---
title: Expressions
shortTitle: Expressions
intro: You can evaluate expressions in workflows and actions.
versions:
  fpt: '*'
  ghes: '*'
  ghec: '*'
---

{% data reusables.actions.enterprise-github-hosted-runners %}

## About expressions

You can use expressions to programmatically set environment variables in workflow files and access contexts. An expression can be any combination of literal values, references to a context, or functions. You can combine literals, context references, and functions using operators. For more information about contexts, see "[AUTOTITLE](/actions/learn-github-actions/contexts)."

Expressions are commonly used with the conditional `if` keyword in a workflow file to determine whether a step should run. When an `if` conditional is `true`, the step will run.

{% data reusables.actions.expressions-syntax-evaluation %}

{% raw %}
`${{ <expression> }}`
{% endraw %}

{% note %}

**Note**: The exception to this rule is when you are using expressions in an `if` clause, where, optionally, you can usually omit {% raw %}`${{`{% endraw %} and {% raw %}`}}`{% endraw %}. For more information about `if` conditionals, see "[AUTOTITLE](/actions/using-workflows/workflow-syntax-for-github-actions#jobsjob_idif)."

{% endnote %}

{% data reusables.actions.context-injection-warning %}

### Example setting an environment variable

{% raw %}

```yaml
env:
  MY_ENV_VAR: ${{ <expression> }}
```

{% endraw %}

## Literals

As part of an expression, you can use `boolean`, `null`, `number`, or `string` data types.

| Data type | Literal value |
|-----------|---------------|
| `boolean` | `true` or `false` |
| `null`    | `null` |
| `number`  | Any number format supported by JSON. |
| `string`  | You don't need to enclose strings in `{% raw %}${{{% endraw %}` and `{% raw %}}}{% endraw %}`. However, if you do, you must use single quotes (`'`) around the string. To use a literal single quote, escape the literal single quote using an additional single quote (`''`). Wrapping with double quotes (`"`) will throw an error. |

Note that in conditionals, falsy values (`false`, `0`, `-0`, `""`, `''`, `null`) are coerced to `false` and truthy (`true` and other non-falsy values) are coerced to `true`.

### Example of literals

{% raw %}

```yaml
env:
  myNull: ${{ null }}
  myBoolean: ${{ false }}
  myIntegerNumber: ${{ 711 }}
  myFloatNumber: ${{ -9.2 }}
  myHexNumber: ${{ 0xff }}
  myExponentialNumber: ${{ -2.99e-2 }}
  myString: Mona the Octocat
  myStringInBraces: ${{ 'It''s open source!' }}
```

{% endraw %}

## Operators

| Operator    | Description |
| ---         | ---         |
| `( )`       | Logical grouping |
| `[ ]`       | Index |
| `.`         | Property de-reference |
| `!`         | Not |
| `<`         | Less than |
| `<=`        | Less than or equal |
| `>`         | Greater than |
| `>=`        | Greater than or equal |
| `==`        | Equal |
| `!=`        | Not equal |
| `&&`        | And |
|  <code>\|\|</code> | Or |

  {% note %}

  **Notes:**
  * {% data variables.product.company_short %} ignores case when comparing strings.
  * `steps.<step_id>.outputs.<output_name>` evaluates as a string. {% data reusables.actions.expressions-syntax-evaluation %} For more information, see "[AUTOTITLE](/actions/learn-github-actions/contexts#steps-context)."
  * For numerical comparison, the `fromJSON()` function can be used to convert a string to a number. For more information on the `fromJSON()` function, see "[fromJSON](#fromjson)."

  {% endnote %}

{% data variables.product.prodname_dotcom %} performs loose equality comparisons.

* If the types do not match, {% data variables.product.prodname_dotcom %} coerces the type to a number. {% data variables.product.prodname_dotcom %} casts data types to a number using these conversions:

  | Type    | Result |
  | ---     | ---    |
  | Null    | `0` |
  | Boolean | `true` returns `1` <br /> `false` returns `0` |
  | String  | Parsed from any legal JSON number format, otherwise `NaN`. <br /> Note: empty string returns `0`. |
  | Array   | `NaN` |
  | Object  | `NaN` |
* When `NaN` is one of the operands of any relational comparison (`>`, `<`, `>=`, `<=`), the result is always `false`. For more information, see the "[NaN Mozilla docs](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/NaN)."
* {% data variables.product.prodname_dotcom %} ignores case when comparing strings.
* Objects and arrays are only considered equal when they are the same instance.

{% data variables.product.prodname_dotcom %} offers ternary operator like behaviour that you can use in expressions. By using a ternary operator in this way, you can dynamically set the value of an environment variable based on a condition, without having to write separate if-else blocks for each possible option.

