refactor: use template strings, unify testing, remove dottie

[SimonSchick]

Pull Request check-list

Please make sure to review and check all of these items:

• Does npm run test or npm run test-DIALECT pass with this change (including linting)?
• Does the description below contain a link to an existing issue (Closes #[issue]) or a description of the issue you are solving?
• Have you added new tests to prevent regressions?
• Is a documentation update included (if this change modifies existing APIs, or introduces new ones)?
• Did you follow the commit message conventions explained in CONTRIBUTING.md?

Description of change

Last big PR for a while.

1. Used eslint to convert all string concat to template strings, see https://stackoverflow.com/questions/29055518/are-es6-template-literals-faster-than-string-concatenation
2. Moved all usage of suite and test within tests to use the describe and it functions.
3. Replaced dottie with _.set/get (one less dependency 🎉)

[sushantdhiman]

Please keep dottie dependency for now or atleast we will remove that in different PR. Dottie is used in some performance sensitive codebase. I want to measure the impace before we replace it with lodash get/set

[sushantdhiman]

Some tests for Postgres are failing

[codecov]

Codecov Report

Merging #10055 into master will decrease coverage by <.01%.
The diff coverage is 89.91%.

@@            Coverage Diff             @@
##           master   #10055      +/-   ##
==========================================
- Coverage    96.3%    96.3%   -0.01%     
==========================================
  Files          63       63              
  Lines        9412     9410       -2     
==========================================
- Hits         9064     9062       -2     
  Misses        348      348

Impacted Files	Coverage Δ
lib/associations/belongs-to-many.js	97.03% <ø> (ø)	⬆️
lib/associations/belongs-to.js	100% <ø> (ø)	⬆️
lib/associations/has-one.js	97.11% <ø> (ø)	⬆️
lib/associations/has-many.js	97.67% <ø> (ø)	⬆️
lib/dialects/abstract/connection-manager.js	87.6% <100%> (ø)	⬆️
lib/dialects/sqlite/query-interface.js	100% <100%> (ø)	⬆️
lib/dialects/mssql/data-types.js	98.13% <100%> (ø)	⬆️
lib/dialects/sqlite/data-types.js	97.15% <100%> (ø)	⬆️
lib/dialects/mssql/query.js	94.87% <100%> (ø)	⬆️
lib/dialects/sqlite/connection-manager.js	97.77% <100%> (ø)	⬆️
... and 24 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update be958ac...b17941c. Read the comment docs.

[SimonSchick]

Re dottie, I did a rudimentary benchmark with 10mil iterations.

get
lodash: 2487.456ms
dottie: 1422.263ms
set
lodash: 2952.551ms
dottie: 2644.326ms

Reason I removed it is that you use _.get _.set everywhere else, not sure if the performance impact is relevant at all.

[sushantdhiman]

We use lodash.get to merely get some deep config path, messages etc stuff, nothing critical. lodash.set is not used anywhere.

Dottie still got upper hand (as per your benchmark), I would like to measure impact at-least independently before deciding to remove it

[SimonSchick]

I don't think it will have any measureable impact in actual consumer use but I will revert the specific changes.

Here is my benchmark:

'use strict';

const { get: lget, set: lset } = require('lodash');
const { get: dget, set: dset } = require('dottie');

console.log('get');
{
  const lookup = 'a.b.c.d.e.f.g';
  const obj = {
    a: {
      b: {
        d: {
          e: {
            f: {
              g: 123,
            },
          },
        },
      },
    },
  };
  console.time('lodash');
  for (let i = 0;i<10000000;++i) {
    lget(obj, lookup);
  }
  console.timeEnd('lodash');
}

{
  const lookup = 'a.b.c.d.e.f.g';
  const obj = {
    a: {
      b: {
        d: {
          e: {
            f: {
              g: 123,
            },
          },
        },
      },
    },
  };
  console.time('dottie');
  for (let i = 0;i<10000000;++i) {
    dget(obj, lookup);
  }
  console.timeEnd('dottie');
}

console.log('set');
{
  const lookup = 'a.b.c.d.e.f.g';
  const obj = {};
  console.time('lodash');
  for (let i = 0;i<10000000;++i) {
    lset(obj, lookup, 123);
  }
  console.timeEnd('lodash');
}

{
  const lookup = 'a.b.c.d.e.f.g';
  const obj = {};
  console.time('dottie');
  for (let i = 0;i<10000000;++i) {
    dset(obj, lookup);
  }
  console.timeEnd('dottie');
}

[sushantdhiman]

Thanks @SimonSchick