\n\n**[Memphis.dev](https://memphis.dev)** is a highly scalable, painless, and effortless data streaming platform. \nMade to enable developers and data teams to collaborate and build \nreal-time and streaming apps fast.\n\n
\n \n
.\n\n## Installation\n\n```sh\n$ npm install memphis-dev\n```\n\n## Importing\n\nFor Javascript, you can choose to use the import or required keyword. This library exports a singleton instance of `memphis` with which you can consume and produce messages.\n\n```js\nconst { memphis } = require('memphis-dev');\n```\n\nFor Typescript, use the import keyword. You can import `Memphis` to aid for typechecking assistance.\n\n```js\nimport { memphis, Memphis } from 'memphis-dev';\n```\n\nTo leverage Nestjs dependency injection feature\n\n```js\nimport { Module } from '@nestjs/common';\nimport { Memphis, MemphisModule, MemphisService } from 'memphis-dev';\n```\n\n### Connecting to Memphis\n\nFirst, we need to connect with Memphis by using `memphis.connect`.\n\n```js\n/* Javascript and typescript project */\nawait memphis.connect({\n host: \"\",\n port: , // defaults to 6666\n username: \"\", // (root/application type user)\n accountId: //You can find it on the profile page in the Memphis UI. This field should be sent only on the cloud version of Memphis, otherwise it will be ignored\n connectionToken: \"\", // you will get it on application type user creation\n password: \"\", // depends on how Memphis deployed - default is connection token-based authentication\n reconnect: true, // defaults to true\n maxReconnect: 10, // The default value is -1 which means reconnect indefinitely\n reconnectIntervalMs: 1500, // defaults to 1500\n timeoutMs: 15000, // defaults to 15000\n // for TLS connection:\n keyFile: '',\n certFile: '',\n caFile: ''\n suppressLogs: false // defaults to false - indicates whether to suppress logs or not\n });\n```\n\nThe connect function in the Memphis class allows for the connection to Memphis. Connecting to Memphis (cloud or open-source) will be needed in order to use any of the other functionality of the Memphis class. Upon connection, all of Memphis' features are available.\n\nNest injection\n\n```js\n@Module({\n imports: [MemphisModule.register()],\n})\n\nclass ConsumerModule {\n constructor(private memphis: MemphisService) {}\n\n startConnection() {\n (async function () {\n let memphisConnection: Memphis;\n\n try {\n memphisConnection = await this.memphis.connect({\n host: \"\",\n username: \"\",\n connectionToken: \"\",\n });\n } catch (ex) {\n console.log(ex);\n memphisConnection.close();\n }\n })();\n }\n}\n```\n\nWhat arguments are used with the Memphis.connect function change depending on the type of connection being made. Memphis uses password-based connections by default. An example of a password-based connection would look like this (using the defualt root memphis login with Memphis open-source):\n\n```typescript\nasync function connectPassword(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({\n host: \"localhost\",\n username: \"root\", // (root/application type user)\n password: \"memphis\" \n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\nIf you wanted to connect to Memphis cloud instead, simply add your account ID and change the host. The host and account_id can be found on the Overview page in the Memphis cloud UI under your name at the top. Here is an example to connecting to a cloud broker that is located in US East: \n\n```typescript\nasync function connectPasswordCloud(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({\n host: \"aws-us-east-1.cloud.memphis.dev\",\n username: \"my_client_username\", // (root/application type user)\n password: \"my_client_password\",\n accountId: 123456789\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\nIt is possible to use a token-based connection to memphis as well, where multiple users can share the same token to connect to memphis. Here is an example of using memphis.connect with a token:\n\n```typescript\nasync function connectToken(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({\n host: \"localhost\",\n username: \"root\", // (root/application type user)\n connectionToken: \"token\"\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\nThe token will be presented when creating new users.\n\nMemphis needs to be configured to use token based connection. See the [docs](https://docs.memphis.dev/memphis/memphis-broker/concepts/security) for help doing this.