### Example

{% raw %}

```yaml
env:
  MY_ENV_VAR: ${{ github.ref == 'refs/heads/main' && 'value_for_main_branch' || 'value_for_other_branches' }}
```

{% endraw %}

In this example, we're using a ternary operator to set the value of the `MY_ENV_VAR` environment variable based on whether the {% data variables.product.prodname_dotcom %} reference is set to `refs/heads/main` or not. If it is, the variable is set to `value_for_main_branch`. Otherwise, it is set to `value_for_other_branches`.
It is important to note that the first value after the `&&` must be truthy. Otherwise, the value after the `||` will always be returned.

## Functions

{% data variables.product.prodname_dotcom %} offers a set of built-in functions that you can use in expressions. Some functions cast values to a string to perform comparisons. {% data variables.product.prodname_dotcom %} casts data types to a string using these conversions:

| Type    | Result |
| ---     | ---    |
| Null    | `''` |
| Boolean | `'true'` or `'false'` |
| Number  | Decimal format, exponential for large numbers |
| Array   | Arrays are not converted to a string |
| Object  | Objects are not converted to a string |

### contains

`contains( search, item )`

Returns `true` if `search` contains `item`. If `search` is an array, this function returns `true` if the `item` is an element in the array. If `search` is a string, this function returns `true` if the `item` is a substring of `search`. This function is not case sensitive. Casts values to a string.

#### Example using a string

`contains('Hello world', 'llo')` returns `true`.

#### Example using an object filter

`contains(github.event.issue.labels.*.name, 'bug')` returns `true` if the issue related to the event has a label "bug".

For more information, see "[Object filters](#object-filters)."

#### Example matching an array of strings

Instead of writing `github.event_name == "push" || github.event_name == "pull_request"`, you can use `contains()` with `fromJSON()` to check if an array of strings contains an `item`.

For example, `contains(fromJSON('["push", "pull_request"]'), github.event_name)` returns `true` if `github.event_name` is "push" or "pull_request".

### startsWith

`startsWith( searchString, searchValue )`

Returns `true` when `searchString` starts with `searchValue`. This function is not case sensitive. Casts values to a string.

#### Example of `startsWith`

`startsWith('Hello world', 'He')` returns `true`.

### endsWith

`endsWith( searchString, searchValue )`

Returns `true` if `searchString` ends with `searchValue`. This function is not case sensitive. Casts values to a string.

#### Example of `endsWith`

`endsWith('Hello world', 'ld')` returns `true`.

### format

`format( string, replaceValue0, replaceValue1, ..., replaceValueN)`

Replaces values in the `string`, with the variable `replaceValueN`. Variables in the `string` are specified using the `{N}` syntax, where `N` is an integer. You must specify at least one `replaceValue` and `string`. There is no maximum for the number of variables (`replaceValueN`) you can use. Escape curly braces using double braces.

#### Example of `format`

{% raw %}

```javascript
format('Hello {0} {1} {2}', 'Mona', 'the', 'Octocat')
```

{% endraw %}

Returns 'Hello Mona the Octocat'.

#### Example escaping braces

{% raw %}

```javascript
format('{{Hello {0} {1} {2}!}}', 'Mona', 'the', 'Octocat')
```

{% endraw %}

Returns '{Hello Mona the Octocat!}'.

### join

`join( array, optionalSeparator )`

The value for `array` can be an array or a string. All values in `array` are concatenated into a string. If you provide `optionalSeparator`, it is inserted between the concatenated values. Otherwise, the default separator `,` is used. Casts values to a string.

#### Example of `join`

`join(github.event.issue.labels.*.name, ', ')` may return 'bug, help wanted'

### toJSON

`toJSON(value)`

Returns a pretty-print JSON representation of `value`. You can use this function to debug the information provided in contexts.

#### Example of `toJSON`

`toJSON(job)` might return `{ "status": "success" }`

### fromJSON

`fromJSON(value)`

Returns a JSON object or JSON data type for `value`. You can use this function to provide a JSON object as an evaluated expression or to convert any data type that can be represented in JSON or JavaScript, such as strings, booleans, null values, arrays, and objects.

#### Example returning a JSON object

This workflow sets a JSON matrix in one job, and passes it to the next job using an output and `fromJSON`.

{% raw %}

```yaml copy
name: build
on: push
jobs:
  job1:
    runs-on: ubuntu-latest
    outputs:
      matrix: ${{ steps.set-matrix.outputs.matrix }}
    steps:
      - id: set-matrix
        run: echo "matrix={\"include\":[{\"project\":\"foo\",\"config\":\"Debug\"},{\"project\":\"bar\",\"config\":\"Release\"}]}" >> $GITHUB_OUTPUT
  job2:
    needs: job1
    runs-on: ubuntu-latest
    strategy:
      matrix: ${{ fromJSON(needs.job1.outputs.matrix) }}
    steps:
      - run: echo "Matrix - Project ${{ matrix.project }}, Config ${{ matrix.config }}"
```

{% endraw %}

#### Example returning a JSON data type

This workflow uses `fromJSON` to convert environment variables from a string to a Boolean or integer.

```yaml copy
name: print
on: push
env:
  continue: true
  time: 3
jobs:
  job1:
    runs-on: ubuntu-latest
    steps:
      - continue-on-error: {% raw %}${{ fromJSON(env.continue) }}{% endraw %}
        timeout-minutes: {% raw %}${{ fromJSON(env.time) }}{% endraw %}
        run: echo ...
```

The workflow uses the `fromJSON()` function to convert the environment variable `continue` from a string to a boolean, allowing it to determine whether to continue-on-error or not. Similarly, it converts the `time` environment variable from a string to an integer, setting the timeout for the job in minutes.

