Kingsland.ArmLinter

Overview

Kingsland.ArmLinter is a C# library for parsing and validating the contents of Azure ARM Template files.

The motiviation for this library was too many late nights committing an ARM template with an embedded expression along the lines of:

"name": "[concat('storage', uniqueString(resourceGroup().id))]"

and pushing it through my build and deployment pipeline only to find out 5 minutes later that I'd missed a ')' somewhere in the expression, or misspelled a built-in function name.

Lexing

The project can be used to lex and parse all of the expressions in a deployment template (see the Kingsland.ArmValidator project for an example).

You can currently detect the following types of errors:

• [concat('storage)] - mismatched quotes in strings
• [concat('storage'] - mismatched brackets:
• [concat()'storage'] - unexpected token sequences

At present, it won't validate the argument signature of functions - e.g.

• toLower('value1', 'value2') - too many arguments
• toLower(100) - incorrect argument type

Evaluating

The library supports evaluating a limited subset of ARM Template functions inside expression strings - for example:

ArmExpressionEvaluator.Evaluate("concat('hello', 'brave', 'new', 'world')");
// returns "hellobravenewworld"

ArmExpressionEvaluator.Evaluate("concat('hello', concat('brave', 'new'), 'world')");
// also returns "hellobravenewworld"

ArmExpressionEvaluator.Evaluate("base64('one, two, three')");
// returns "b25lLCB0d28sIHRocmVl"

ArmExpressionEvaluator.Evaluate("toUpper('One Two Three')");
// returns "ONE TWO THREE"

ArmExpressionEvaluator.Evaluate("concat(toLower('ONE'), '-', toUpper('two'), '-', base64('three'))");
// returns "one-TWO-dGhyZWU="

The following functions are currently implemented:

• Array functions: concat, createArray

• String functions: base64, base64ToString, concat, dataUri, dataUriToString, empty, endsWith, first, format, indexOf, last, lastIndexOf, length, padLeft, replace, skip, startsWith, substring, take, toLower, toUpper, trim

but the code is structured to allow easy addition of new functions, so post an issue (or send a PR) if there's one you'd like to be supported.

Roadmap

This project still currently a bit of a work in progress - here's the things it still needs...

Target	Progress
Write the lexer	Done!
Write the parser	Done!
Write an evaluator	In Progress
Write an analyzer	Not Started