\n\nA TLS based connection would look like this:\n\n```typescript\nasync function connectTLS(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({\n host: \"localhost\",\n username: \"root\", // (root/application type user)\n keyFile: \"~/tls_file_path.key\",\n certFile: \"~/tls_cert_file_path.crt\",\n caFile: \"~/tls_ca_file_path.crt\"\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\nMemphis needs to configured for these use cases. To configure memphis to use TLS see the [docs](https://docs.memphis.dev/memphis/open-source-installation/kubernetes/production-best-practices#memphis-metadata-tls-connection-configuration). \n\n### Disconnecting from Memphis\n\nTo disconnect from Memphis, call `close()` on the memphis object.\n\n```js\nmemphisConnection.close();\n```\n\n### Creating a Station\n\nStations are distributed units that store messages. Producers add messages to stations and Consumers take messages from them. Each station stores messages until their retention policy causes them to either delete the messages or move them to [remote storage](https://docs.memphis.dev/memphis/integrations-center/storage/s3-compatible). \n\n**A station will be automatically created for the user when a consumer or producer is used if no stations with the given station name exist.**
\n_If the station trying to be created exists when this function is called, nothing will change with the exisitng station_\n\n```js\nconst station = await memphis.station({\n name: '',\n schemaName: '',\n retentionType: memphis.retentionTypes.MAX_MESSAGE_AGE_SECONDS, // defaults to memphis.retentionTypes.MAX_MESSAGE_AGE_SECONDS\n retentionValue: 3600, // defaults to 3600\n storageType: memphis.storageTypes.DISK, // defaults to memphis.storageTypes.DISK\n replicas: 1, // defaults to 1\n idempotencyWindowMs: 0, // defaults to 120000\n sendPoisonMsgToDls: true, // defaults to true\n sendSchemaFailedMsgToDls: true, // defaults to true\n tieredStorageEnabled: false, // defaults to false\n partitionsNumber: 1, // defaults to 1\n dlsStation:'' // defaults to \"\" (no DLS station) - If selected DLS events will be sent to selected station as well\n});\n```\n\nThe station function is used to create a station. Using the different arguemnts, one can programically create many different types of stations. The Memphis UI can also be used to create stations to the same effect. \n\nCreating a station with Nestjs dependency injection\n\n```js\n@Module({\n imports: [MemphisModule.register()],\n})\n\nclass stationModule {\n constructor(private memphis: MemphisService) { }\n\n createStation() {\n (async function () {\n const station = await this.memphis.station({\n name: \"\",\n schemaName: \"\",\n retentionType: memphis.retentionTypes.MAX_MESSAGE_AGE_SECONDS, // defaults to memphis.retentionTypes.MAX_MESSAGE_AGE_SECONDS\n retentionValue: 604800, // defaults to 604800\n storageType: memphis.storageTypes.DISK, // defaults to memphis.storageTypes.DISK\n replicas: 1, // defaults to 1\n idempotencyWindowMs: 0, // defaults to 120000\n sendPoisonMsgToDls: true, // defaults to true\n sendSchemaFailedMsgToDls: true, // defaults to true\n tieredStorageEnabled: false, // defaults to false\n dlsStation:'' // defaults to \"\" (no DLS station) - If selected DLS events will be sent to selected station as well\n });\n })();\n }\n}\n```\n\nA minimal example, using all default values would simply create a station with the given name:\n\n```typescript\nasync function stationDefault(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({...});\n await memphisConnection.station({\n name: \"myStation\"\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\nTo change what criteria the station uses to decide if a message should be retained in the station, change the retention type. The different types of retention are documented [here](https://github.com/memphisdev/memphis.js#retention-types) in the node README. \n\nThe unit of the rentention value will vary depending on the RetentionType. The [previous link](https://github.com/memphisdev/memphis.js#retention-types) also describes what units will be used. \n\nHere is an example of a station which will only hold up to 10 messages:\n\n```typescript\nasync function stationRetentionType(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({...});\n await memphisConnection.station({\n name: \"myStation\",\n retentionType: memphis.retentionTypes.MESSAGES\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\nMemphis stations can either store Messages on disk or in memory. A comparison of those types of storage can be found [here](https://docs.memphis.dev/memphis/memphis-broker/concepts/storage-and-redundancy#tier-1-local-storage).\n\nHere is an example of how to create a station that uses Memory as its storage type:\n\n```typescript\nasync function stationMemoryStorage(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({...