### hashFiles

`hashFiles(path)`

Returns a single hash for the set of files that matches the `path` pattern. You can provide a single `path` pattern or multiple `path` patterns separated by commas. The `path` is relative to the `GITHUB_WORKSPACE` directory and can only include files inside of the `GITHUB_WORKSPACE`. This function calculates an individual SHA-256 hash for each matched file, and then uses those hashes to calculate a final SHA-256 hash for the set of files. If the `path` pattern does not match any files, this returns an empty string. For more information about SHA-256, see "[SHA-2](https://en.wikipedia.org/wiki/SHA-2)."

You can use pattern matching characters to match file names. Pattern matching for `hashFiles` follows glob pattern matching and is case-insensitive on Windows. For more information about supported pattern matching characters, see the [Patterns](https://www.npmjs.com/package/@actions/glob#patterns) section in the `@actions/glob` documentation.

#### Example with a single pattern

Matches any `package-lock.json` file in the repository.

`hashFiles('**/package-lock.json')`

#### Example with multiple patterns

Creates a hash for any `package-lock.json` and `Gemfile.lock` files in the repository.

`hashFiles('**/package-lock.json', '**/Gemfile.lock')`

## Status check functions

You can use the following status check functions as expressions in `if` conditionals. A default status check of `success()` is applied unless you include one of these functions. For more information about `if` conditionals, see "[AUTOTITLE](/actions/using-workflows/workflow-syntax-for-github-actions#jobsjob_idif)" and "[AUTOTITLE](/actions/creating-actions/metadata-syntax-for-github-actions#runsstepsif)".

### success

Returns `true` when all previous steps have succeeded.

#### Example of `success`

```yaml
steps:
  ...
  - name: The job has succeeded
    if: {% raw %}${{ success() }}{% endraw %}
```

### always

Causes the step to always execute, and returns `true`, even when canceled. The `always` expression is best used at the step level or on tasks that you expect to run even when a job is canceled. For example, you can use `always` to send logs even when a job is canceled.

{% warning %}

**Warning:** Avoid using `always` for any task that could suffer from a critical failure, for example: getting sources, otherwise the workflow may hang until it times out. If you want to run a job or step regardless of its success or failure, use the recommended alternative: `if: {% raw %}${{ !cancelled() }}{% endraw %}`

{% endwarning %}

#### Example of `always`

```yaml
if: {% raw %}${{ always() }}{% endraw %}
```

### cancelled

Returns `true` if the workflow was canceled.

#### Example of `cancelled`

```yaml
if: {% raw %}${{ cancelled() }}{% endraw %}
```

### failure

Returns `true` when any previous step of a job fails. If you have a chain of dependent jobs, `failure()` returns `true` if any ancestor job fails.

#### Example of `failure`

```yaml
steps:
  ...
  - name: The job has failed
    if: {% raw %}${{ failure() }}{% endraw %}
```

#### failure with conditions

You can include extra conditions for a step to run after a failure, but you must still include `failure()` to override the default status check of `success()` that is automatically applied to `if` conditions that don't contain a status check function.

##### Example of `failure` with conditions

```yaml
steps:
  ...
  - name: Failing step
    id: demo
    run: exit 1
  - name: The demo step has failed
    if: {% raw %}${{ failure() && steps.demo.conclusion == 'failure' }}{% endraw %}
```

## Object filters

You can use the `*` syntax to apply a filter and select matching items in a collection.

For example, consider an array of objects named `fruits`.

```json
[
  { "name": "apple", "quantity": 1 },
  { "name": "orange", "quantity": 2 },
  { "name": "pear", "quantity": 1 }
]
```

The filter `fruits.*.name` returns the array `[ "apple", "orange", "pear" ]`.

You may also use the `*` syntax on an object. For example, suppose you have an object named `vegetables`.

```json

{
  "scallions":
  {
    "colors": ["green", "white", "red"],
    "ediblePortions": ["roots", "stalks"],
  },
  "beets":
  {
    "colors": ["purple", "red", "gold", "white", "pink"],
    "ediblePortions": ["roots", "stems", "leaves"],
  },
  "artichokes":
  {
    "colors": ["green", "purple", "red", "black"],
    "ediblePortions": ["hearts", "stems", "leaves"],
  },
}
```

The filter `vegetables.*.ediblePortions` could evaluate to:

```json

[
  ["roots", "stalks"],
  ["hearts", "stems", "leaves"],
  ["roots", "stems", "leaves"],
]
```

Since objects don't preserve order, the order of the output cannot be guaranteed.