Typescriptify - Conversion to TypeScript. #133

BurtHarris · 2017-07-30T21:07:56Z

Readme still needs to be updated.

BREAKING CHANGE: Deprecated decorators removed: @debounce, @throttle, @mixin, and @memoize, and associated tests
All source files now in TypeScript, tests remain in ECMAScript
Includes a fast, gulp-based build using gulp-tsb
Includes eslint checking on all .js files, no tslint yet, but soon

The tests remain in ECMAScript, so that they can be run for both Babel and TypeScript generated versions. Some of the test cases aren't working in TypeScript yet, and may require a typescript upgrade to support them. But all the remaining tests continue to work when called from Babel processed sources.

Tests now require the core-decorators package by name, so they are like any dependent code. This is supported at test-time by creating a symlink in ./test/node_modules/core-decorators pointing to the project root directory.

This file is supposed to allow checking that exactly the intended versions of dependencies are being used.

@debounce

Removes the @debounce, @Throttle, @mixin, and @memoize decorators which have been deprecated for some time. The corresponding test cases have been removed as well, bringing the current number of tests expected to 56. BREAKING CHANGE: but we may not yet increment the major version! SemVer is currently not applicable to this package due to major version of 0.

Some edits needed, but nothing major. Left the hard ones as .js files.

This includes changin the output directories to cjs and esm.

They build into directory testBabel. With this change, we no longer need to use mocha switchs `--compilers js:babel-core/register` or `--require babel-polyfill`

Switched default output directory back to `lib` Switch clean to use del rather than rimraf Update VSCode launch & tasks config Fixup top-level tests for TypeScript

setup VSCode launch to test typescript

This reverts commit 4ae646f.

Typescript buidls with gulp-tsb for speed Integration with babel, eslint and mocha

There are a few warnings left.

Some of the TypeScript runs are still failing...

BurtHarris · 2017-07-30T21:15:26Z

.gitignore

-/built/
+/esm/
+/test/babel/
+/test/typescript/


Tests, from both transpilers, now build into these test subdirectories so that they can be compared.

BurtHarris · 2017-07-30T21:17:53Z

index.d.ts

@@ -1,69 +0,0 @@
-// Type definitions for core-decorators.js 0.19


index.d.ts is no longer need this because TypeScript is generating individual .d.ts files per source file.

BurtHarris · 2017-07-30T21:18:23Z

index.ts

@@ -1,4 +0,0 @@
-// index.ts - facade for core-decorators


Replaced index.ts/index.js with a change to main setting in package.json

BurtHarris · 2017-07-30T21:31:44Z

.editorconfig

 indent_size = 2
 indent_style = space
+charset = utf-8
+end_of_line = lf
+insert_final_newline = true


Ahh, I had the editorconfig add-in for vscode installed but disabled... That explains some things.

jayphelps

It's a pretty big PR so it'll take me a bit to really dive deep into it and confirm, but I did an initial pass and semantically everything appeared great. I only found code style consistency issues.

jayphelps · 2017-08-02T19:23:20Z

src/core-decorators.ts

 export { default as lazyInitialize } from './lazy-initialize';
 export { default as time } from './time';
 export { default as extendDescriptor } from './extendDescriptor';
 export { default as profile } from './profile';

+// Exported for testing purposes
+export  {default as defaultConsole} from './defaultConsole';


please indent like the others.

export { default as defaultConsole } from './defaultConsole';

P.S. I'd like to add tslint too, currently these escaped eslint because of the file extension.

jayphelps · 2017-08-02T19:24:00Z

src/deprecate.ts

  return decorate(handleDescriptor, args);
 }
+
+export {deprecate, deprecate as deprecated} ;


jayphelps · 2017-08-02T19:24:35Z

src/lazy-initialize.ts

  return decorate(handleDescriptor, args);
 }
+
+export {lazyInitialize};


jayphelps · 2017-08-02T19:25:33Z

src/private/utils.ts

-  throttleTrailingArgs = null;
-
-  @lazyInitialize
+  // @lazyInitialize


I think this is extraneous?

I remember commenting out the @lazyInitialize part was intentional. Its broken due to TypeScript's limitations on property initializers, even tests from Babel fail with the @lazyInitialize in place here.

As I remember removing the throttleTrailingArgs = null; part was part of removing the obsolete decorators and their tests.

sorry if I'm misunderstanding but it's still not clear if this comment extraneous? It seems like it is

Oh! You mean I could simply delete the comment. Of course you are right.

jayphelps · 2017-08-02T19:25:49Z

src/profile.ts

@@ -1,15 +1,9 @@
-import { decorate, metaFor, warn, bind } from './private/utils';
+import { decorate, metaFor } from './private/utils';
+import {defaultConsole} from './defaultConsole';


jayphelps · 2017-08-02T19:27:43Z

test/exports.spec.js

+const glob = require('glob');
+const camelCase = require('camelCase');
+const interopRequire = require('interop-require');
+const decorators = require('core-decorators');


Why did we lose the ability to use ESM imports? sorry if I missed a previous explanation.

