# Export to Flat File

Zuar Runner can output data to delimited (e.g. comma, tab, pipe, etc) flat files.

.. image:: assets/flat-file-1.png
   :alt: database table to csv

## Outputting to a flat file with a custom IO job

Here's an example SQL database table we will export to CSV using a modified
[Query job](/jobs/query/).

.. figure:: assets/export-to-flat-file__table.png

Sample data for toCSV

Use the [Generic plugin](/jobs/generic/) to create a custom IO
job.

Set the `type` to `io`.

Use the job config below:
```json
{
	"input": {
		"dbo": "postgresql://db/analytics",
		"query": [
			"SELECT * from excel.tennis_products;"
		],
		"use": "query.io#QueryInput"
	},
	"output": {
		"path": "/var/mitto/data/output_{year}_{month}_{day}.csv",
	  "delimiter": ",",
		"use": "call:mitto.iov2#tocsv"
	},
	"steps": [
		{
			"transforms": [
				{
					"rename_columns": false,
					"use": "mitto.iov2.transform#ExtraColumnsTransform",
					"include_empty_columns": true
				}
			],
			"use": "mitto.iov2.steps#Input"
		},
		{
			"transforms": [
				{
					"use": "mitto.iov2.transform#FlattenTransform"
				}
			],
			"use": "mitto.iov2.steps#Output"
		}
	]
}
```

## Job Config Keys

The `input` below is used for a [Query job](/jobs/query/), but the
key components when exporting to a flat file are the `output` and the
`steps`. The input can be an input from any IO job.

### input

Contains the database and query: `SELECT * from excel.tennis_products;`. Any
`SELECT` statement will work. Zuar Runner will use the column names for headers.

### output

Contains options that affect the output file the job will create,
including the output path, and the delimiter to use.

**Output path**: `/var/mitto/data/` is the top-level directory in Runner's file system. Files
in this directory (or sub-directories) will be visible in the `Files` page within Runner's admin UI.

**Delimiter**: In this example we're using `,` as a delimiter. For tabs use: `\t`

**Exclude the file header**: use `write_header: false` in the output section to
prevent column names from being included in the output file.

The output code uses the Python `csv` package. Any key word arguments can be
passed to `csv.writer`
([https://docs.python.org/3/library/csv.html#csv.writer](https://docs.python.org/3/library/csv.html#csv.writer)).

#### Path Variables

It is possible to include date and time variables in your output
path/filenames. In our example
`/var/mitto/data/output_{year}_*{month}_*{day}.csv`, the strings inside
curly brackets `{}` are variables; in this case the current year, month and
day.

Available Variables:

- **year** - The current year (2020)
- **month** - The current month (05)
- **day** - The current day (26)
- **hour** - The current hour (07)
- **start_time** - Date and time the job started (05-26-2020T07:30:30)
- **last_year** - 2019
- **last_month** - 04
- **last_day** - 25
- **last_hour** - 06
- **next_year** - 2021
- **next_month** - 06
- **next_day** - 27
- **next_hour** - 08
- **week** - The week number for week of the year. (01-52)

You can also pull out values from the **last** **and** **next** **values**:

- Yesterday's day is `{last_day.day}`, month is `{last_day.month}`, and year is `{last_day.year}`.

These are more complex examples using [Python's
**strftime**](https://docs.python.org/3/library/datetime.html#strftime-strptime-behavior)
on date values:

- Today is `{now:%A}, {now:%B} {now:%d}, {now:%Y}`
- One year ago from today was `{last_year:%c}`
- The time is `{now:%X}`
- The time is `{now:%I}:{minute}:{second} {now:%p}`

### steps

Any steps for processing the data. The steps included in the example above are
default steps which will include empty columns (via `"include_empty_columns":
true`) and will not rename columns (via `"rename_columns": false`).

## Resulting Flat File

The resulting flat file is created on [Zuar Runner's file system](/user-interface/files/).

.. figure:: assets/export-to-flat-file__output_file.png

Output from toCSV job

You can download the file manually or use another job
(e.g. [email](/jobs/email-smtp/),
[Rclone](/jobs/rclone/), [command
line](/jobs/command-line/), etc) to send the file elsewhere
automatically.