terraform/website/docs/configuration/functions/regex.html.md

168 lines
10 KiB
Markdown
Raw Normal View History

---
layout: "functions"
page_title: "regex - Functions - Configuration Language"
sidebar_current: "docs-funcs-string-regex"
description: |-
The regex function applies a regular expression to a string and returns the
matching substrings.
---
# `regex` Function
-> **Note:** This page is about Terraform 0.12 and later. For Terraform 0.11 and
earlier, see
[0.11 Configuration Language: Interpolation Syntax](../../configuration-0-11/interpolation.html).
`regex` applies a
[regular expression](https://en.wikipedia.org/wiki/Regular_expression)
to a string and returns the matching substrings.
```hcl
regex(pattern, string)
```
The return type of `regex` depends on the capture groups, if any, in the
pattern:
- If the pattern has no capture groups at all, the result is a single string
covering the substring matched by the pattern as a whole.
- If the pattern has one or more _unnamed_ capture groups, the result is a
list of the captured substrings in the same order as the definition of
the capture groups.
- If the pattern has one or more _named_ capture groups, the result is a
map of the captured substrings, using the capture group names as map keys.
It's not valid to mix both named and unnamed capture groups in the same pattern.
If the given pattern does not match at all, the `regex` raises an error. To
_test_ whether a given pattern matches a string, use
[`regexall`](./regexall.html) and test that the result has length greater than
zero.
The pattern is a string containing a mixture of literal characters and special
matching operators as described in the following table. Note that when giving a
regular expression pattern as a literal quoted string in the Terraform
language, the quoted string itself already uses backslash `\` as an escape
character for the string, so any backslashes intended to be recognized as part
of the pattern must be escaped as `\\`.
| Sequence | Matches |
| -------------- | -------------------------------------------------------------------------------- |
| `.` | Any character except newline |
| `[xyz]` | Any character listed between the brackets (`x`, `y`, and `z` in this example) |
| `[a-z]` | Any character between `a` and `z`, inclusive |
| `[^xyz]` | The opposite of `[xyz]` |
| `\d` | ASCII digits (0 through 9, inclusive) |
| `\D` | Anything except ASCII digits |
| `\s` | ASCII spaces (space, tab, newline, carriage return, form feed) |
| `\S` | Anything except ASCII spaces |
| `\w` | The same as `[0-9A-Za-z_]` |
| `\W` | Anything except the characters matched by `\w` |
| `[[:alnum:]]` | The same as `[0-9A-Za-z]` |
| `[[:alpha:]]` | The same as `[A-Za-z]` |
| `[[:ascii:]]` | Any ASCII character |
| `[[:blank:]]` | ASCII tab or space |
| `[[:cntrl:]]` | ASCII/Unicode control characters |
| `[[:digit:]]` | The same as `[0-9]` |
| `[[:graph:]]` | All "graphical" (printable) ASCII characters |
| `[[:lower:]]` | The same as `[a-z]` |
| `[[:print:]]` | The same as `[[:graph:]]` |
| `[[:punct:]]` | The same as `` [!-/:-@[-`{-~] `` |
| `[[:space:]]` | The same as `[\t\n\v\f\r ]` |
| `[[:upper:]]` | The same as `[A-Z]` |
| `[[:word:]]` | The same as `\w` |
| `[[:xdigit:]]` | The same as `[0-9A-Fa-f]` |
| `\pN` | Unicode character class by using single-letter class names ("N" in this example) |
| `\p{Greek}` | Unicode character class by unicode name ("Greek" in this example) |
| `\PN` | The opposite of `\pN` |
| `\P{Greek}` | The opposite of `\p{Greek}` |
| `xy` | `x` followed immediately by `y` |
| `x|y` | either `x` or `y`, preferring `x` |
| `x*` | zero or more `x`, preferring more |
| `x*?` | zero or more `x`, preferring fewer |
| `x+` | one or more `x`, preferring more |
| `x+?` | one or more `x`, preferring fewer |
| `x?` | zero or one `x`, preferring one |
| `x??` | zero or one `x`, preferring zero |
| `x{n,m}` | between `n` and `m` repetitions of `x`, preferring more |
| `x{n,m}?` | between `n` and `m` repetitions of `x`, preferring fewer |
| `x{n,}` | at least `n` repetitions of `x`, preferring more |
| `x{n,}?` | at least `n` repetitions of `x`, preferring fewer |
| `x{n}` | exactly `n` repetitions of `x` |
| `(x)` | unnamed capture group for sub-pattern `x` |
| `(?P<name>x)` | named capture group, named `name`, for sub-pattern `x` |
| `(?:x)` | non-capturing sub-pattern `x` |
| `\*` | Literal `*` for any punctuation character `*` |
| `\Q...\E` | Literal `...` for any text `...` as long as it does not include literally `\E` |
In addition to the above matching operators that consume the characters they
match, there are some additional operators that _only_ match, but consume
no characters. These are "zero-width" matching operators:
| Sequence | Matches |
| -------- | ------------------------------------------------------------------------------------------------ |
| `^` | At the beginning of the given string |
| `$` | At the end of the given string |
| `\A` | At the beginning of the given string |
| `\z` | At the end of the given string |
| `\b` | At an ASCII word boundary (transition between `\w` and either `\W`, `\A` or `\z`, or vice-versa) |
| `\B` | Not at an ASCII word boundary |
Terraform uses the
[RE2](https://github.com/google/re2/wiki/Syntax) regular expression language.
This engine does not support all of the features found in some other regular
expression engines; in particular, it does not support backreferences.
## Matching Flags
Some of the matching behaviors described above can be modified by setting
matching flags, activated using either the `(?flags)` operator (to activate
within the current sub-pattern) or the `(?flags:x)` operator (to match `x` with
the modified flags). Each flag is a single letter, and multiple flags can be
set at once by listing multiple letters in the `flags` position.
The available flags are listed in the table below:
| Flag | Meaning |
| ---- | ----------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `i` | Case insensitive: a literal letter in the pattern matches both lowercase and uppercase versions of that letter |
| `m` | The `^` and `$` operators also match the beginning and end of lines within the string, marked by newline characters; behavior of `\A` and `\z` is unchanged |
| `s` | The `.` operator also matches newline |
| `U` | The meaning of presence or absense `?` after a repetition operator is inverted. For example, `x*` is interpreted like `x*?` and vice-versa. |
## Examples
```
> regex("[a-z]+", "53453453.345345aaabbbccc23454")
aaabbbccc
> regex("(\\d\\d\\d\\d)-(\\d\\d)-(\\d\\d)", "2019-02-01")
[
"2019",
"02",
"01",
]
> regex("^(?:(?P<scheme>[^:/?#]+):)?(?://(?P<authority>[^/?#]*))?", "https://terraform.io/docs/")
{
"authority" = "terraform.io"
"scheme" = "https"
}
> regex("[a-z]+", "53453453.34534523454")
Error: Error in function call
Call to function "regex" failed: pattern did not match any part of the given
string.
```
## Related Functions
- [`regexall`](./regexall.html) searches for potentially multiple matches of a given pattern in a string.
- [`replace`](./replace.html) replaces a substring of a string with another string, optionally matching using the same regular expression syntax as `regex`.
If Terraform already has a more specialized function to parse the syntax you
are trying to match, prefer to use that function instead. Regular expressions
can be hard to read and can obscure your intent, making a configuration harder
to read and understand.