aboutsummaryrefslogtreecommitdiffhomepage
path: root/src/ruby/lib
diff options
context:
space:
mode:
authorGravatar Jan Tattermusch <jtattermusch@users.noreply.github.com>2015-04-20 13:16:00 -0700
committerGravatar Jan Tattermusch <jtattermusch@users.noreply.github.com>2015-04-20 13:16:00 -0700
commit291800b78f73ad974a4e472555ac4845a2b32aa0 (patch)
tree423fe665a68350b46f8872a89628678702af922a /src/ruby/lib
parent3bb40596c5303d7a7055504e20a53c465c0ac8e4 (diff)
parenta80aa7d86a1aa1ae64780c23341fecfb06fac640 (diff)
Merge pull request #1309 from tbetbetbe/grpc_ruby_rpc_server_md
Grpc ruby add support for returning metadata to the rpc server
Diffstat (limited to 'src/ruby/lib')
-rw-r--r--src/ruby/lib/grpc/errors.rb7
-rw-r--r--src/ruby/lib/grpc/generic/active_call.rb34
-rw-r--r--src/ruby/lib/grpc/generic/rpc_desc.rb12
-rw-r--r--src/ruby/lib/grpc/generic/rpc_server.rb334
4 files changed, 216 insertions, 171 deletions
diff --git a/src/ruby/lib/grpc/errors.rb b/src/ruby/lib/grpc/errors.rb
index 35e9c02a94..f1201c1704 100644
--- a/src/ruby/lib/grpc/errors.rb
+++ b/src/ruby/lib/grpc/errors.rb
@@ -36,14 +36,15 @@ module GRPC
# error should be returned to the other end of a GRPC connection; when
# caught it means that this end received a status error.
class BadStatus < StandardError
- attr_reader :code, :details
+ attr_reader :code, :details, :metadata
# @param code [Numeric] the status code
# @param details [String] the details of the exception
- def initialize(code, details = 'unknown cause')
+ def initialize(code, details = 'unknown cause', **kw)
super("#{code}:#{details}")
@code = code
@details = details
+ @metadata = kw
end
# Converts the exception to a GRPC::Status for use in the networking
@@ -51,7 +52,7 @@ module GRPC
#
# @return [Status] with the same code and details
def to_status
- Status.new(code, details)
+ Struct::Status.new(code, details, @metadata)
end
end
diff --git a/src/ruby/lib/grpc/generic/active_call.rb b/src/ruby/lib/grpc/generic/active_call.rb
index 8d63de4145..43ba549905 100644
--- a/src/ruby/lib/grpc/generic/active_call.rb
+++ b/src/ruby/lib/grpc/generic/active_call.rb
@@ -39,7 +39,10 @@ class Struct
return nil if status.nil?
fail GRPC::Cancelled if status.code == GRPC::Core::StatusCodes::CANCELLED
if status.code != GRPC::Core::StatusCodes::OK
- fail GRPC::BadStatus.new(status.code, status.details)
+ # raise BadStatus, propagating the metadata if present.
+ md = status.metadata
+ with_sym_keys = Hash[md.each_pair.collect { |x, y| [x.to_sym, y] }]
+ fail GRPC::BadStatus.new(status.code, status.details, **with_sym_keys)
end
status
end
@@ -119,6 +122,12 @@ module GRPC
@metadata_tag = metadata_tag
end
+ # output_metadata are provides access to hash that can be used to
+ # save metadata to be sent as trailer
+ def output_metadata
+ @output_metadata ||= {}
+ end
+
# multi_req_view provides a restricted view of this ActiveCall for use
# in a server client-streaming handler.
def multi_req_view
@@ -161,10 +170,12 @@ module GRPC
def finished
batch_result = @call.run_batch(@cq, self, INFINITE_FUTURE,
RECV_STATUS_ON_CLIENT => nil)
- if @call.metadata.nil?
- @call.metadata = batch_result.metadata
- elsif !batch_result.metadata.nil?
- @call.metadata.merge!(batch_result.metadata)
+ unless batch_result.status.nil?
+ if @call.metadata.nil?
