* Specify a timeout when making the HTTP request to refresh access tokens.
Without the timeout, some requests are silently never made.
Fixesrealm/realm-js-private#338.
* Add a changelog entry.
* origin/master: (23 commits)
Fix api doc error
fix progress notifications registrations
Fix typo in Github issue template
Fix port conflict between RN >= 0.48 and RPC server (#1294)
Disable testAddListener when running in chrome
Ensure RN has an event loop running for async tests
Make permission tests better handle server delays
Fix race conditions in testAddListener
Separate build and test steps in xcode to reduce chance of hitting "Early unexpected exit"
Ignore errors when sourcing nvm.sh
Don't forward arguments to nvh.sh
Skip sourcing nvm.sh if it's already available
Fix some shellcheck warnings
Ensure node 6.5.0 is installed on CI
Build realm from source for each test-runner test
Improve error reporting for incorrect argument counts for Realm methods
Use the same error messages in the RPC code as the regular code
Silence an unused variable warning when building with sync disabled
Check the exception message in all Realm tests which assert an exception is thrown
Don't discard the actual error message in validated_get_X
...
* Add basic permissions skeleton
* ...
* Update permissions api
* Wait for server to process management realm, add offer api
* Fix test of apply and get permissions, add offer test
* Accept permission offer
* Accept permission offer test
* Invalidate permission offer
* Add basic docs (still need some links etc)
* Refactor tests
* Only run permission tests where sync is enabled
* Use legal syntax for user name generation
* Add changelog entry
* Streamline permission tests
* Fix casing for access level names etc.
* Add basic definitions to index.d.ts
* Use settimeout for resolving promise from listener
* Complete typescript defs
* Improve docs
* Allow 'any' as default recipient in getGrantedPermissions
* Fix getSpecialPurposeRealm on iOS
* Response to PR comments
* Respond to PR comments
* Fix offer description
* Skip permission tests in chrome debugger :-/
* Stop trying to kill the server twice after testing
* Fix test targets to run before publishing
* ...
* Remove electron runner from publish for now
* Update changelog
Instead of rethrowing the JavaScript error emitted from a callback function, we should instead pass it to `node::FatalException`.
Callbacks are employed when there is no other JavaScript frame on the stack prior so rethrowing the JS error as a C++ exception is not going to propagate the error back to JavaScript. `node::FatalException` will raise the `uncaughtError` event on the `process` object , print the error and stacktrace, and alert the debugger is there is one attached. This would make our async callbacks behave consistently with Node’s own async callbacks such as `setTimeout` when encountering an error.