});\n await memphisConnection.station({\n name: \"myStation\",\n storageType: memphis.storageTypes.MEMORY\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\nIn order to make a station more redundant, replicas can be used. Read more about replicas [here](https://docs.memphis.dev/memphis/memphis-broker/concepts/storage-and-redundancy#replicas-mirroring). Note that replicas are only available in cluster mode. Cluster mode can be enabled in the [Helm settings](https://docs.memphis.dev/memphis/open-source-installation/kubernetes/1-installation#appendix-b-helm-deployment-options) when deploying Memphis with Kubernetes.\n\nHere is an example of creating a station with 3 replicas:\n\n```typescript\nasync function stationWithReplicas(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({...});\n await memphisConnection.station({\n name: \"myStation\",\n replicas: 3\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\nIdempotency defines how Memphis will prevent duplicate messages from being stored or consumed. The duration of time the message ID's will be stored in the station can be set with idempotencyWindowsMs. If the environment Memphis is deployed in has unreliably connection and/or a lot of latency, increasing this value might be desiriable. The default duration of time is set to two minutes. Read more about idempotency [here](https://docs.memphis.dev/memphis/memphis-broker/concepts/idempotency).\n\nHere is an example of changing the idempotency window to 3 seconds:\n\n```typescript\nasync function stationIdempotency(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({...});\n await memphisConnection.station({\n name: \"myStation\",\n idempotencyWindowMs: 180000\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\nThe schema name is used to set a schema to be enforced by the station. The default value of \"\" ensures that no schema is enforced. Here is an example of changing the schema to a defined schema in schemaverse called \"sensorLogs\":\n\n```typescript\nasync function stationWithSchema(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({...});\n await memphisConnection.station({\n name: \"myStation\",\n schemaName: \"sensorLogs\"\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\nThere are two parameters for sending messages to the [dead-letter station(DLS)](https://docs.memphis.dev/memphis/memphis-broker/concepts/dead-letter#terminology). These are sendPoisonMsgToDls and sendSchemaFailedMsgToDls. \n\nHere is an example of sending poison messages to the DLS but not messages which fail to conform to the given schema.\n\n```typescript\nasync function stationWithDeadLetter(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({...});\n await memphisConnection.station({\n name: \"myStation\",\n sendPoisonMsgToDls: true,\n sendSchemaFailedMsgToDls: false\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\nWhen either of the DLS flags are set to True, a station can also be set to handle these events. To set a station as the station to where schema failed or poison messages will be set to, use the DlsStation parameter:\n\n```typescript\nasync function stationWithDeadLetterToStation(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({...});\n await memphisConnection.station({\n name: \"myStation\",\n sendPoisonMsgToDls: true,\n sendSchemaFailedMsgToDls: false,\n dlsStation: \"badSensorLogsStation\"\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\nWhen the retention value is met, Memphis by default will delete old messages. If tiered storage is setup, Memphis can instead move messages to tier 2 storage. Read more about tiered storage [here](https://docs.memphis.dev/memphis/memphis-broker/concepts/storage-and-redundancy#storage-tiering). Enable this setting with the respective flag:\n\n```typescript\nasync function stationWithTieredStorage(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({...});\n await memphisConnection.station({\n name: \"myStation\",\n tieredStorageEnabled: true\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\n[Partitioning](https://docs.memphis.dev/memphis/memphis-broker/concepts/station#partitions) might be useful for a station. To have a station partitioned, simply change the partitions number:\n\n```typescript\nasync function stationWithPartitions(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({...});\n await memphisConnection.station({\n name: \"myStation\",\n partitionsNumber: 3\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\n\n### Retention types\n\nRetention types define the methodology behind how a station behaves with its messages. Memphis currently supports the following retention types:\n\n```js\nmemphis.retentionTypes.MAX_MESSAGE_AGE_SECONDS;\n```\n\nWhen the retention type is set to MAX_MESSAGE_AGE_SECONDS, messages will persist in the station for the number of seconds specified in the retention_value. \n\n```js\nmemphis.retentionTypes.MESSAGES;\n```\n\nWhen the retention type is set to MESSAGES, the station will only hold up to retention_value messages. The station will delete the oldest messsages to maintain a retention_value number of messages.