+ @call.metadata = batch_result.status.metadata
+ else
+ @call.metadata.merge!(batch_result.status.metadata)
+ end
end
batch_result.check_status
end
@@ -192,9 +203,13 @@ module GRPC
# @param details [String] details
# @param assert_finished [true, false] when true(default), waits for
# FINISHED.
- def send_status(code = OK, details = '', assert_finished = false)
+ #
+ # == Keyword Arguments ==
+ # any keyword arguments are treated as metadata to be sent to the server
+ # if a keyword value is a list, multiple metadata for it's key are sent
+ def send_status(code = OK, details = '', assert_finished = false, **kw)
ops = {
- SEND_STATUS_FROM_SERVER => Struct::Status.new(code, details)
+ SEND_STATUS_FROM_SERVER => Struct::Status.new(code, details, kw)
}
ops[RECV_CLOSE_ON_SERVER] = nil if assert_finished
@call.run_batch(@cq, self, INFINITE_FUTURE, ops)
@@ -438,12 +453,13 @@ module GRPC
# SingleReqView limits access to an ActiveCall's methods for use in server
# handlers that receive just one request.
- SingleReqView = view_class(:cancelled, :deadline, :metadata)
+ SingleReqView = view_class(:cancelled, :deadline, :metadata,
+ :output_metadata)
# MultiReqView limits access to an ActiveCall's methods for use in
# server client_streamer handlers.
MultiReqView = view_class(:cancelled, :deadline, :each_queued_msg,
- :each_remote_read, :metadata)
+ :each_remote_read, :metadata, :output_metadata)
# Operation limits access to an ActiveCall's methods for use as
# a Operation on the client.
diff --git a/src/ruby/lib/grpc/generic/rpc_desc.rb b/src/ruby/lib/grpc/generic/rpc_desc.rb
index 3e48b8e51d..10211ae239 100644
--- a/src/ruby/lib/grpc/generic/rpc_desc.rb
+++ b/src/ruby/lib/grpc/generic/rpc_desc.rb
@@ -80,12 +80,12 @@ module GRPC
else # is a bidi_stream
active_call.run_server_bidi(mth)
end
- send_status(active_call, OK, 'OK')
+ send_status(active_call, OK, 'OK', **active_call.output_metadata)
rescue BadStatus => e
- # this is raised by handlers that want GRPC to send an application
- # error code and detail message.
+ # this is raised by handlers that want GRPC to send an application error
+ # code and detail message and some additional app-specific metadata.
logger.debug("app err: #{active_call}, status:#{e.code}:#{e.details}")
- send_status(active_call, e.code, e.details)
+ send_status(active_call, e.code, e.details, **e.metadata)
rescue Core::CallError => e
# This is raised by GRPC internals but should rarely, if ever happen.
# Log it, but don't notify the other endpoint..
@@ -135,9 +135,9 @@ module GRPC
"##{mth.name}: bad arg count; got:#{mth.arity}, want:#{want}, #{msg}"
end
- def send_status(active_client, code, details)
+ def send_status(active_client, code, details, **kw)
details = 'Not sure why' if details.nil?
- active_client.send_status(code, details, code == OK)
+ active_client.send_status(code, details, code == OK, **kw)
rescue StandardError => e
logger.warn("Could not send status #{code}:#{details}")
logger.warn(e)
diff --git a/src/ruby/lib/grpc/generic/rpc_server.rb b/src/ruby/lib/grpc/generic/rpc_server.rb
index c7c8267fa3..88c24aa92b 100644
--- a/src/ruby/lib/grpc/generic/rpc_server.rb
+++ b/src/ruby/lib/grpc/generic/rpc_server.rb
@@ -37,6 +37,120 @@ $grpc_signals = []
# GRPC contains the General RPC module.
module GRPC
+ # Handles the signals in $grpc_signals.
+ #
+ # @return false if the server should exit, true if not.
+ def handle_signals
+ loop do
+ sig = $grpc_signals.shift
+ case sig
+ when 'INT'
+ return false
+ when 'TERM'
+ return false
+ end
+ end
+ true
+ end
+ module_function :handle_signals
+
+ # Pool is a simple thread pool.
