cel-bind 0.1.2

No project description provided

Common Expression Language (Python)

The Common Expression Language (CEL) is a non-Turing complete language designed for simplicity, speed, safety, and portability. CEL's C-like syntax looks nearly identical to equivalent expressions in C++, Go, Java, and TypeScript. CEL is ideal for lightweight expression evaluation when a fully sandboxed scripting language is too resource intensive.

// Check whether a resource name starts with a group name.
resource.name.startsWith("/groups/" + auth.claims.group)

// Determine whether the request is in the permitted time window.
request.time - resource.age < duration("24h")

// Check whether all resource names in a list match a given filter.
auth.claims.email_verified && resources.all(r, r.startsWith(auth.claims.email))

Installing

pip install cel-bind

Overview

Determine the variables and functions you want to provide to CEL. Parse and check an expression to make sure it's valid. Then evaluate the output AST against some input.

Environment setup

Let's expose name and group variables to CEL:

import cel

env = {
    "name": cel.StringType(),
    "group": cel.StringType(),
}

Parse and Check

The parsing phase indicates whether the expression is syntactically valid and expands any macros present within the environment. Parsing and checking are more computationally expensive than evaluation, and it is recommended that expressions be parsed and checked ahead of time.

The parse and check phases are combined for convenience into the compile_to_checked_expr step:

pool = cel.DescriptorPool()
compiler = cel.Compiler(pool, env, None)
try:
    checked_expr = compiler.compile_to_checked_expr('name.startsWith("/groups/" + group)')
except Exception as e:
    logging.fatal("compiling error", exc_info=True)
    exit(1)

Evaluate

Build the expression plan with build_expression_plan and reuse it for multiple evaluations of the same expression:

interpreter = cel.Interpreter(pool, env, None)
expr_plan = interpreter.build_expression_plan(checked_expr)

res = interpreter.evaluate(expr_plan, {
    "name": "/groups/acme.co/documents/secret-stuff",
    "group": "acme.co"
})

print(res) # True

Objects

cel-bind supports the use of structured objects in its CEL type-checking and evaluation. First, declare the object type and add it to a cel.DescriptorPool:

schema = """
{
  "type": "Person",
  "properties": {
    "name": {
      "type": "string"
    },
    "age": {
      "type": "integer"
    },
    "hobbies": {
      "type": "array",
      "items": {
        "type": "string"
      }
    }
  },
  "required": ["name", "age"]
}
"""


pool = cel.DescriptorPool()
person_type = pool.add_json_schema("person", schema)

Then, use the returned object type when declaring your CEL environment:

env = {
    "person": person_type,
}

compiler = cel.Compiler(pool, env, None)
checked_expr = compiler.compile_to_checked_expr('person.age > 18')

Currently, only JSON schema is supported when declaring objects.

Extending CEL

Extension functions can be written in Python and then made available to CEL by registering it to a cel.FunctionRegistry:

def is_hiker(person):
    return "hiking" in person["hobbies"]

function_registry = cel.FunctionRegistry()
function_registry.add_function(
    "is_hiker",
    is_hiker,
    cel.BoolType(), # return type
    [person_type], # arguments type
)

Then use it in your CEL expression:

env = {
    "person": person_type,
}

compiler = cel.Compiler(pool, env, function_registry)
try:
    checked_expr = compiler.compile_to_checked_expr('person.age > 18 && is_hiker(person)')
except Exception as e:
    logging.fatal("compiling error", exc_info=True)
    exit(1)

interpreter = cel.Interpreter(pool, env, function_registry)
expr_plan = interpreter.build_expression_plan(checked_expr)

res = interpreter.evaluate(expr_plan, {
    "person": {
        "name": "Alice",
        "age": 20,
        "hobbies": ["cooking", "hiking"]
    }
})
print(res) # True

Type checking

Compiler.compile_to_checked_expr will raise an Exception if type checking fails. Some examples of of type checking errors are:

• undefined object fields

  RuntimeError: ERROR: :1:7: undefined field 'last_name' not found in struct 'person_pkg.person'
  | person.last_name
  | ......^
  

• undeclared reference:

  RuntimeError: ERROR: :1:1: undeclared reference to 'undeclared_var' (in container '')
  | undeclared_var > 5
  | ^
  

• functions applied to incorrect types:

  RuntimeError: ERROR: :1:12: found no matching overload for '_==_' applied to '(int, string)'
  | person.age == "5"
  | ...........^