\n\n```js\nmemphis.retentionTypes.BYTES;\n```\n\nWhen the retention type is set to BYTES, the station will only hold up to retention_value BYTES. The oldest messages will be deleted in order to maintain at maximum retention_vlaue BYTES in the station.\n\n```js\nmemphis.retentionTypes.ACK_BASED; // for cloud users only\n```\n\nWhen the retention type is set to ACK_BASED, messages in the station will be deleted after they are acked by all subscribed consumer groups.\n\n### Retention Values\n\nThe unit of the `retention value` changes depending on the `retention type` specified. \n\nAll retention values are of type `int`. The following units are used based on the respective retention type:\n\n`memphis.retentionTypes.MAX_MESSAGE_AGE_SECONDS` is **in seconds**, \n`memphis.retentionTypes.MESSAGES` is a **number of messages**, \n`memphis.retentionTypes.BYTES` is a **number of bytes** \nWith `memphis.retentionTypes.ACK_BASED`, the `retentionValue` is ignored.\n\n### Storage types\n\nMemphis currently supports the following types of messages storage:\n\n```js\nmemphis.storageTypes.DISK;\n```\n\nWhen storage is set to DISK, messages are stored on disk.\n\n```js\nmemphis.storageTypes.MEMORY;\n```\n\nWhen storage is set to MEMORY, messages are stored in the system memory.\n\n### Destroying a Station\n\nDestroying a station will remove all its resources (producers/consumers)\n\n```js\nawait station.destroy();\n```\n\n### Creating a new schema \nIn case schema is already exist a new version will be created\n\n```js\nawait memphisConnection.createSchema({schemaName: \"\", schemaType: \"\", schemaFilePath: \"\" });\n```\n\n### Enforcing a schema on an existing Station\n\n```js\nawait memphisConnection.enforceSchema({ name: '', stationName: '' });\n```\n\n### Deprecated - Use enforceSchema instead\n\n```js\nawait memphisConnection.attachSchema({ name: '', stationName: '' });\n```\n\n### Detaching a schema from Station\n\n```js\nawait memphisConnection.detachSchema({ stationName: '' });\n```\n\n### Produce and Consume messages\n\nThe most common client operations are using `produce` to send messages and `consume` to\nreceive messages.\n\nMessages are published to a station with a Producer and consumed from it by a Consumer. \n\nConsumers are poll based and consume all the messages in a station. Consumers can also be grouped into consumer groups. When consuming with a consumer group, all consumers in the group will receive each message.\n\nMemphis messages are payload agnostic. Payloads are always `bytearray`s.\n\nIn order to stop getting messages, you have to call `consumer.destroy()`. Destroy will terminate the consumer even if messages are currently being sent to the consumer.\n\nIf a station is created with more than one partition, producing to and consuming from the station will happen in a round robin fashion. \n\n### Creating a Producer\n\n```js\nconst producer = await memphisConnection.producer({\n stationName: '',\n producerName: '',\n});\n```\n\nCreating producers with nestjs dependecy injection\n\n```js\n@Module({\n imports: [MemphisModule.register()],\n})\n\nclass ProducerModule {\n constructor(private memphis: MemphisService) { }\n\n createProducer() {\n (async function () {\n const producer = await memphisConnection.producer({\n stationName: \"\",\n producerName: \"\"\n });\n })();\n }\n}\n```\n\n### Producing a message\n\nBoth producers and connections can use the produce function. To produce a message from a connection, simply call `memphis.produce`. This function will create a producer if none with the given name exists, otherwise it will pull the producer from a cache and use it to produce the message. \n\n```js\nawait memphisConnection.produce({\n stationName: '',\n producerName: '',\n message: 'Uint8Arrays/object/string/DocumentNode graphql', // Uint8Arrays/object (schema validated station - protobuf) or Uint8Arrays/object (schema validated station - json schema) or Uint8Arrays/string/DocumentNode graphql (schema validated station - graphql schema) or Uint8Arrays/object (schema validated station - avro schema)\n ackWaitSec: 15, // defaults to 15\n asyncProduce: true // defaults to true. For better performance. The client won't block requests while waiting for an acknowledgment.\n headers: headers, // defults to empty\n msgId: 'id', // defaults to null\n producerPartitionKey: \"key\", // produce to specific partition. defaults to null\n producerPartitionNumber: -1 // produce to specific partition number. defaults to -1\n});\n```\n\nCreating a producer and calling produce on it will increase the performance of producing messages as it removes the overhead of pulling created producers from the cache.