Update TransformEvaluator.ts to include decomposeDiacriticalMarks

[Semperverus]

Adds a very rudimentary decomposeDiacriticalMarks function to the Transform Evaluators.

The way this functions is by using the built-in javascript normalize() method in 'NFKD' mode to first break a character with diacritical marks up into its constituent unicode components in compatibility mode, and then it performs a regex .replace() to strip out the diacritical marks in the string. These diacritical marks are located in unicode range \u0300 through \u036F as referenced in the official unicode documenation (or alternatively the Wikipedia page which is much easier to read)

This code requires testing within the plugin itself, but the function to perform the decomposition has been tested independently.

Official SailPoint Developer documenation page for decomposeDiacriticalMarks

[Semperverus]

Note: I am not great with JavaScript, and I essentially stole existing functions' design and hopefully put together something acceptable. I mostly needed to wrap the input.normalize('NFKD').replace(/[\u0300-\u036f]/g, "") code into a decomposeDiacriticalMarks method to make it work, as not having this is problematic for work I am doing.

I tested the actual core of the method on Mozilla's documentation page here:
https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/String/normalize and replaced the existing normalize lines (lines 11 and 12) with the above. This validated and works correctly.

If you see any changes that need to be made, please feel free to do so, especially from lines 913 to lines 925 regarding the following code:

            if (attributes.input !== undefined) {
                input = attributes.input;

                if (typeof input === 'object') {
                    input = await this.evaluateChildTransform(input);
                }

                if (input === undefined) {
                    return;
                }
            }

I have a general understanding of what this does, but do not yet fully understand what all might need to go in here. Assuming that evaluating if it's of type object is a safe bet.

I hope this helps!

[Semperverus]

Update: I went ahead and tested the official documentation's usage of the input.replaceAll("[\\p{InCombiningDiacriticalMarks}]", "") regex pattern. This seems to be flavor-specific to Java and is unavailable in Javascript, therefore we will need to use the literal regex match of [\u0300-\u036f] instead.

A more complete list might be [\u0300–\u036F\u1AB0–\u1AFF\u1DC0–\u1DFF\u20D0–\u20FF\uFE20–\uFE2F\u0483-\u0486\u05C7\u0610-\u061A\u0656-\u065F\u0670\u06D6-\u06ED\u0711\u0730-\u073F\u0743-\u074A\u0F18-\u0F19\u0F35\u0F37\u0F72-\u0F73\u0F7A-\u0F81\u0F84\u0e00-\u0eff\uFC5E-\uFC62] (Source), but this also may be overzealous and I think removes some characters that IdentityNow's transform does not.

[yannick-beot-sp]

Thank you for your contribution.
Can you update the README and CHANGELOG files to reflect this update? Don't forget to mention yourself as a contributor (there are example of Pull Requests and contributions in those files).

Also, what would be great is to:

• Refactor your code to use a function that is responsible for cleaning the diacritics (basically, your line result = input.normalize('NFKD').replace(/[\u0300-\u036f]/g, ""); . Such function would find its place in https://github.com/yannick-beot-sp/vscode-sailpoint-identitynow/blob/main/src/utils/stringUtils.ts for instance This may seems silly but this is just to be able to test it.
• Test it (cf. https://github.com/yannick-beot-sp/vscode-sailpoint-identitynow/blob/main/src/test/unit/toCamelCase.test.ts)

[Semperverus]

Okay, these are all uploaded. I am not sure on how I would test locally. Is this something you would be able to run @yannick-beot-sp ?

[yannick-beot-sp]

I'm using the task "Unit Tests" or "Run current mocha test" to execute the test.
So, the tests shall succeed