Update README.md

[adibaba]

The current README.md seems to be incomplete.

Please check:

• WARNING: The DATASET_FOLDER variable is not set. Defaulting to a blank string.
• WARNING: The ES_INDEX variable is not set. Defaulting to a blank string.
• please add ex_logstash and bind q_logstash to it
• ERROR: for opal-converter_dataset-file-fetcher_1 Cannot create container for service dataset-file-fetcher: create .: volume name is too short, names should be at least two alphanumeric characters
• ERROR: for dataset-file-fetcher Cannot create container for service dataset-file-fetcher: create .: volume name is too short, names should be at least two alphanumeric characters

Successfully built c8f51b4e1f75
Successfully tagged opal-converter_elasticsearch-writer:latest
WARNING: The DATASET_FOLDER variable is not set. Defaulting to a blank string.
WARNING: The ES_INDEX variable is not set. Defaulting to a blank string.
Starting opal-converter_rabbitmq_1           ... done
Recreating opal-converter_discovery-server_1 ... done
Recreating opal-converter_es_1               ... done
==============
please add ex_logstash and bind q_logstash to it
==============
Press [ENTER] to continue,...: 
WARNING: The DATASET_FOLDER variable is not set. Defaulting to a blank string.
WARNING: The ES_INDEX variable is not set. Defaulting to a blank string.
opal-converter_rabbitmq_1 is up-to-date
opal-converter_es_1 is up-to-date
Recreating opal-converter_config-server_1 ... done
Starting opal-converter_logstash_1        ... done
Starting opal-converter_kibana_1          ... done
WARNING: The DATASET_FOLDER variable is not set. Defaulting to a blank string.
WARNING: The ES_INDEX variable is not set. Defaulting to a blank string.
opal-converter_discovery-server_1 is up-to-date
opal-converter_es_1 is up-to-date
opal-converter_rabbitmq_1 is up-to-date
opal-converter_config-server_1 is up-to-date
opal-converter_kibana_1 is up-to-date
opal-converter_logstash_1 is up-to-date
Recreating opal-converter_data-cleaner-service_1 ... 
Recreating opal-converter_opal-confirm-conversion-service_1 ... 
Recreating opal-converter_data-cleaner-service_1            ... done
Recreating opal-converter_opal-confirm-conversion-service_1 ... done
Recreating opal-converter_triplestore-writer_1              ... done
Recreating opal-converter_elasticsearch-writer_1            ... done
Recreating opal-converter_dataset-fetcher_1                 ... done
Recreating opal-converter_quality-metrics-service_1         ... done
ERROR: for opal-converter_dataset-file-fetcher_1  Cannot create container for service dataset-file-fetcher: create .: volume name is too short, names should be at least two alphanumeric characters

ERROR: for dataset-file-fetcher  Cannot create container for service dataset-file-fetcher: create .: volume name is too short, names should be at least two alphanumeric characters
ERROR: Encountered errors while bringing up the project.

[adibaba]

Regarding to the current docker-compose.yml and README.md:

These properties seem to be missing:

• DATASET_FOLDER (code)
• ES_INDEX (code)

If more than 1 developer is working on the component, should CONFIG_REPO (code) also become a variable? Then everybody can have an own configuration.
Additonally, there are already 2 branches, master and developer. Maybe it is handeled like this already?

Question: To have some local tests on a machine without much power, the CONFIG_REPO has to be used to configure the required resources, correct?

Question: are values like http://discovery-server:8761/eureka/ (code) replaced internally?

[mnafshin]

For CONFIG_REPO there are different ways to solve it,

1. You can set the branch that you want to work on it with $CONFIG_REPO_DEFAULT_BRANCH (if it is not set, the fallback value is master)
2. You can update the application.yml file in configuration component to read from a folder instead of an online git as it is explained in:
   https://cloud.spring.io/spring-cloud-config/multi/multi__spring_cloud_config_server.html#_file_system_backend
3. Or one can provide their own instead of URL,
   If you insist on making it a variable, I am OK to do that, but as I prefer to work with branch name, I think we can make the branch name a variable would be more readable and maintainable

[mnafshin]

For being able to run test, you can provide another profile in application.yml such as
/{application}-{profile}.yml
or inside the application.yml such as this link
or if it is a unit test you can provide your test version of application.yml inside
test/resources to replace the one in main/resources

then inside that, you can provide the config to read the configuration from a local git folder

(I believe there is no difference between them and you can choose one way that you prefer more)

[mnafshin]

EUREKA_URL is setting discovery server address in every module. if it is not set, the fallback value is localhost:8761/eureka
you can check them in converter-configuration, they will have a value like:

eureka:
  client:
    fetch-registry: true
    register-with-eureka: true
    serviceUrl:
      defaultZone: ${EUREKA_URL:http://localhost:8761/eureka/}

[mnafshin]

The ReadMe.md is updated with mentioning of the following variables:

DATASET_FOLDER
ES_INDEX