This commit updates the short connect timeout tests to tolerate either a
ConnectException or SocketTimeoutException as a success condition.
Connections could fail with either exception depending upon timing or
network setup.
This commit enables http clients to optionally configure a connect
and/or socket timeout for requests. For persistent clients, the
timeout values can only be configured at the client and not a
per-request level. Non-persistent client requests support the
configuration of these timeouts.
This commit also bumps a few dependencies - tk-jetty9 to 1.2.0,
trapperkeeper to 1.10, ssl-utils to 0.8.0, and clojure to 1.6.0 -
and adds the 'test' tk-jetty9 library as a dev dependency for
testing.
Add a PersistentSyncHttpClient class, which is a wrapper around
CloseableHttpAsyncClient and allows multiple requests to be made
with a persistent synchronous client.
Separate the Java RequestOptions class into two new classes:
RequestOptions, which contains options related to the request,
and ClientOptions, which contains options related to the client.
Both are now required when calling the request methods in
AsyncHttpClient and SyncHttpClient.
This commit encodes a request body string per the content of the
Content-Type header. In the event that no charset is specified in the
Content-Type header, the charset is set to UTF-8. Previously, the
request body string was always encoded to ISO-8859-1 and was not
necessarily in sync with the value of the Content-Type header.
Add query parameter support to the clojure client. User can now
specify a :query-params option in the options map containing a
map of query parameters.
Add a persistent sync client to the persistent-sync namespace.
With this namespace, a user can now create a persistent
sync client for reuse. Also, add persistent sync client test.
* Ports the code over to use the Apache HttpAsyncClient library
instead of http-kit, as we were getting some weird and inconsistent
SSL errors from http-kit, and
* Simplify the API (by eliminating a lot of the supported configuration
options). Some of these options we'll eventually want to add back in,
but for now, getting rid of them makes the API less cluttered and
also gives us the opportunity to do a better job writing tests
for new options when we do add support for them.
For now it only supports constructing a new client on every request.
I intend to add API for creating a client explicitly, and a protocol
for how to interact with the client. This will allow users to
re-use an existing client as they see fit... but this hasn't been
implemented yet.
