Oops I missed important changes from
https://github.com/grpc/grpc/pull/32712. And it turned out that there
are two problems that I couldn't fix at this point.
- Windows Bazel RBE Linker Error: This may be caused by how new Bazel 6
invokes build tools chain but it's not clear. I put workaround to use
Bazel 5 by using `OVERRIDE_BAZEL_VERSION=5.4.1`
- Rule `rules_pods` to fetch CronetFramework from CocoaPod has
incompatibility with sort of built-in apple toolchain.
(https://github.com/bazel-xcode/PodToBUILD/issues/232): I couldn't find
a workaround to fix this so I ended up disabling all tests depending
this target.
<!--
If you know who should review your pull request, please assign it to
that
person, otherwise the pull request would get assigned randomly.
If your pull request is for a specific language, please add the
appropriate
lang label.
-->
I'm opening this PR to (hopefully!) stimulate a discussion. In brief,
I'd like to amend the gRPC-Web protocol docs to encourage
implementations to follow HTTP semantics and compare HTTP field names
case-insensitively.
The gRPC-Web specification is a nicely-designed way for proxies to
expose standard HTTP/2 gRPC servers to clients using less
tightly-controlled HTTP stacks, such as web browsers. To serve that
goal, it seems valuable to have the gRPC-Web specification follow [RFC
9110 (HTTP
Semantics)](https://www.rfc-editor.org/rfc/rfc9110.html#name-field-names).
Like previous RFCs, 9110 specifies that "field names are
case-insensitive." However, the current gRPC-Web specification requires
that servers and proxies "use lower-case header/trailer names" on the
wire. In principle, mandating casing on the wire is normal for HTTP/2
and fine (if unusual) for HTTP/1.1; however, it encourages
implementations to violate HTTP semantics and require lower-case names
when _reading_ headers and trailers.
I'd like to loosen the gRPC-Web specification to permit any casing on
the wire for HTTP/1.1. I'd also like to emphasize that gRPC-Web
implementations ought to follow standard HTTP semantics when _reading_
fields and compare names case-insensitively. Implementations that can't
treat names case-insensitively without breaking backward compatibility
should instead normalize field names to lowercase. Among the
Google-maintained gRPC implementations, at least `grpc-go` and
`grpc-java` already compare names case-insensitively (even though
they're HTTP/2-only). `grpc-dart` does the opposite and compares names
case-sensitively. `grpc-web` is sometimes case-insensitive (when reading
`grpc-status` and `grpc-message` from trailers-only responses) and
sometimes case-sensitive (when hand-parsing a block of length-prefixed
trailers).
The proposed amendment does not affect the correctness of Envoy (which
may continue to use lower-case field names). It partially affects
`grpc-web`, which would require a small patch to always normalize names.
(Both patched and unpatched versions of `grpc-web` would work with
Envoy.) `grpc-dart` would need to either begin treating field names
case-insensitively or normalize names, depending on what's possible in
Dart without breaking backward compatibility.
Relates to https://github.com/improbable-eng/grpc-web/issues/228,
https://github.com/bufbuild/connect-go/issues/453/, and
https://github.com/grpc/grpc-dart/issues/594.
cc @markdroth
<!--
If you know who should review your pull request, please assign it to
that
person, otherwise the pull request would get assigned randomly.
If your pull request is for a specific language, please add the
appropriate
lang label.
-->
* WRR: port StaticStrideScheduler to OSS
* WIP
* Automated change: Fix sanity tests
* fix build
* remove unused aliases
* fix another type mismatch
* remove unnecessary include
* move benchmarks to their own file, and don't run it on windows
* Automated change: Fix sanity tests
* add OOB reporting
* generate_projects
* clang-format
* add config parser test
* clang-tidy and minimize lock contention
* add config defaults
* add oob_reporting_period config field and add basic test
* Automated change: Fix sanity tests
* fix test
* change test to use basic RR
* WIP: started exposing peer address to LB policy API
* first WRR test passing!
* small cleanup
* port RR fix to WRR
* test helper refactoring
* more test helper refactoring
* WIP: trying to fix test to have the right weights
* more WIP -- need to make pickers DualRefCounted
* fix timer ref handling and get tests working
* clang-format
* iwyu and generate_projects
* fix build
* add test for OOB reporting
* keep only READY subchannels in the picker
* add file missed in a previous commit
* fix sanity
* iwyu
* add weight expiration period
* add tests for weight update period and OOB reporting period
* Automated change: Fix sanity tests
* lower bound for timer interval
* consistently apply grpc_test_slowdown_factor()
* cache time in test
* add blackout_period tests
* avoid some unnecessary copies
* clang-format
* add field to config test
* simplify orca watcher tracking
* attempt to fix build
* iwyu
* generate_projects
* add "_experimental" suffix to policy name
* WRR: update tests to cover qps plumbing
* WIP
* more WIP
* basic WRR e2e test working
* add OOB test
* fix sanity
* ignore duplicate addresses
* Automated change: Fix sanity tests
* add new tracer to doc/environment_variables.md
* retain scheduler state across pickers
* Automated change: Fix sanity tests
* use separate mutexes for scheduler and timer
* sort addresses to avoid index churn
* remove fetch_sub for wrap around in RR case
Co-authored-by: markdroth <markdroth@users.noreply.github.com>
Mark D. Roth
Yash Tibrewal
Esun Kim
Eugene Ostroukhov
Stanley Cheung
Terry Wilson
Akshay Shah
Terry Wilson
Sergii Tkachenko
AJ Heller
Craig Tiller
Jan Tattermusch
Clément Jean
Richard Belleville
Cheng-Yu Chung
tony
Michael Lumish
apolcyn
ktr
Jundol
yifeizhuang
stdll00