# Configurations

Hemera indexer can read configuration from cmd line arguments or environment variables.

* If you run Hemera Indexer in Docker, then the environment variable is easier to configure.
* If you prefer running from Source Code, command line arguments are more intuitive.

Run with `python hemera.py stream --help` to get the latest instructions for arguments.

### Parameters

* If the name of the parameter is in `UPPER_CASE` then it's an environment variable.
* If the name of the parameter starts with `--` then it's a command line argument.

Avoid specifying the same parameter from both the environment variable and the command line argument.

**`PROVIDER_URI` or `--provider-uri`**

\[**Default**: `https://mainnet.infura.io`]

The URI of the web3 rpc provider, e.g. `file://$HOME/Library/Ethereum/geth.ipc` or `https://mainnet.infura.io`.

**`DEBUG_PROVIDER_URI` or `--debug-provider-uri`**

\[**Default**: `https://mainnet.infura.io`]

The URI of the web3 debug rpc provider, e.g. `file://$HOME/Library/Ethereum/geth.ipc` or `https://mainnet.infura.io`.

**`POSTGRES_URL` or `--postgres-url`**

\[**Required**] The PostgreSQL connection URL that the Hemera Indexer used to maintain its state. e.g. `postgresql+psycopg2://user:password@127.0.0.1:5432/postgres`.

**`OUTPUT` or `--output` or `-o`**

\[**Required**] You may specify the output parameter so Hemera Indexer will export the data to CSV or JSON files. If not specified the data will be printed to the console.

If you have multiple outputs, use "," to concat the files. The file location will be relative to your current location if you run from source code, or the `output` folder as configured in `docker-compose.yaml`.

e.g.

* `postgresql+psycopg2://user:password@localhost:5432/hemera_indexer`: Output will be exported to your postgres.
* `jsonfile://output/json`: Json files will be exported to folder `output/json`
* `csvfile://output/csv`: Csv files will be exported to folder `output/csv`
* `console,jsonfile://output/json,csvfile://output/csv`: Multiple destinations are supported.

**`ENTITY_TYPES` or `--entity-types` or `-E`**

\[**Default**: `EXPLORER_BASE,EXPLORER_TOKEN`] The list of entity types to export. e.g. `EXPLORER_BASE`, `EXPLORER_TOKEN`, `EXPLORER_TRACE`.

**`OUTPUT_TYPES` or `--output-types` or `-O`**

The list of output types to export, corresponding to more detailed data models. Specifying this option will prioritize these settings over the entity types specified in -E. Available options include: block, transaction, log, token, address\_token\_balance, erc20\_token\_transfer, erc721\_token\_transfer, erc1155\_token\_transfer, trace, contract, coin\_balance.

You may spawn up multiple Hemera Indexer processes, each of them specifying different output types to accelerate the indexing process. For example, indexing `trace` data may take much longer than other entities, you may want to run a separate process to index `trace` data. Checkout `docker-compose/docker-compose.yaml` for examples.

**`DB_VERSION` or `--db-version`**

\[**Default**: `head`] The database version to initialize the database. Using the Alembic script's revision ID to specify a version.\
e.g. `head`, indicates the latest version.\
Or `base`, indicates the empty database without any table.\
Default value: `head`

**`START_BLOCK` or `--start-block`**

The block number to start from, e.g. `0`, `1000`, etc. If you don't specify this, Hemera Indexer will read the last synced block from the PostgreSQL database and resume from it.

**`END_BLOCK` or `--end-block`**

The block number that ends at, e.g. `10000`, `20000`, etc.

**`PARTITION_SIZE` or `--partition-size`**

\[**Default**: `50000`] The number of records to write to each file.

**`PERIOD_SECONDS` or `--period-seconds`**

\[**Default**: `10`] Seconds to sleep between each sync with the latest blockchain state.

**`BATCH_SIZE` or `--batch-size`**

\[**Default**: `10`] The number of non-debug rpc requests to batch in a single request.

**`DEBUG_BATCH_SIZE` or `--debug-batch-size`**

\[**Default**: `1`] The number of debug rpc to batch in a single request.

**`BLOCK_BATCH_SIZE` or `--block-batch-size`**

\[**Default**: `1`] The number of blocks to batch in a single sync round.

**`MAX_WORKERS` or `--max-workers`**

\[**Default**: `5`] The number of workers, e.g. `4`, `5`, etc.

**`LOG_FILE` or `--log-file`**

The log file to use. e.g. `path/to/logfile.log`.