\n\n```js\nawait producer.produce({\n message: 'Uint8Arrays/object/string/DocumentNode graphql', // Uint8Arrays/object (schema validated station - protobuf) or Uint8Arrays/object (schema validated station - json schema) or Uint8Arrays/string/DocumentNode graphql (schema validated station - graphql schema) or Uint8Arrays/object (schema validated station - avro schema)\n ackWaitSec: 15, // defaults to 15,\n producerPartitionKey: \"key\", // produce to specific partition. defaults to null\n producerPartitionNumber: -1 // produce to specific partition number. defaults to -1\n});\n```\n\nNote:\nWhen producing to a station with more than one partition, the producer will produce messages in a Round Robin fashion between the different partitions.\n\nFor message data formats see [here](https://docs.memphis.dev/memphis/memphis-schemaverse/formats/produce-consume). \n\nWhen producing many messages with a producer, asyncProduce may be used to help increase the performance of the producer. By default, a producer will produce messages asynchronously. This can be set to false to increase reliability. \n\nHere is an example of a produce function call that waits up to 30 seconds for an acknowledgement from the memphis broker and does so in an async manner:\n\n```typescript\nasync function produceAsync(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({...});\n await memphisConnection.produce({\n stationName: \"myStation\",\n producerName: \"tempProducer\",\n message: {some: \"message\"},\n ackWaitSec: 30,\n asyncProduce: true\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\nAs discussed before in the station section, idempotency is an important feature of memphis. To achieve idempotency, an id must be assigned to messages that are being produced. Use the msgId parameter for this purpose.\n\n```typescript\nasync function produceWithIdempotency(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({...});\n await memphisConnection.produce({\n stationName: \"myStation\",\n producerName: \"tempProducer\",\n message: {some: \"message\"},\n msgId: \"42\"\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\nTo add message headers to the message, use the headers function to create a headers object which you can add headers to. Headers can help with observability when using certain 3rd party to help monitor the behavior of memphis. See [here](https://docs.memphis.dev/memphis/memphis-broker/comparisons/aws-sqs-vs-memphis#observability) for more details.\n\n```typescript\nasync function produceWithHeaders(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({...});\n\n let headers = memphis.headers()\n headers.add(\"trace_header_example\", \"track_me_123\");\n\n await memphisConnection.produce({\n stationName: \"myStation\",\n producerName: \"tempProducer\",\n message: {some: \"message\"},\n headers: headers\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\nor:\n\n```typescript\nasync function produceWithHeaders(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({...});\n\n let headers = {trace_header_example, \"track_me_123\"};\n\n await memphisConnection.produce({\n stationName: \"myStation\",\n producerName: \"tempProducer\",\n message: {some: \"message\"},\n headers: headers\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\nLastly, memphis can produce to a specific partition in a station. To do so, use the producerPartitionKey parameter:\n\n```typescript\nasync function produceWithPartition(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({...});\n await memphisConnection.produce({\n stationName: \"myStation\",\n producerName: \"tempProducer\",\n message: {some: \"message\"},\n producerPartitionKey: \"somePartition\"\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\n### Produce to multiple stations\n\nProducing to multiple stations can be done by creating a producer with multiple stations and then calling produce on that producer.\n\n```js\nconst producer = await memphisConnection.producer({\n stationName: ['', ''],\n producerName: '',\n});\n\nawait producer.produce({\n message: {some: \"message\"},\n ackWaitSec: 15,\n producerPartitionKey: \"key\", \n producerPartitionNumber: -1 \n});\n```\n\nAlternatively, it also possible to produce to multiple stations using the connection:\n\n```js\nawait memphisConnection.produce({\n stationName: ['', ''],\n producerName: '',\n message: {some: \"message\"},\n ackWaitSec: 15,\n asyncProduce: true \n});\n```\n\n### Destroying a Producer\n\n```js\nawait producer.destroy();\n```\n\n### Creating a Consumer\n\n```js\nconst consumer = await memphisConnection.consumer({\n stationName: '',\n consumerName: '',\n consumerGroup: '', // defaults to the consumer name.