Packer
Expressions
Note: This page is about HCL2 Packer templates. HCL2 templates were first introduced as a beta feature into Packer version 1.5. As of v1.7, HCL2 support is no longer in beta, and is the preferred way to write Packer configuration. For the old-style stable configuration language see template docs. As of v1.6.2, you can convert your legacy JSON template into an HCL2 config file using the hcl2_upgrade command.
Expressions are used to refer to or compute values within a configuration.
The simplest expressions are just literal values, like "hello"
or 5
, but
HCL also allows more complex expressions such as references to data exported by
sources, arithmetic, conditional evaluation, and a number of built-in
functions.
Expressions can be used in a number of places in HCL, but some contexts limit which expression constructs are allowed, such as requiring a literal value of a particular type or forbidding. Each language feature's documentation describes any restrictions it places on expressions.
The rest of this page describes all of the features of Packer's expression syntax.
Types and Values
The result of an expression is a value. All values have a type, which dictates where that value can be used and what transformations can be applied to it.
HCL uses the following types for its values:
string
: a sequence of Unicode characters representing some text, like"hello"
.number
: a numeric value. Thenumber
type can represent both whole numbers like15
and fractional values like6.283185
.bool
: eithertrue
orfalse
.bool
values can be used in conditional logic.list
(ortuple
): a sequence of values, like["us-west-1a", "us-west-1c"]
. Elements in a list or tuple are identified by consecutive whole numbers, starting with zero.map
(orobject
): a group of values identified by named labels, like{name = "Mabel", age = 52}
.
Strings, numbers, and bools are sometimes called primitive types. Lists/tuples and maps/objects are sometimes called complex types, structural types, or collection types.
Finally, there is one special value that has no type:
null
: a value that represents absence or omission. If you set an argument of a source or module tonull
, Packer behaves as though you had completely omitted it — it will use the argument's default value if it has one, or raise an error if the argument is mandatory.null
is most useful in conditional expressions, so you can dynamically omit an argument if a condition isn't met.
Advanced Type Details
In most situations, lists and tuples behave identically, as do maps and objects. Whenever the distinction isn't relevant, the Packer documentation uses each pair of terms interchangeably (with a historical preference for "list" and "map").
However, plugin authors should understand the differences between these similar
types (and the related set
type), since they offer different ways to restrict
the allowed values for input variables and source arguments.
Type Conversion
Expressions are most often used to set values for arguments. In these cases, the argument has an expected type and the given expression must produce a value of that type.
Where possible, Packer automatically converts values from one type to another in order to produce the expected type. If this isn't possible, Packer will produce a type mismatch error and you must update the configuration with a more suitable expression.
Packer automatically converts number and bool values to strings when needed. It also converts strings to numbers or bools, as long as the string contains a valid representation of a number or bool value.
true
converts to"true"
, and vice-versafalse
converts to"false"
, and vice-versa15
converts to"15"
, and vice-versa
Literal Expressions
A literal expression is an expression that directly represents a particular constant value. Packer has a literal expression syntax for each of the value types described above:
Strings are usually represented by a double-quoted sequence of Unicode characters,
"like this"
. There is also a "heredoc" syntax for more complex strings. String literals are the most complex kind of literal expression in Packer, and have additional documentation on this page:- See String Literals below for information about escape sequences and the heredoc syntax.
- See String Templates below for information about interpolation and template directives.
Numbers are represented by unquoted sequences of digits with or without a decimal point, like
15
or6.283185
.Bools are represented by the unquoted symbols
true
andfalse
.The null value is represented by the unquoted symbol
null
.Lists/tuples are represented by a pair of square brackets containing a comma-separated sequence of values, like
["a", 15, true]
.List literals can be split into multiple lines for readability, but always require a comma between values. A comma after the final value is allowed, but not required. Values in a list can be arbitrary expressions.
Maps/objects are represented by a pair of curly braces containing a series of
<KEY> = <VALUE>
pairs:{ name = "John" age = 52 }
Key/value pairs can be separated by either a comma or a line break. Values can be arbitrary expressions. Keys are strings; they can be left unquoted if they are a valid identifier, but must be quoted otherwise. You can use a non-literal expression as a key by wrapping it in parentheses, like
(var.business_unit_tag_name) = "SRE"
.