What I was shooting for was that the exports.spec.js and test.spec.js files should not require any transpiler. I think as long as we're running tests on a modern version of node.js, we could use ESM imports here, but I wasn't sure. I can restore them if you want. Please advise.

@BurtHarris I wasn't going to burden you right now with it, but I was hoping to also switch the tests over to TypeScript so that we're actually also testing the type definitions at the same time. This can come later (if ever). For now your plan is OK with me, thanks for clarifying.

@jayphelps I had shared your hope, just want to make sure you understand that as of TypeScript@2.4, the limited support for property decorators means we can't really test all of core-decorators from a TypeScript file. This is the quote from the TypeScript handbook that I think reflects the limitation:

NOTE A Property Descriptor is not provided as an argument to a property decorator due to how property decorators are initialized in TypeScript. This is because there is currently no mechanism to describe an instance property when defining members of a prototype, and no way to observe or modify the initializer for a property. As such, a property decorator can only be used to observe that a property of a specific name has been declared for a class.

I've tried to find issues tracking this restriction, microsoft/TypeScript#11866 seems one, but it's been closed, making reference to possibly reopening once the Public class fields TC39 proposal is ratified. I don't fully understand that proposal, but it seems different from the stage 3 decorators one. Can you see any reason they couldn't/shouldn't address it the way Babel does, by passing a synthetic (transpiler generated) descriptor to the first decorator in the list?

Oh yes..I forgot they didn't support it. I say we punt.

Re TS updating to stage-3: I haven't heard, but I would bet money TS is waiting for the decorators proposal to be finalized stage-4 because in its current stage-3 form it's going to be quite the pain in the ass of a breaking change for the TS community aka mostly Angular people, so they want to be 100% sure of the changes so they only have to do it once. If it goes stage-4 fairly soon, there's not a lot of point in them adding the old initializer way that babel mostly just made up because the early spec didn't clarify

But doesn't getting to stage 4 require "Two compatible implementations which pass the acceptance tests"? I was thinking TypeScript & Babel might be the likely implementations, have I misread the TC39 process and practices?

jayphelps · 2017-08-02T19:29:33Z

test/unit/override.spec.js

@@ -36,7 +36,7 @@ describe('@override', function () {
    (function () {
      class Child extends Parent {
        @override
-        speak(first, second) {}
+        speak (first, second) {}


hmm why add the space in-between all of these? I'd prefer they weren't.

Personally, I agree with you. The requirement for a space here is part of the eslint "semistandard" ruleset, and I'm not strongly advocating it.... As I said somewhere else, I picked that because it seemed a close match to existing code, but this whitespace was one change needed to conform to "semistandard".

I think picking a canned standard is easier than maintaining our own ruleset, but I don't have broad knowledge of all the rulesets out there.

BurtHarris · 2017-08-02T21:52:43Z

Yes, sorry for the size of the PR @jayphelps.

I've fixed the indentation issues. They slipped by because eslint doesn't understand typescript files, the long-term fix is to add tslint, but I didn't want to bloat the PR even more. I'll add a workitem to track that.

On the space after function name issue, I agree with you. The requirement for a space there is part of the eslint "semistandard" ruleset. As I said somewhere else, I picked that ruleset because it seemed a close match to existing code, but this whitespace was one change needed to conform to "semistandard".

I think picking a canned standard is easier than maintaining our own ruleset, but I don't have broad knowledge of all the rulesets out there. If you want to choose a different ruleset, or have me tweak that rule I'm happy to oblige.

jayphelps · 2017-08-04T17:32:46Z

Yes, sorry for the size of the PR @jayphelps.

No need to apologize, by its nature it has to be huge--but I appreciate it.

If you want to choose a different ruleset, or have me tweak that rule I'm happy to oblige.

I don't have a preference on standardized ruleset vs custom either way; I've always maintained my own, and IMO it's not a huge deal. I do however personally have a problem with the space in the CallExpressions, that we're talking about. If you'd prefer it, we can of course discuss it further, but if it's all the same to you let's not have the space.

BurtHarris · 2017-08-04T18:44:20Z

On the topic of code style, I think that the "recommended" style that comes with eslint may be a better match for both of our sensibilities. I'll try it out and update the branch.

P.S. I wish there were a "do-not-merge" label I could attach to reflect the fact I still plan on touchups. Is there a way for me to add such a label?

jayphelps · 2017-08-04T19:02:55Z

You should have permissions to create new labels and then add them to any tickets, if not lmk. That won't truly prevent you or I from merging it though. I've changed the settings now to disabling merge ability until at least one review approval. We'll see how it works.

