does not need or use this
wrongdoc was difficult to maintain because of the tidy-ffi
dependency and the HTML5 changes in Darkfish could not be
handled well by Tidy.
olddoc is superior as it generates leaner HTML which loads faster,
requires less scrolling and less processing power to render.
Aesthetic comparisons are subjective of course but completely
unimportant compared to speed and accessibility.
The presence of images and CSS on the old (Darkfish-based) site
probably set unreasonable expectations as to my ability and
willingness to view such things. No more, the new website is
entirely simple HTML which renders well with even the wimpiest
browser (hell, olddoc even tries to generate readable raw HTML).