References to Named Values
Packer makes one named values available.
The following named values are available:
source.<SOURCE TYPE>.<NAME>
is an object representing a source of the given type and name.
Available Functions
For a full list of available functions, see the function reference.
for
Expressions
A for
expression creates a complex type value by transforming
another complex type value. Each element in the input value
can correspond to either one or zero values in the result, and an arbitrary
expression can be used to transform each input element into an output element.
For example, if var.list
is a list of strings, then the following expression
produces a list of strings with all-uppercase letters:
[for s in var.list : upper(s)]
This for
expression iterates over each element of var.list
, and then
evaluates the expression upper(s)
with s
set to each respective element.
It then builds a new tuple value with all of the results of executing that
expression in the same order.
The type of brackets around the for
expression decide what type of result
it produces. The above example uses [
and ]
, which produces a tuple. If
{
and }
are used instead, the result is an object, and two result
expressions must be provided separated by the =>
symbol:
{for s in var.list : s => upper(s)}
This expression produces an object whose attributes are the original elements
from var.list
and their corresponding values are the uppercase versions.
A for
expression can also include an optional if
clause to filter elements
from the source collection, which can produce a value with fewer elements than
the source:
[for s in var.list : upper(s) if s != ""]
The source value can also be an object or map value, in which case two temporary variable names can be provided to access the keys and values respectively:
[for k, v in var.map : length(k) + length(v)]
Finally, if the result type is an object (using {
and }
delimiters) then
the value result expression can be followed by the ...
symbol to group
together results that have a common key:
{for s in var.list : substr(s, 0, 1) => s... if s != ""}
Splat Expressions
A splat expression provides a more concise way to express a common operation
that could otherwise be performed with a for
expression.
If var.list
is a list of objects that all have an attribute id
, then a list
of the IDs could be produced with the following for
expression:
[for o in var.list : o.id]
This is equivalent to the following splat expression:
var.list[*].id
The special [*]
symbol iterates over all of the elements of the list given to
its left and accesses from each one the attribute name given on its right. A
splat expression can also be used to access attributes and indexes from lists
of complex types by extending the sequence of operations to the right of the
symbol:
var.list[*].interfaces[0].name
The above expression is equivalent to the following for
expression:
[for o in var.list : o.interfaces[0].name]
Splat expressions are for lists only (and thus cannot be used to reference
resources created with
for_each
, which
are represented as maps). However, if a splat expression is applied to a value
that is not a list or tuple then the value is automatically wrapped in a
single-element list before processing.
For example, var.single_object[*].id
is equivalent to
[var.single_object][*].id
, or effectively [var.single_object.id]
. This
behavior is not interesting in most cases, but it is particularly useful when
referring to resources that may or may not have count
set, and thus may or
may not produce a tuple value:
aws_instance.example[*].id
The above will produce a list of IDs whether aws_instance.example
has count
set or not, avoiding the need to revise various other expressions in the
configuration when a particular resource switches to and from having count
set.
dynamic
blocks
Within top-level block constructs like source, expressions
can usually be used only when assigning a value to an argument using the name = expression
or key = expression
form.
This covers many uses, but some source types include repeatable nested blocks in their arguments, which do not accept expressions:
source "amazon-ebs" "example" {
name = "pkr-test-name" # can use expressions here
tag {
# but the "tag" block is always a literal block
}
}
You can dynamically construct repeatable nested blocks like tag
using a
special dynamic
block type, which is supported in top-level block constructs, example:
locals {
standard_tags = {
Component = "user-service"
Environment = "production"
}
}
source "amazon-ebs" "example" {
# ...
tag {
key = "Name"
value = "example-asg-name"
}
dynamic "tag" {
for_each = local.standard_tags
content {
key = tag.key
value = tag.value
}
}
}
A dynamic
block acts much like a for
expression, but produces nested blocks
instead of a complex typed value. It iterates over a given complex value, and
generates a nested block for each element of that complex value.
- The label of the dynamic block (
"tag"
in the example above) specifies what kind of nested block to generate. - The
for_each
argument provides the complex value to iterate over. - The
iterator
argument (optional) sets the name of a temporary variable that represents the current element of the complex value. If omitted, the name of the variable defaults to the label of thedynamic
block ("tag"
in the example above). - The
labels
argument (optional) is a list of strings that specifies the block labels, in order, to use for each generated block. You can use the temporary iterator variable in this value. - The nested
content
block defines the body of each generated block. You can use the temporary iterator variable inside this block.