BurtHarris · 2017-08-04T19:16:16Z

@jayphelps I've updated eslint to extend the "reccomended" set and added a rule for the function-parens spacing, fixing the violations. There are 15 lint warnings remaining, these are issues I'd like your input on at some time, particularly the no-unused-expressions ones.

BurtHarris · 2017-08-04T19:45:27Z

@jayphelps My tests match up well with Appveyor, but Travis.CI seems to have other problems. I'm uncomfortable that neither of these were considered a failure and that the PR claims that "All checks have passed." Thoughts?

If you are a Linux guy, perhaps you can look into what's happening on Travis.CI. I'll take a fresh look at the 10 TypeScript failures this weekend.

jayphelps · 2017-08-04T20:36:48Z

Travis is reporting success because gulp test returned exit status 0, indicating success by POSIX convention. I'm not immediately familiar with gulp enough to know how to bubble up a non-zero exit status e.g. 1

The separate issue that there's an error to begin with I commented inline on the problem (case sensitive camelCase -> camelcase on linux_

jayphelps · 2017-08-04T20:39:16Z

test/exports.spec.js

+const chai = require('chai');
+const path = require('path');
+const glob = require('glob');
+const camelCase = require('camelCase');


This is what is causing the error on Linux. This import string needs to be camelcase all lowercase. Linux filesystem is case-sensitive.

Due to a complicated bug, npm is acting strange for me. I've found the Yarn package manager doesn't have the same prolem. Added a yarn.lock file, and a note in README.md. Having both a yarn and npm lock file seems like it might be undesirable long term, but we can reexamine this once some one of the npm/VSCode/typescript bugs are resolved.

Some tests have been marked so that they are skipped in typescript, but these show up as pending rather than failed. These include: - Two tests in autobind.spec.js (needs review) - Two in lazyIntialize.spec.json - One in readonly.spec.json - One in nonenmerable.spec.json

This matters on Linux.

BurtHarris · 2017-08-06T09:53:15Z

@jayphelps,

I've fixed that test error exit status problem by eliminating the gulp-mocha interop layer. just launch mocha to run the full set if tests
I've fixed all the code style issues
I've fix all tests running, with six exceptions. These have been marked as pending for the TypeScript tests, so they don't block successful completion of the test suite. I think four of the remaining problem tests have something to do with the TypeScript property decorator restriction. Two other, in autobind.spec.js I'm less sure about, I think they may be detecting something else. In general I think autobind is working.
I've made an update to the README to reflect the level of TypeScript support and mention the restrictions above.
I think it's close to ready. I can't tell if there are any pending changes I haven't addressed, I think I got them all. Please update your review.

jayphelps · 2017-08-12T01:20:44Z

Unfortunately we may have jumped the gun here. Babel is actively working on adding support for stage-2 decorators and when they do, I think core-decorators needs to follow suit. AFAIK TypeScript has not yet offered any public indication of when (or even if) they're going to do so as well. Thoughts?

jayphelps · 2017-08-13T02:55:04Z

One thing we could do, that I don't like (but this situation sucks no matter what) is indeed convert core-decorators to TypeScript, release it and continue to maintain it as normal then when babel gets stage-2+ spec we go back to babel but maintain the TypeScript version in a branch. We could continue to make bug fixes on it (bumping the semver patch on that same locked minor version).

jayphelps · 2017-08-13T02:59:30Z

Just had another thought: I haven't looked into deeply yet, but it may be possible for us to detect which spec version is being used by the user simply by the arguments passed in. Duck-type checking or instanceof. If that's the case, we could still in fact write this in TypeScript and have it check at runtime which one is being used. The trouble is we need to test the Babel codepath, and I don't think it's possible for TypeScript to compile to something that would work in that case--but we might be able to get away with only using syntax extensions that are also supported by babel's Flow preset.

BurtHarris · 2017-12-20T04:16:25Z

Reviewing TypeScript PRs scheduled for inclusion in the next version I found microsoft/TypeScript#19675, which looks like it may be related the remaining issue blocking this. @jayphelps could you review the comments on that PR?

If I get some cycles I'll try building this with TypeScript@next and verify.

Burt Harris added 14 commits July 14, 2017 15:51

added package-lock.json generated by npm@5

9122175

This file is supposed to allow checking that exactly the intended versions of dependencies are being used.

further work on removing obsolete decorators

55de621

Update build scripts to use TypeScript exclusively

7b96d89

Convert the easy functions to TypeScript

b70f978

Some edits needed, but nothing major. Left the hard ones as .js files.

Build src with gulp

b93c248

This includes changin the output directories to cjs and esm.

