thombashi / elasticsearch-faker

elasticsearch-faker is a CLI tool to generate fake data for Elasticsearch.

elasticsearch-faker

Summary

elasticsearch-faker is a CLI tool to generate fake data for Elasticsearch.

Installation

Installation: pip

pip install elasticsearch-faker

Installation: dpkg (Ubuntu)

1. Navigate to Releases page
2. Download the latest deb package
3. Install with dpkg -i command

Installation: Docker container

Packages page

Usage

Command help

Usage: elasticsearch-faker [OPTIONS] COMMAND [ARGS]...

  Faker for Elasticsearch.

Options:
  --version                       Show the version and exit.
  --debug                         For debug print.
  -q, --quiet                     Suppress execution log messages.
  -v, --verbose
  --locale [ar_EG|ar_PS|ar_SA|bs_BA|bg_BG|cs_CZ|de_DE|dk_DK|el_GR|en_AU|en_CA|en_GB|en_NZ|en_US|es_ES|es_MX|et_EE|fa_IR|fi_FI|fr_FR|hi_IN|hr_HR|hu_HU|it_IT|ja_JP|ko_KR|lt_LT|lv_LV|ne_NP|nl_NL|no_NO|pl_PL|pt_BR|pt_PT|ro_RO|ru_RU|sl_SI|sv_SE|tr_TR|uk_UA|zh_CN|zh_TW|ka_GE]
                                  Specify localization for fake data. Defaults
                                  to en_US.
  --seed INTEGER                  Random seed for faker.
  --basic-auth-user TEXT          User name for Elasticsearch basic
                                  authentication. Or you can set the value via
                                  ES_BASIC_AUTH_USER environment variable.
  --basic-auth-password TEXT      Password for Elasticsearch basic
                                  authentication. Or you can set the value via
                                  ES_BASIC_AUTH_PASSWORD environment variable.
  --verify-certs                  Verify Elasticsearch server certificate. Or
                                  you can set the value via
                                  ES_SSL_ASSERT_FINGERPRINT environment
                                  variable.
  --ssl-assert-fingerprint TEXT   SSL certificate fingerprint to verify.
  --ignore-es-warn                Ignore ElasticsearchWarning.
  -h, --help                      Show this message and exit.

Commands:
  generate    Generate fake data and put it to an Elasticsearch index.
  provider    Show or search providers for doc templates.
  show-stats  Fetch and show statistics of an index.
  validate    Check that a faker doc template file is well formed.
  version     Show version information.

  Issue tracker: https://github.com/thombashi/elasticsearch-faker/issues

Usage: elasticsearch-faker generate [OPTIONS] ENDPOINT

  Generate fake data and put it to an Elasticsearch index.

Options:
  --index NAME                    Name of an index to create. Defaults to
                                  'test_index'.
  --mapping PATH                  Path to a mapping file. See also https://www
                                  .elastic.co/guide/en/elasticsearch/reference
                                  /current/explicit-mapping.html
  --doc-template, --template PATH
                                  Path to a faker doc template file.
  -n, --num-doc INTEGER           Number of generating documents. The command
                                  uses bulk API if the value equals or is
                                  greater than two. Defaults to 1000.
  --bulk-size INTEGER             Number of creating documents for a single
                                  bulk API call. Defaults to 200.
  --delete-index                  Delete the index if already exists before
                                  generating documents.
  -j, --jobs INTEGER              Number of workers that create docs. Defaults
                                  to 1.
  --stdin                         Read a faker doc template from stdin.
  --dry-run                       Do no harm.
  -h, --help                      Show this message and exit.

