> ## Documentation Index > Fetch the complete documentation index at: https://docs.fingerprint.com/llms.txt > Use this file to discover all available pages before exploring further. # Expression language reference Rule expressions are written using a simple expression language. It provides a rich set of operators and builtin functions, and allows you to create flexible conditions and criteria. You can use available data fields with available operators and functions. ## **Data fields** | **Key** | **Type** | **Notes** | | :------------------------------- | :------------- | :----------------------------------------------------------------------------------------- | | `http.user_agent.is_ai_bot` | `bool` | Whether the UserAgent string matches the criteria for AI Bots. | | `http.user_agent.is_crawler_bot` | `bool` | Whether the UserAgent string matches the criteria for Crawler Bots. | | `http.user_agent.string` | `string` | The full UserAgent string. | | `http.request.headers` | `http_headers` | The map of HTTP request headers. | | `http.request.ip` | `ip` | IP address of the client making the request. | | `sdk.platform.name` | `string` | Name of the SDK platform (i.e. one of `js`, `android`, `ios`, `unknown`). | | `sdk.version` | `string` | Version of the SDK platform. Example value: `1.2.3`. | | `sdk.platform.app_package_name` | `string` | Name of the mobile app bundle, if the request came from a mobile SDK (`android` or `ios`). | | `sdk.platform.is_android` | `bool` | Flag indicating if request came from Android device. | | `sdk.platform.is_browser` | `bool` | Flag indicating if request came from a web browser. | | `sdk.platform.is_ios` | `bool` | Flag indicating if request came from iOS device. | | `sdk.platform.is_unknown` | `bool` | Flag indicating if request came from an unknown platform. | ## **Operators** `LHS` stands for left-hand side of the operation, `RHS` stands for right-hand side of the operation. ### Equality * `LHS == RHS` - returns `true` if both sides are equal * `LHS != RHS` - returns `true` if both sides are different ### Numerical When `LHS` and `RHS`are numerical values. * `LHS < RHS` - returns `true` if `LHS` is less than `RHS` * `LHS > RHS` - returns `true` if `LHS` is more than `RHS` * `LHS <= RHS` - returns `true` if `LHS` is less than or equal to `RHS` * `LHS >= RHS` - returns `true` if `LHS` is more than or equal to `RHS` ### Logical When `LHS` and `RHS`are boolean values. * `!LHS` - returns the opposite boolean value of `LHS` * `LHS && RHS` - returns `true` if both `LHS` and `RHS` are `true` * `LHS || RHS` - returns `true` if either `LHS` or `RHS` are `true` ### Strings When `LHS` and `RHS`are strings. * `LHS contains RHS` - returns `true` if `RHS` is a contained in `LHS` * `LHS not contains RHS` - returns `true` if `RHS` is not contained in `LHS` ### RegEx matching When `LHS` is a string, and `RHS` is a string literal with a [RegEx](https://github.com/google/re2/wiki/Syntax) pattern. Pattern needs to be enclosed with backticks (\`). * `LHS matches RHS` - returns `true` if `LHS` matches the regular expression in the `RHS` * `LHS not matches RHS` - returns `true` if `LHS` does not match the regular expression in the `RHS` ### Arrays When `LHS`is an array. * `LHS in RHS` - returns `true` whether `LHS` is in (or belongs to) `RHS` * `LHS not in RHS` - returns the opposite of `LHS in RHS` * `LHS[RHS]` - returns the value at *numeric* index `RHS` in the array `LHS`, returns `nil` for missing values ### Objects When `LHS`is an object. * `LHS in RHS` - returns `true` whether `LHS` is in (or belongs to) * `RHS`:`LHS not in RHS` - returns the opposite of `LHS in RHS`. * `LHS[RHS]` - returns the value at string field `RHS` in the object `LHS`. Returns `nil` for missing values * `LHS.RHS` is equivalent to `LHS[RHS]` where `LHS` is an `object`, and `RHS` is an unquoted `string` literal that can be used as an identifier. *Examples:* 1. `http.request` is equivalent to `http["request"]`. 2. `http.request.headers.origin.0` produces an error because the unquoted `string` literal `0` cannot be used as an identifier. ## **Functions and custom data types** Describes additional functions and special data types available during rule evaluation. ### Strings * `upper(string) -> string` - transforms the string to all uppercase characters * `lower(string) -> string` - - transforms the string to all lowercase characters * `starts_with(source string, substring string) -> bool` * `ends_with(source string, substring string) -> bool` ### HTTP headers The *Custom Type* `http_headers` behaves similar to an `object`, but it’s keys are normalized to be in MIME Header format in all operations (so usage is case-insensitive), and its values are of type `array` and contain the values for the given header name, preserving the original order. \ \ *Examples*: 1. Check if a header was set: `"cOnTeNt-TyPe" in http.request.headers` 2. Get the first cookie that was set: `http.request.headers["accept"][0]` ### IP address and CIDR The *Custom Type* `ip` represents an IPv4 or IPv6 address. The *Custom Type* `cidr` represents an IPv4 or IPv6 CIDR range. * `cidr(string) -> cidr` parses an IPv4 or IPv6 CIDR. * The `in` operator can be used to check if an `ip` is in a given `cidr`. *Example*: 1. Check if IP address belongs to a CIDR range: `http.request.ip in cidr("1.1.1.1/10")` ### Semantic versioning * `semver_is_valid(string) -> bool` - checks if given string is a valid semver string * `semver_compare(s1, s2) -> int` - compares two semver strings * returns `-1` if `s2 < s1`, `1` if `s2 > s1`, or `0` if they are equal