\n pullIntervalMs: 1000, // defaults to 1000\n batchSize: 10, // defaults to 10\n batchMaxTimeToWaitMs: 1000, // defaults to 1000\n maxAckTimeMs: 30000, // defaults to 30000\n maxMsgDeliveries: 2, // defaults to 2\n startConsumeFromSequence: 1, // start consuming from a specific sequence. defaults to 1\n lastMessages: -1, // consume the last N messages, defaults to -1 (all messages in the station)\n consumerPartitionKey: \"key\", // consume by specific partition key. Defaults to null\n consumerPartitionNumber: -1 // consume by specific partition number. Defaults to -1\n});\n```\n\nNote:\nWhen consuming from a station with more than one partition, the consumer will consume messages in Round Robin fashion from the different partitions.\n\nHere is an example on how to create a consumer with all of the default options:\n\n```typescript\nasync function consumerDefualt(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({...});\n await memphisConnection.consumer({\n stationName: \"myStation\",\n consumerName: \"newConsumer\"\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\nTo create a consumer in a consumer group, add the consumerGroup parameter:\n\n```typescript\nasync function consumerGroup(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({...});\n await memphisConnection.consumer({\n stationName: \"myStation\",\n consumerName: \"newConsumer\",\n consumerGroup: \"consumerGroup1\"\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\nWhen using Consumer.consume, the consumer will continue to consume in an infinite loop. To change the rate at which the consumer polls, change the pullIntervalMs parameter:\n\n```typescript\nasync function consumerPollInterval(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({...});\n await memphisConnection.consumer({\n stationName: \"myStation\",\n consumerName: \"newConsumer\",\n pullIntervalMs: 2000\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\nEvery time the consumer pulls messages from a station, the consumer will try to take batchSize number of elements from the station. However, sometimes there are not enough messages in the station for the consumer to consume a full batch. In this case, the consumer will continue to wait until either batchSize messages are gathered or the time in milliseconds specified by batchMaxTimeToWaitMs is reached. \n\nHere is an example of a consumer that will try to pull 100 messages every 10 seconds while waiting up to 15 seconds for all messages to reach the consumer.\n\n```typescript\nasync function consumerBatched(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({...});\n await memphisConnection.consumer({\n stationName: \"myStation\",\n consumerName: \"newConsumer\",\n pullIntervalMs: 10000,\n batchSize: 100,\n batchMaxTimeToWaitMs: 1000\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\nThe maxMsgDeliveries parameter allows the user how many messages the consumer is able to consume before consuming more. \n\n```typescript\nasync function consumerMaxMessages(){\n let memphisConnection;\n try{\n memphisConnection = await memphis.connect({...});\n await memphisConnection.consumer({\n stationName: \"myStation\",\n consumerName: \"newConsumer\",\n pullIntervalMs: 10000,\n batchSize: 100,\n batchMaxTimeToWaitMs: 1000,\n maxMsgDeliveries: 2\n });\n }catch(exception){\n // Handle exception\n }\n}\n```\n\n### Passing context to message handlers\n\n```js\nconsumer.setContext({ key: 'value' });\n```\n\n### Processing messages\n\n```js\nconsumer.on('message', (message, context) => {\n // processing\n console.log(message.getData());\n message.ack();\n});\n```\n\n#### Processing schema deserialized messages\nTo get messages deserialized, use `message.getDataDeserialized()`. \n\n```js\nconsumer.on('message', (message, context) => {\n // processing\n console.log(message.getDataDeserialized());\n message.ack();\n});\n```\n\nThere may be some instances where you apply a schema *after* a station has received some messages. In order to consume those messages get_data_deserialized may be used to consume the messages without trying to apply the schema to them. As an example, if you produced a string to a station and then attached a protobuf schema, using get_data_deserialized will not try to deserialize the string as a protobuf-formatted message.\n\n### Fetch a single batch of messages\n\nUsing fetch_messages or fetch will allow the user to remove a specific number of messages from a given station. This behavior could be beneficial if the user does not want to have a consumer actively poll from a station indefinetly.\n\n```js\nconst msgs = await memphis.fetchMessages({\n stationName: '',\n consumerName: '',\n consumerGroup: '', // defaults to the consumer name.