  Issue tracker: https://github.com/thombashi/elasticsearch-faker/issues

Execution example

Create 1000 docs to an Elasticsearch index

Execution

$ elasticsearch-faker generate --doc-template doc_template.jinja2 https://localhost:9200 -n 1000
document generator #0: 100%|█████████████████████| 1000/1000 [00:01<00:00, 590.53docs/s]
[INFO] generate 1000 docs to test_index

[Results]
target index: test_index
completed in 10.4 secs
current store.size: 3.0 MB
current docs.count: 1,000
generated store.size: 3.0 MB
average size[byte]/doc: 3,164
generated docs.count: 1,000
generated docs/secs: 96.3
bulk size: 200
$ curl -sS localhost:9200/test_index/_search | jq .hits.hits[:2]
[
  {
    "_index": "test_index",
    "_id": "4bdd73c0-7744-4c6f-9736-50e3e8515f1c-0",
    "_score": 1,
    "_source": {
      "name": "jennifer17",
      "userId": 56561230,
      "createdAt": "2009-07-17T06:31:04.000+0000",
      "body": "Present blue happen thus miss toward. Itself race so successful build real beyond score. Look different she receive.Compare miss federal lawyer. Herself prevent approach east.",
      "ext": "course",
      "blobId": "c35769a9-3468-43fc-93c7-3c2f27ec9f64"
    }
  },
  {
    "_index": "test_index",
    "_id": "88238d96-5ecc-4639-bb8f-c3f816027560-0",
    "_score": 1,
    "_source": {
      "name": "dnicholson",
      "userId": 457,
      "createdAt": "2008-08-29T22:14:43.000+0000",
      "body": "I sit another health president bring. Very expect international television job parent into.Authority read few stock. International hope yard left measure.Player them get move.",
      "ext": "trial",
      "blobId": "e43faf58-9b66-4a43-b1b7-7540b3996cde"
    }
  }
]

doc template file (doc_template.jinja2)

{
  "name": "{{ user_name }}",
  "userId": {{ random_number }},
  "createdAt": "{{ date_time }}",
  "body": "{{ text }}",
  "ext": "{{ word }}",
  "blobId": "{{ uuid4 }}"
}

{{ XXX }} in the template file indicates the used providers of Faker to generate data. The available providers can be listed by elasticsearch-faker provider list / elasticsearch-faker provider example subcommands.

Use Elasticsearch authentication

Execution

$ export ES_BASIC_AUTH_USER=elastic
$ export ES_BASIC_AUTH_PASSWORD=<PASSWORD>
$ export ES_SSL_ASSERT_FINGERPRINT=<HTTP CA certificate SHA-256 fingerprint>

$ elasticsearch-faker --verify-certs generate --doc-template doc_template.jinja2 https://localhost:9200 -n 1000
[INFO] generate 1000 docs to test_index

[Results]
target index: test_index
completed in 0.7 secs
current store.size: 3.9 MB
current docs.count: 6,000
generated store.size: 0.0 MB
average size[byte]/doc: 690
generated docs.count: 1,000
generated docs/secs: 1,338.6
bulk size: 200

$ curl --insecure -sS https://${ES_BASIC_AUTH_USER}:${ES_BASIC_AUTH_PASSWORD}@localhost:9200/test_index/_search | jq .hits.hits[:2]
[
  {
    "_index": "test_index",
    "_id": "8PMd9ocBtCWmUGxHBM9L",
    "_score": 1,
    "_source": {
      "name": "lclarke",
      "userId": 331837,
      "createdAt": "1980-07-18T23:42:30.000+0000",
      "body": "Large address animal husband present. In act call animal.Yes plant pressure year me.",
      "ext": "series",
      "blobId": "ede46099-ac97-4447-b86b-0a87ef0180f1"
    }
  },
  {
    "_index": "test_index",
    "_id": "71b76118-91fa-4ed3-a1e0-305694b3d34d-0",
    "_score": 1,
    "_source": {
      "name": "shawnyoder",
      "userId": 80039293,
      "createdAt": "1972-09-28T19:04:31.000+0000",
      "body": "Book television political surface fill position security itself. Not man support attorney attorney which amount finish. Ground mother board natural wait about lot.",
      "ext": "before",
      "blobId": "8913b0a4-dd44-442a-8961-a6be87eb68a6"
    }
  }
]

Or without --verify-certs option:

Execution

$ export ES_BASIC_AUTH_USER=elastic
$ export ES_BASIC_AUTH_PASSWORD=<PASSWORD>

$ elasticsearch-faker generate --doc-template doc_template.jinja2 https://localhost:9200 -n 1000

Dependencies

• Elasticsearch 8 or newer
• Python 3.7+
• Python package dependencies (automatically installed)