mongocloner

A simple utility to drop, clone, append or backup data from a mongo URI to another.

With this tool you can do different actions:

• backup_json: backup specified collections from the source in json format
• backup_yaml: backup specified collections from the source in yaml format
• append: append specified collections from a wellknown mongo uri to another
• delete_target: drop specified collections from the target
• clone: drop and append specified collections from a wellknown mongo uri to another
• list_source: list all collection names from the source
• list_target: list all collection names from the target
• size: list every document with size > 100kb

Fully BASED on

First of all, thanks to Matheus Mendes de Sant'Ana (profile) for share NodeJS_Copy_Collection.

Important note

This code is not optimized and is not designed for production. It's is usable in development contest with small collections. Feel fre to submit a PR to improve this utility. And feel free to use it in your project (send me a message if you want, it's really appreciated).

Installation

1. fork or clone this repo
2. create your own .env
3. execute yarn to install all the stuff
4. execute yarn dev or yarn start (it's the same) to execute it
5. that's all

Command line

Not necessary.

Content of .env

# Possible ACTIONs --> backup_json, backup_yaml, append, clone, delete_target, list_source, list_target
ACTION=backup_source
CHUNK_SIZE=10000
TIMEOUT=3000

SOURCE_URI=mongodb+srv://YOUR-SOURCE-USR:[email protected]/YOUR-SOURCE-DB
SOURCE_DATABASE=YOUR-SOURCE-DB

TARGET_URI=mongodb+srv://YOUR-TARGET-USR:[email protected]/YOUR-TARGET-DB
TARGET_DATABASE=YOUR-TARGET-DB

# List of collections to read (comma separated)
SOURCE_COLLECTIONS=collection1,collection2,collection3,collection4,collection5

# Void if the same of SOURCE or is possible to specify different names
# (but with the same number of collections, and pay attention to the order because matters)
TARGET_COLLECTIONS=

# Enable processor (if present)
RUN_PROCESSOR=true

How to process documents before inserting

In the class processor.js is possible to add specific logic to manipulate documents before inserting into target.

Note: you can enable/disable the processor via .env (property RUN_PROCESSOR=true)

For this purpose you can customize code into class processor, here you can change these 2 methods:

let fields = []

async function initialize(data) {
  const { source, target } = data

  // do something like load static data to use in process method
  console.log('Load [myfields] from source')
  await await source
    .collection('myfields')
    .find({})
    .toArray()
    .then(async documents => {
      await documents.map(field => {
        fields[field._id] = field.name
      })
    })
}

async function process(collectionName, documents) {
  if (collectionName === 'myothercollection') {
    // do something with collection myothercollection ..
    // for example: change '_id' with the explicit 'name'
    documents = await documents.map(document => {
      const { field } = document
      return {
        ...document,
        field: fields[field],
      }
    })
  } else if (collectionName === 'other') {
    // do something with collection other ..
  }

  // .. BUT remember to return documents
  return documents
}

In this example, the method initialize read all static data that are necessary to the second method for substitute field id with the name.