feat: separate methods for object key value

[xDivisionByZerox]

Created in relation to #492.

Thinks I have done:

• Create and implement 2 new methods faker.random.objectValue and faker.random.objectKey
• Wrote documentation for both methods
• Wrote test for both methods
• Added a deprecation flag and log for faker.random.objectElement

[codecov]

Codecov Report

Merging #503 (f9b8225) into main (6cfd1e1) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##             main     #503   +/-   ##
=======================================
  Coverage   99.41%   99.41%           
=======================================
  Files        1959     1959           
  Lines      210830   210905   +75     
  Branches      901      906    +5     
=======================================
+ Hits       209600   209675   +75     
  Misses       1172     1172           
  Partials       58       58           

Impacted Files	Coverage Δ
src/finance.ts	99.29% <100.00%> (-0.01%)	⬇️
src/helpers.ts	99.32% <100.00%> (+0.02%)	⬆️
src/random.ts	99.54% <100.00%> (+0.03%)	⬆️

[xDivisionByZerox]

The failing CI doesn't seem to be related to the changes I made, rather than with the fake module.

2022-03-26T09:57:08.4524974Z  > Fake.fake src/fake.ts:85:12
2022-03-26T09:57:08.4525598Z
2022-03-26T09:57:08.4526162Z     if (this.faker[parts[0]] == null) {
2022-03-26T09:57:08.4527269Z           throw new Error('Invalid module: ' + parts[0]);
2022-03-26T09:57:08.4527942Z                             ^
2022-03-26T09:57:08.4528536Z     }

Error: invalid module: {datatype

I guess a rebase could fix this. (?)

[Shinigami92]

1. We need to add test for the deprecation warning
2. We need to swap out all internal usages like in finance.currencyCode (please use search to find more occurrences)

[Shinigami92]

The failing CI doesn't seem to be related to the changes I made, rather than with the fake module.

2022-03-26T09:57:08.4524974Z  > Fake.fake src/fake.ts:85:12
2022-03-26T09:57:08.4525598Z
2022-03-26T09:57:08.4526162Z     if (this.faker[parts[0]] == null) {
2022-03-26T09:57:08.4527269Z           throw new Error('Invalid module: ' + parts[0]);
2022-03-26T09:57:08.4527942Z                             ^
2022-03-26T09:57:08.4528536Z     }

Error: invalid module: {datatype

I guess a rebase could fix this. (?)

I'm not totally sure what's going on here, I hope @ST-DDT has some more in-depth inside into the fake function and what could be the problem here related to this PR changes
main branch is green and I did not see this previously somewhere

Here is a local failed test result

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ Failed Tests 1 ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯

 FAIL  test/all_functional.spec.ts > faker.fake functional tests > uk > datatype > array()
Error: Invalid module: {datatype
 ❯ Fake.fake src/fake.ts:85:12
     83| 
     84|     if (this.faker[parts[0]] == null) {
     85|       throw new Error('Invalid module: ' + parts[0]);
       |            ^
     86|     }
     87| 
 ❯ Fake.fake src/fake.ts:124:16
 ❯ test/all_functional.spec.ts:96:35

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯[1/1]⎯

[ST-DDT]

I have identified the issue.
The generated string contains a $ which causes the string replace method to insert the token again. In this case the $ was directly after a { which then caused the issue.
I will provide a fix for the faker method.

[ST-DDT]

Fix and test provided: https://github.com/faker-js/faker/pull/688/files#diff-d764addde9b944dba41e8666051bf020b5f44349c6043d6d7770bd58073eca25R83

[Shinigami92]

I will take this PR over so we can merge it soon

[Shinigami92]

I just found out that we need JSDoc with @deprecated on ALL signatures!

In the one line with not using 'key', it is not strikedthrough...

[ST-DDT]

Please also change the helpers module js to something like this:

Module with various helper methods that transform the method input rather than returning values from locales. The transformation process may call methods that use the locale data.