+ class Pool
+ # Default keep alive period is 1s
+ DEFAULT_KEEP_ALIVE = 1
+
+ def initialize(size, keep_alive: DEFAULT_KEEP_ALIVE)
+ fail 'pool size must be positive' unless size > 0
+ @jobs = Queue.new
+ @size = size
+ @stopped = false
+ @stop_mutex = Mutex.new
+ @stop_cond = ConditionVariable.new
+ @workers = []
+ @keep_alive = keep_alive
+ end
+
+ # Returns the number of jobs waiting
+ def jobs_waiting
+ @jobs.size
+ end
+
+ # Runs the given block on the queue with the provided args.
+ #
+ # @param args the args passed blk when it is called
+ # @param blk the block to call
+ def schedule(*args, &blk)
+ fail 'already stopped' if @stopped
+ return if blk.nil?
+ logger.info('schedule another job')
+ @jobs << [blk, args]
+ end
+
+ # Starts running the jobs in the thread pool.
+ def start
+ fail 'already stopped' if @stopped
+ until @workers.size == @size.to_i
+ next_thread = Thread.new do
+ catch(:exit) do # allows { throw :exit } to kill a thread
+ loop_execute_jobs
+ end
+ remove_current_thread
+ end
+ @workers << next_thread
+ end
+ end
+
+ # Stops the jobs in the pool
+ def stop
+ logger.info('stopping, will wait for all the workers to exit')
+ @workers.size.times { schedule { throw :exit } }
+ @stopped = true
+ @stop_mutex.synchronize do # wait @keep_alive for works to stop
+ @stop_cond.wait(@stop_mutex, @keep_alive) if @workers.size > 0
+ end
+ forcibly_stop_workers
+ logger.info('stopped, all workers are shutdown')
+ end
+
+ protected
+
+ # Forcibly shutdown any threads that are still alive.
+ def forcibly_stop_workers
+ return unless @workers.size > 0
+ logger.info("forcibly terminating #{@workers.size} worker(s)")
+ @workers.each do |t|
+ next unless t.alive?
+ begin
+ t.exit
+ rescue StandardError => e
+ logger.warn('error while terminating a worker')
+ logger.warn(e)
+ end
+ end
+ end
+
+ # removes the threads from workers, and signal when all the
+ # threads are complete.
+ def remove_current_thread
+ @stop_mutex.synchronize do
+ @workers.delete(Thread.current)
+ @stop_cond.signal if @workers.size == 0
+ end
+ end
+
+ def loop_execute_jobs
+ loop do
+ begin
+ blk, args = @jobs.pop
+ blk.call(*args)
+ rescue StandardError => e
+ logger.warn('Error in worker thread')
+ logger.warn(e)
+ end
+ end
+ end
+ end
+
# RpcServer hosts a number of services and makes them available on the
# network.
class RpcServer
@@ -69,6 +183,32 @@ module GRPC
%w(INT TERM).each { |sig| trap(sig) { $grpc_signals << sig } }
end
+ # setup_cq is used by #initialize to constuct a Core::CompletionQueue from
+ # its arguments.
+ def self.setup_cq(alt_cq)
+ return Core::CompletionQueue.new if alt_cq.nil?
+ unless alt_cq.is_a? Core::CompletionQueue
+ fail(TypeError, '!CompletionQueue')
+ end
+ alt_cq
+ end
+
+ # setup_srv is used by #initialize to constuct a Core::Server from its
+ # arguments.
+ def self.setup_srv(alt_srv, cq, **kw)
+ return Core::Server.new(cq, kw) if alt_srv.nil?
+ fail(TypeError, '!Server') unless alt_srv.is_a? Core::Server
+ alt_srv
+ end
+
+ # setup_connect_md_proc is used by #initialize to validate the
+ # connect_md_proc.
+ def self.setup_connect_md_proc(a_proc)
+ return nil if a_proc.nil?
