Brickflow / amqp2solr

amqp2solr is a wrapper for delayed population of a solr database

amqp2solr

amqp2solr is a library both for querying a SOLR core and queueing (add) queries for delayed execution. It is a wrapper, which aims to make delayed/remote execution of solr queries more seamless.

Basic usage

var dependencies = { config, logger, amqp };
var amqp2solr = require('amqp2solr')(dependencies);

The dependencies parameter can be used for overriding config and/or logger.

• logger should have info, error and debug methods. It defaults to console.log.
• config assumed to have a get method. (nconf is used)

getResource(options) and query locally

var resourceParams = { core: 'blogs' };
var blogResource = amqp2solr.getResource(resourceParams);
// ... and then ...
blogResource.add({id: 'example', someField: 1}, cb);
blogResource.createOrUpdate({id: 'example'}, {someOtherField: 1}, cb);

getQueue([queueName,] blogResource) and queue queries

// Create a resource ...
var blogResource = amqp2solr.getResource(resourceParams);

// ... and map it to the resource ...
var blogQueue = amqp2solr.getQueue(blogResource);
// ... optionally with explicitly given AMQP queue name ...
var blogQueue = amqp2solr.getQueue(queueName, blogResource);

You can give resourceParams instead of an actual resource (can be useful in environments where you don't want to use the resource locally.

var blogQueue = amqp2solr.getQueue(queueName, resourceParams);

You can get the resource var blogResource = blogQueue.resource;.

solrQueue.listen() to consume a queue

amqp2solr.getQueue(queueName, resourceParams).listen(); 

getResource(options)

Returns a solrResource instance.

The options parameter accepts the following fields:

• core: the name of the core to connect to, defaults to ''
• fields: an object consisting of key-value pairs to transform field names.
  • it's keys are keys of your existing model
  • it's values are field names in SOLR
  • Typically, this can be used to adapt to the default solr schema.xml, eg: {email: 'email_s'}
• transformations is an optional object which can be used for transforming the data before/after solr.
  • it's keys are keys of your existing model
  • it's values are objects with 2 fields:
  • formSolr: function(value) {return value; }
  • toSolr: function(value) {return value; }
• mlt is an optional more like this setting to be used as default in the recommend method.

It returns a solrResource object;

solrResource

• add(doc [,cb]): creates/replaces a document
• find(q [,cb]): find documents both by query or id
• findAndModify(q, updateParams [, opts ,cb]): update some fields of existing documents matched by find(q)
• createOrUpdate(q, updateParams, [, cb]): similar to findAndModify, but this creates a new document if none found.
• moreLikeThis(q, [mlt], cb): find similar documents to any document which matches
• delete(id, cb) deletes by a query, mapped from node-solr-client
• deleteById(id, cb) deletes by id
• encode formats the document to solr (transposes field names and performs transformations)
• deocde formats the document came back from solr (reverses field names and performs reverse transformations)
• solr exposes the wrapped node-solr-client

getQueue([queueName,] solrResourceOrOptions)

Returns a solrQueue instance. If solrResourceOrOptions is not a solrResource, it calls getResource with solrResourceOrOptions.

getAsymmetric([queueName,] solrResourceOrOptions)

Returns an asymmetric (queued-write, direct-read), mixed resource which queues inserts/updates/modifies but directly queries selects/moreLikeThis/...

solrQueue

solrQueue has exactly the same methods than a solrResource, but it pushes the task to the queue rather than executing it locally.