Expand Support for Eliding Resources at Build Time

[agubler]

Type: feature

The following has been addressed in the PR:

• There is a related issue
• All code has been formatted with prettier as per the readme code style guidelines
• Unit or Functional tests are included in the PR

Description:

Extend static has build time processing to support additional scenarios such as import.

Support for template literals

For applications that are not built in legacy mode, template literals are not down emitted therefore need to be supported when detecting the static has pragma

`has('flag')`

Support for require with a variable declaration

When an require contains a variable declaration, it should be elided.

"!has('include-my-require')"
var my_require = require('my-require');

The variable declaration is replaced with an undefined variable:

var my_require = undefined;

Support for import syntax

Applications not built in legacy mode will not down emit the import syntax but should be supported.

"!has('include-my-require')"
import 'my-require';

For imports that returns a declaration (either default or named), replaces import with an undefined var:

"!has('include-my-require')"
import defaultValue, { namedValue } from 'my-require';

Is transformed to:

var defaultValue = undefined, namedValue = undefined;

Has Flag Logic

Inverted the logic used by the static has processor to include the next import then the has check resolves as to true, this is more natural and reads like you would expect.

`has('include-foo')`
import 'foo';

Previously this would have elided the foo import, however logically the statement appears to indicate that if include-foo is true then it should import foo. Meaning that the has pragma should be negated to elide the import, i.e. only for cases where foo does not exist.

`!has('include-foo')`
import 'foo';

Consolidation of feature sets

We only leverage a single feature (currently chrome) set that describes modern browsers, therefore we can normalise this to a single modern feature set.

Related to dojo/framework#183

[codecov]

Codecov Report

Merging #94 into master will increase coverage by 0.75%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master      #94      +/-   ##
==========================================
+ Coverage   84.72%   85.47%   +0.75%     
==========================================
  Files          33       33              
  Lines        1211     1260      +49     
  Branches      332      353      +21     
==========================================
+ Hits         1026     1077      +51     
+ Misses        185      183       -2

Impacted Files	Coverage Δ
src/static-build-loader/loader.ts	99.35% <100%> (+4.03%)	⬆️
src/static-build-loader/getFeatures.ts	91.3% <0%> (-8.7%)	⬇️

---

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 4f21fb4...45e5cec. Read the comment docs.

[agubler]

Inverts the logic of the has, so now it will only elide the import when the has expression evaluates to false