\n batchSize: 10, // defaults to 10\n batchMaxTimeToWaitMs: 1000, // defaults to 1000\n maxAckTimeMs: 30000, // defaults to 30000\n maxMsgDeliveries: 2, // defaults to 2\n startConsumeFromSequence: 1, // start consuming from a specific sequence. defaults to 1\n lastMessages: -1, // consume the last N messages, defaults to -1 (all messages in the station)\n consumerPartitionKey: \"key\", // consume by specific partition key. Defaults to null\n consumerPartitionNumber: -1 // consume by specific partition number. Defaults to -1\n});\n```\n\n### Fetch a single batch of messages after creating a consumer\n\n```js\nconst msgs = await consumer.fetch({\n batchSize: 10, // defaults to 10\n consumerPartitionKey: \"key\", // fetch by specific partition key. Defaults to null\n consumerPartitionNumber: -1 // fetch by specific partition number. Defaults to -1\n});\n```\n\nTo set up a connection in nestjs\n\n```js\nimport { MemphisServer } from 'memphis-dev'\n\nasync function bootstrap() {\n const app = await NestFactory.createMicroservice(\n AppModule,\n {\n strategy: new MemphisServer({\n host: '',\n username: '',\n connectionToken: ''\n }),\n },\n );\n\n await app.listen();\n}\nbootstrap();\n```\n\nTo consume messages in NestJS\n\n```js\nexport class Controller {\n import { MemphisConsume, Message } from 'memphis-dev';\n\n @MemphisConsume({\n stationName: '',\n consumerName: '',\n consumerGroup: ''\n })\n async messageHandler(message: Message) {\n console.log(message.getData().toString());\n message.ack();\n }\n}\n```\n\n### Acknowledge a message\n\nAcknowledge a message indicates the Memphis server to not re-send the same message again to the same consumer/consumers group\n\n```js\nmessage.ack();\n```\n\n### Nacking a Message\n\nMark the message as not acknowledged - the broker will resend the message immediately to the same consumers group, instead of waiting to the max ack time configured.\n\n```js\nmsg.nack();\n```\n\n### Sending a message to the dead-letter\n\nSending the message to the dead-letter station (DLS) - the broker won't resend the message again to the same consumers group and will place the message inside the dead-letter station (DLS) with the given reason.\nThe message will still be available to other consumer groups\n\n```js\nmessage.deadLetter(\"reason\");\n```\n\n### Delay and resend the message after a given duration\n\nDelay the message and tell the Memphis server to re-send the same message again to the same consumer group. The message will be redelivered only in case `Consumer.maxMsgDeliveries` is not reached yet.\n\n```js\nmessage.delay(delayInMilliseconds);\n```\n\n### Get message payload\n\nAs Uint8Array\n\n```js\nmsg = message.getData();\n```\n\nAs Json\n\n```js\nmsg = message.getDataAsJson();\n```\n\n### Get headers\n\nGet headers per message\n\n```js\nheaders = message.getHeaders();\n```\n\n### Get message sequence number\n\nGet message sequence number\n\n```js\nsequenceNumber = message.getSequenceNumber();\n```\n\n### Get message sent time\n\nGet the time the message was sent\n\n```js\nsentTime = message.getTimeSent();\n```\n\n\n### Catching async errors\n\n```js\nconsumer.on('error', (error) => {\n // error handling\n});\n```\n\n### Stopping a Consumer \n\nStopping a consumer simply stops it from consuming messages in the code. \n\nLet's say you don't want listeners of a consumer to receive messages anymore (even if messages are still being produced to its station), stop the consumer and that's it.\n\n```js\nawait consumer.stop();\n```\n\n### Destroying a Consumer\n\nThis is different from stopping a consumer. Destroying a consumer destroys it from the station and the broker itself. It won't exist again.\n\n```js\nawait consumer.destroy();\n```\n\n### Check if the broker is connected\n\n```js\nmemphisConnection.isConnected();\n```\n\n### Create memphis instance \n\n```js\nconst memphisInstance = createMemphisInstance()\n```\n","readmeFilename":"README.md","homepage":"https://memphis.dev","repository":{"type":"git","url":"git+https://github.com/memphisdev/memphis.js.git"},"bugs":{"url":"https://github.com/memphisdev/memphis.js/issues"}}
![\"Discord\"](\"https://img.shields.io/discord/963333392844328961?color=6557ff&label=discord\")
\n![](\"https://img.shields.io/github/issues-closed/memphisdev/memphis?color=6557ff\")
\n ![](\"https://img.shields.io/npm/dw/memphis-dev?color=ffc633&label=installations\")
\n![\"Code](\"https://img.shields.io/badge/Code%20of%20Conduct-v1.0-ff69b4.svg?color=ffc633\"){kind=icon}
\n![\"GitHub](\"https://img.shields.io/github/v/release/memphisdev/memphis?color=61dfc6\")
\n![](\"https://img.shields.io/github/last-commit/memphisdev/memphis?color=61dfc6&label=last%20commit\")
\n