Build tests w/ Babel at build time

1752052

They build into directory testBabel. With this change, we no longer need to use mocha switchs `--compilers js:babel-core/register` or `--require babel-polyfill`

Add gulp build of tests for Typescript

2e95f60

Switched default output directory back to `lib` Switch clean to use del rather than rimraf Update VSCode launch & tasks config Fixup top-level tests for TypeScript

Fix gulpfile for typescript test sourcemaps

10b7b78

setup VSCode launch to test typescript

Push test sources into test/src

4ae646f

Revert "Push test sources into test/src"

c2473c8

This reverts commit 4ae646f.

continued gulp/TypeScript conversion

64d5f65

Typescript buidls with gulp-tsb for speed Integration with babel, eslint and mocha

Enable eslint in build

88519c0

There are a few warnings left.

Wrapup making all tests run under both Babel and TypeScript

c339ce2

Some of the TypeScript runs are still failing...

BurtHarris commented Jul 30, 2017

View reviewed changes

Burt Harris added 2 commits July 30, 2017 14:22

insert_final_newline

585ca8a

Whitespace fixes

1804235

BurtHarris commented Jul 30, 2017

View reviewed changes

Burt Harris added 3 commits July 30, 2017 14:38

Removed unneeded change to .editorconfig

a81f8ac

+ comment re: yarn's upcoming 1.0 release

4ea03a0

Clarifying yarn comment in gulpfile

559b393

jayphelps suggested changes Aug 2, 2017

View reviewed changes

jayphelps mentioned this pull request Aug 2, 2017

Typescript build failing tests. (@lazy-initialize not working correctly with tsc) #130

Open

Whitespace/indentation fixes

9324ab1

BurtHarris self-assigned this Aug 4, 2017

BurtHarris added the enhancement label Aug 4, 2017

Removed extranious comment in util.ts

ea71047

eslint tune-up

b83569e

BurtHarris added the do not merge label Aug 4, 2017

jayphelps reviewed Aug 4, 2017

View reviewed changes

BurtHarris mentioned this pull request Aug 4, 2017

autobind not working with async function #98

Open

BurtHarris and others added 4 commits August 5, 2017 14:25

Update describing TypeScript support in README.md

5ac29b6

fix module capitilization

4533023

This matters on Linux.

BurtHarris force-pushed the typescriptify branch from 62fb5b9 to 4533023 Compare August 6, 2017 09:41

BurtHarris removed the do not merge label Aug 6, 2017

		@@ -1,69 +0,0 @@
		// Type definitions for core-decorators.js 0.19

Typescriptify - Conversion to TypeScript. #133

Are you sure you want to change the base?

Typescriptify - Conversion to TypeScript. #133

Conversation

BurtHarris commented Jul 30, 2017 • edited Loading

BurtHarris Jul 30, 2017 • edited Loading

Choose a reason for hiding this comment

BurtHarris Jul 30, 2017 • edited Loading

Choose a reason for hiding this comment

BurtHarris Jul 30, 2017 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jayphelps left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

BurtHarris Aug 2, 2017 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jayphelps Aug 4, 2017 • edited Loading

Choose a reason for hiding this comment

BurtHarris Aug 4, 2017 • edited Loading

Choose a reason for hiding this comment

jayphelps Aug 4, 2017 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

BurtHarris commented Aug 2, 2017 • edited Loading

jayphelps commented Aug 4, 2017

BurtHarris commented Aug 4, 2017

jayphelps commented Aug 4, 2017

BurtHarris commented Aug 4, 2017

BurtHarris commented Aug 4, 2017 • edited Loading

jayphelps commented Aug 4, 2017 • edited Loading

Choose a reason for hiding this comment

BurtHarris commented Aug 6, 2017 • edited Loading

jayphelps commented Aug 12, 2017

jayphelps commented Aug 13, 2017

jayphelps commented Aug 13, 2017

BurtHarris commented Dec 20, 2017 • edited Loading

BurtHarris commented Jul 30, 2017 •

edited

Loading

BurtHarris Jul 30, 2017 •

edited

Loading

BurtHarris Jul 30, 2017 •

edited

Loading

BurtHarris Jul 30, 2017 •

edited

Loading

BurtHarris Aug 2, 2017 •

edited

Loading

jayphelps Aug 4, 2017 •

edited

Loading

BurtHarris Aug 4, 2017 •

edited

Loading

jayphelps Aug 4, 2017 •

edited

Loading

BurtHarris commented Aug 2, 2017 •

edited

Loading

BurtHarris commented Aug 4, 2017 •

edited

Loading

jayphelps commented Aug 4, 2017 •

edited

Loading

BurtHarris commented Aug 6, 2017 •

edited

Loading

BurtHarris commented Dec 20, 2017 •

edited

Loading