This is a repository in the microservice demonstration system for the Tao of Microservices book (chapter 9). This code is live at nodezoo.com. To get started, visit the nodezoo/tao repository.
This microservice provides the search functionality.
To run this microservice normally, use the tooling describing in the nodezoo/tao repository, which shows you how to run the entire system of microservices (of which this is only one of many) in production (Kubernetes), staging (Docker), and development (fuge) modes.
To run from the terminal for testing and debugging, see the Running from the terminal section below.
The table shows how this microservice acts on the Accepted
message
patterns and performs appropriate business Actions
, as a result of
which, new messages are possibly Sent
.
Accepted | Actions | Sent |
---|---|---|
role:search,cmd:insert (AO) |
Insert an entry into the search engine | cmd:add,role:search |
role:search,cmd:search (SC) |
Provide a list of results |
(KEY: A: asynchronous, S: synchronous, O: observed, C: consumed)
Unit tests are in the test folder. To run, use:
$ npm test
Note that this is a learning system, and the tests are not intended to be high coverage.
This microservice is written in node.js, which you
will need to download and install. Fork and checkout this repository,
and then run npm
inside the repository folder to install its dependencies:
$ npm install
To run this microservice separately, for development, debug, or
testing purposes, use the service scripts in the srv
folder:
-
search-dev.js
: run the development configuration with hard-coded network ports.$ node srv/search-dev.js
This script listens for messages on port 9020 and provides a REPL on port 10020 (try
$ telnet localhost 10020
).A seneca-mesh version, for testing purposes, is also shown in the script
search-dev-mesh.js
. For more on this, see the nodezoo-repl repository. -
search-stage.js
: run the staging configuration. This configuration is intended to run in a Docker container so listens on port 9000 by default, but you can change that by providing an optional argument to the script.$ node srv/search-stage.js [PORT]
-
search-prod.js
: run the production configuration. This configuration is intended to run under Kubernetes in a seneca-mesh network. If running in a terminal (only do this for testing), you'll need to provide the mesh base nodes in theBASES
environment variable.$ BASES=x.x.x.x:port node srv/search-prod.js