+ fail(TypeError, '!Proc') unless a_proc.is_a? Proc
+ a_proc
+ end
+
# Creates a new RpcServer.
#
# The RPC server is configured using keyword arguments.
@@ -96,30 +236,21 @@ module GRPC
# * max_waiting_requests: the maximum number of requests that are not
# being handled to allow. When this limit is exceeded, the server responds
# with not available to new requests
+ #
+ # * connect_md_proc:
+ # when non-nil is a proc for determining metadata to to send back the client
+ # on receiving an invocation req. The proc signature is:
+ # {key: val, ..} func(method_name, {key: val, ...})
def initialize(pool_size:DEFAULT_POOL_SIZE,
max_waiting_requests:DEFAULT_MAX_WAITING_REQUESTS,
poll_period:DEFAULT_POLL_PERIOD,
completion_queue_override:nil,
server_override:nil,
+ connect_md_proc:nil,
**kw)
- if completion_queue_override.nil?
- cq = Core::CompletionQueue.new
- else
- cq = completion_queue_override
- unless cq.is_a? Core::CompletionQueue
- fail(ArgumentError, 'not a CompletionQueue')
- end
- end
- @cq = cq
-
- if server_override.nil?
- srv = Core::Server.new(@cq, kw)
- else
- srv = server_override
- fail(ArgumentError, 'not a Server') unless srv.is_a? Core::Server
- end
- @server = srv
-
+ @cq = RpcServer.setup_cq(completion_queue_override)
+ @server = RpcServer.setup_srv(server_override, @cq, **kw)
+ @connect_md_proc = RpcServer.setup_connect_md_proc(connect_md_proc)
@pool_size = pool_size
@max_waiting_requests = max_waiting_requests
@poll_period = poll_period
@@ -179,22 +310,6 @@ module GRPC
t.join
end
- # Handles the signals in $grpc_signals.
- #
- # @return false if the server should exit, true if not.
- def handle_signals
- loop do
- sig = $grpc_signals.shift
- case sig
- when 'INT'
- return false
- when 'TERM'
- return false
- end
- end
- true
- end
-
# Determines if the server is currently stopped
def stopped?
@stopped ||= false
@@ -258,19 +373,7 @@ module GRPC
end
@pool.start
@server.start
- request_call_tag = Object.new
- until stopped?
- deadline = from_relative_time(@poll_period)
- an_rpc = @server.request_call(@cq, request_call_tag, deadline)
- next if an_rpc.nil?
- c = new_active_server_call(an_rpc)
- unless c.nil?
- mth = an_rpc.method.to_sym
- @pool.schedule(c) do |call|
- rpc_descs[mth].run_server_method(call, rpc_handlers[mth])
- end
- end
- end
+ loop_handle_server_calls
@running = false
end
@@ -297,17 +400,35 @@ module GRPC
nil
end
+ # handles calls to the server
+ def loop_handle_server_calls
+ fail 'not running' unless @running
+ request_call_tag = Object.new
+ until stopped?
+ deadline = from_relative_time(@poll_period)
+ an_rpc = @server.request_call(@cq, request_call_tag, deadline)
+ c = new_active_server_call(an_rpc)
+ unless c.nil?
+ mth = an_rpc.method.to_sym
+ @pool.schedule(c) do |call|
+ rpc_descs[mth].run_server_method(call, rpc_handlers[mth])
+ end
+ end
+ end
+ end
+
def new_active_server_call(an_rpc)
- # Accept the call. This is necessary even if a status is to be sent
- # back immediately
return nil if an_rpc.nil? || an_rpc.call.nil?
# allow the metadata to be accessed from the call
handle_call_tag = Object.new
- an_rpc.call.metadata = an_rpc.metadata
- # TODO: add a hook to send md
+ an_rpc.call.metadata = an_rpc.metadata # attaches md to call for handlers
+ connect_md = nil
+ unless @connect_md_proc.nil?
+ connect_md = @connect_md_proc.call(an_rpc.method, an_rpc.metadata)
+ end
an_rpc.call.run_batch(@cq, handle_call_tag, INFINITE_FUTURE,
- SEND_INITIAL_METADATA => nil)
+ SEND_INITIAL_METADATA => connect_md)
return nil unless available?(an_rpc)
return nil unless found?(an_rpc)
@@ -319,93 +440,6 @@ module GRPC
an_rpc.deadline)
end
- # Pool is a simple thread pool for running server requests.