Since the for_each
argument accepts any collection or structural value,
you can use a for
expression or splat expression to transform an existing
collection.
The iterator object (tag
in the example above) has two attributes:
key
is the map key or list element index for the current element. If thefor_each
expression produces a set value thenkey
is identical tovalue
and should not be used.value
is the value of the current element.
A dynamic
block can only generate arguments for nested blocks that belong to
the source type, data source, or provisioner being configured.
The for_each
value must be a map or set with one element per desired nested
block. If you need to declare resource instances based on a nested data
structure or combinations of elements from multiple data structures you can use
expressions and functions to derive a suitable value. For some common examples
of such situations, see the
flatten
and
setproduct
functions.
Best Practices for dynamic
Blocks
Overuse of dynamic
blocks can make configuration hard to read and maintain,
so we recommend using them only when you need to hide details in order to build
a clean user interface for a re-usable code. Always write nested blocks out
literally where possible.
String Literals
HCL has two different syntaxes for string literals. The
most common is to delimit the string with quote characters ("
), like
"hello"
. In quoted strings, the backslash character serves as an escape
sequence, with the following characters selecting the escape behavior:
Sequence | Replacement |
---|---|
\n | Newline |
\r | Carriage Return |
\t | Tab |
\" | Literal quote (without terminating the string) |
\\ | Literal backslash |
\uNNNN | Unicode character from the basic multilingual plane (NNNN is four hex digits) |
\UNNNNNNNN | Unicode character from supplementary planes (NNNNNNNN is eight hex digits) |
The alternative syntax for string literals is the so-called Here Documents or "heredoc" style, inspired by Unix shell languages. This style allows multi-line strings to be expressed more clearly by using a custom delimiter word on a line of its own to close the string:
<<EOF
hello
world
EOF
The <<
marker followed by any identifier at the end of a line introduces the
sequence. Packer then processes the following lines until it finds one that
consists entirely of the identifier given in the introducer. In the above
example, EOF
is the identifier selected. Any identifier is allowed, but
conventionally this identifier is in all-uppercase and begins with EO
, meaning
"end of". EOF
in this case stands for "end of text".
The "heredoc" form shown above requires that the lines following be flush with the left margin, which can be awkward when an expression is inside an indented block:
block {
value = <<EOF
hello
world
EOF
}
To improve on this, Packer also accepts an indented heredoc string variant
that is introduced by the <<-
sequence:
block {
value = <<-EOF
hello
world
EOF
}
In this case, Packer analyses the lines in the sequence to find the one with the smallest number of leading spaces, and then trims that many spaces from the beginning of all of the lines, leading to the following result:
hello
world
Backslash sequences are not interpreted in a heredoc string expression. Instead, the backslash character is interpreted literally.
In both quoted and heredoc string expressions, Packer supports template
sequences that begin with ${
and %{
. These are described in more detail
in the following section. To include these sequences literally without
beginning a template sequence, double the leading character: $${
or %%{
.
String Templates
Within quoted and heredoc string expressions, the sequences ${
and %{
begin
template sequences. Templates let you directly embed expressions into a string
literal, to dynamically construct strings from other values.
The following example demonstrates how a template sequence can be used to embed the value of a variable into a string that can be used as script content:
locals {
packages = ["git", "curl", "vim"]
install_packages = <<-EOF
#!/bin/bash
if [ ${length(local.packages)} -eq 0 ]; then
echo "No packages to install."
exit 1
fi
apt-get update
%{ for package in local.packages ~}
apt-get install -y ${package}
%{ endfor ~}
EOF
}
source "amazon-ebs" "example" {
# ...
}
build {
sources = ["source.amazon-ebs.example"]
provisioner "shell" {
inline = [local.install_packages]
}
}
This can be tested using the packer console
command:
$ packer source.pkr.hcl
> local.install_packages
> #!/bin/bash
if [ 3 -eq 0 ]; then
echo "No packages to install."
exit 1
fi
apt-get update
apt-get install -y git
apt-get install -y curl
apt-get install -y vim