- class Pool
- # Default keep alive period is 1s
- DEFAULT_KEEP_ALIVE = 1
-
- def initialize(size, keep_alive: DEFAULT_KEEP_ALIVE)
- fail 'pool size must be positive' unless size > 0
- @jobs = Queue.new
- @size = size
- @stopped = false
- @stop_mutex = Mutex.new
- @stop_cond = ConditionVariable.new
- @workers = []
- @keep_alive = keep_alive
- end
-
- # Returns the number of jobs waiting
- def jobs_waiting
- @jobs.size
- end
-
- # Runs the given block on the queue with the provided args.
- #
- # @param args the args passed blk when it is called
- # @param blk the block to call
- def schedule(*args, &blk)
- fail 'already stopped' if @stopped
- return if blk.nil?
- logger.info('schedule another job')
- @jobs << [blk, args]
- end
-
- # Starts running the jobs in the thread pool.
- def start
- fail 'already stopped' if @stopped
- until @workers.size == @size.to_i
- next_thread = Thread.new do
- catch(:exit) do # allows { throw :exit } to kill a thread
- loop do
- begin
- blk, args = @jobs.pop
- blk.call(*args)
- rescue StandardError => e
- logger.warn('Error in worker thread')
- logger.warn(e)
- end
- end
- end
-
- # removes the threads from workers, and signal when all the
- # threads are complete.
- @stop_mutex.synchronize do
- @workers.delete(Thread.current)
- @stop_cond.signal if @workers.size == 0
- end
- end
- @workers << next_thread
- end
- end
-
- # Stops the jobs in the pool
- def stop
- logger.info('stopping, will wait for all the workers to exit')
- @workers.size.times { schedule { throw :exit } }
- @stopped = true
-
- @stop_mutex.synchronize do
- @stop_cond.wait(@stop_mutex, @keep_alive) if @workers.size > 0
- end
-
- # Forcibly shutdown any threads that are still alive.
- if @workers.size > 0
- logger.info("forcibly terminating #{@workers.size} worker(s)")
- @workers.each do |t|
- next unless t.alive?
- begin
- t.exit
- rescue StandardError => e
- logger.warn('error while terminating a worker')
- logger.warn(e)
- end
- end
- end
- logger.info('stopped, all workers are shutdown')
- end
- end
-
protected
def rpc_descs
@@ -416,11 +450,9 @@ module GRPC
@rpc_handlers ||= {}
end
- private
-
def assert_valid_service_class(cls)
unless cls.include?(GenericService)
- fail "#{cls} should 'include GenericService'"
+ fail "#{cls} must 'include GenericService'"
end
if cls.rpc_descs.size == 0
fail "#{cls} should specify some rpc descriptions"
@@ -430,21 +462,17 @@ module GRPC
def add_rpc_descs_for(service)
cls = service.is_a?(Class) ? service : service.class
- specs = rpc_descs
- handlers = rpc_handlers
+ specs, handlers = rpc_descs, rpc_handlers
cls.rpc_descs.each_pair do |name, spec|
route = "/#{cls.service_name}/#{name}".to_sym
- if specs.key? route
- fail "Cannot add rpc #{route} from #{spec}, already registered"
+ fail "already registered: rpc #{route} from #{spec}" if specs.key? route
+ specs[route] = spec
+ if service.is_a?(Class)
+ handlers[route] = cls.new.method(name.to_s.underscore.to_sym)
else
- specs[route] = spec
- if service.is_a?(Class)
- handlers[route] = cls.new.method(name.to_s.underscore.to_sym)
- else
- handlers[route] = service.method(name.to_s.underscore.to_sym)
- end
- logger.info("handling #{route} with #{handlers[route]}")
+ handlers[route] = service.method(name.to_s.underscore.to_sym)
end
+ logger.info("handling #{route} with #{handlers[route]}")
end
end
end