Update HTML Reporter to use QUnit.on() instead of callback events #1486
Comments
Krinkle
added
Component: HTML Reporter
Type: Enhancement
|
There is no CRI event for individual assertions. Given we collapse tests by default in the interface, perhaps it's okay to lose this immediacy in the HTML DOM. Batching the DOM/layout to only happen at the end of each test might actually speed up execution as well. The reporter currently uses The I wrote more at #1118 (comment) about untangling I believe Thinking out loud:
|
Switch from `QUnit.begin()` to `QUnit.done()` to the `runStart` and `runEnd` events, as well as the data these provide, such as the test counts and runtime. As part of making the interface less cluttered, I'm proposing we remove the runtime and status-specific counts from the realtime progress text, as these were rapidly changing and are hard to read. These changes, together, would allow removal of a number of things: * Remove internal dependency on Test class (in favour of runStart data). * Remove local counting of test by status (in favour of runEnd data). * Remove local time measuring. I've moved the progress line from being the third line in the testresult-display element, to be the first line. This way, the number of completed tests will be in the same place as where it is displayed when the run has ended. I've also removed the line break between "Running" and the name of the currently running tests. This way the testresult display will generally have the same visual height during and after the test run. (Previously, there was jump ump after the test run, since there are only two lines in the final state, but with the linebreak we had three lines when tests are running.) Ref qunitjs#1486.
Switch from `QUnit.begin()` to `QUnit.done()` to the `runStart` and `runEnd` events, as well as the data these provide, such as the test counts and runtime. As part of making the interface less cluttered, I'm proposing we remove the runtime and status-specific counts from the realtime progress text, as these were rapidly changing and are hard to read. These changes, together, would allow removal of a number of things: * Remove internal dependency on Test class (in favour of runStart data). * Remove local counting of test by status (in favour of runEnd data). * Remove local time measuring. I've moved the progress line from being the third line in the testresult-display element, to be the first line. This way, the number of completed tests will be in the same place as where it is displayed when the run has ended. I've also removed the line break between "Running" and the name of the currently running tests. This way the testresult display will generally have the same visual height during and after the test run. (Previously, there was jump ump after the test run, since there are only two lines in the final state, but with the linebreak we had three lines when tests are running.) Ref #1486.
Internally, HTML Reporter has no strong need for using the testId = data.testId,
current = document.getElementById( "qunit-test-output-" + testId ), renderColors(colors, '#qunit-test-output-' + QUnit.config.current.testId, label);If we want HTML Reporter to use We should then also, in a future 2.x minor release introduce something that faccilitates this kind of plugin in another way. I'm struggling to come up with something that's compatible with having the HTML Reporter not rely on consuming internal state though. For example, if we recommended that plugins use something like As far as I'm concerned, there's no urgency for making the HTML Reporter a pure reporter instead of a QUnit plugin. For one, the HTML Runner portion of it and the toolbar etc would always be a plugin that uses the tighter integration as today. The only urgency in some sense is that we'd like to avoid major versions where we can, and we have one coming up, so if any breaking changes are needed, now is the time. To be continued :) |
Krinkle
changed the title
=== What === Turn html.js into an HtmlReporter class where the HtmlReporter.init static method contains (most) of the browser runner logic that we would run even if the HtmlReporter were turned off. This patch primarily deals with making html.js itself free of side-effects by delaying any browser setup logic to the init() function. In a future patch we can take this further and move the logic to a "browser runner" function of sorts, such that what is left in the reporter can be made conditional on the presence of `div[id=qunit]` and/or `QUnit.config.reporters.html`. This is tracked in qunitjs#1118 and qunitjs#1711. Ref qunitjs#1486. === Why === We need html.js to be side-effect free in order to add native ESM support, tracked at qunitjs#1551. To support native ESM exports, we need to introduce a separate distribution file since ESM export syntax isn't allowed in scripts. I expect numerous projects to involve mixed ESM imports and CJS require usage, especially when integration layers and plugins are involved. To avoid a split-brain problem, this means we need to somehow have both export the same one. The way I intend to do this is by detecting if QUnit was already defined, and if so, discard our local definition and use that one instead. This means we need to only init the browser runner only if we're the first one (e.g. window.onerror, QUnit.begin for fixture, QUnit.testDone for HTML reporting, etc). To accomplish that, start by making html.js side-effect free, deferring its init code to a (for now) unconditional call at the end of qunit.js.
=== What === Turn html.js into an HtmlReporter class where the HtmlReporter.init static method contains (most) of the browser runner logic that we would run even if the HtmlReporter were turned off. This patch primarily deals with making html.js itself free of side-effects by delaying any browser setup logic to the init() function. In a future patch we can take this further and move the logic to a "browser runner" function of sorts, such that what is left in the reporter can be made conditional on the presence of `div[id=qunit]` and/or `QUnit.config.reporters.html`. This is tracked in qunitjs#1118 and qunitjs#1711. Ref qunitjs#1486. === Why === We need html.js to be side-effect free in order to add native ESM support, tracked at qunitjs#1551. To support native ESM exports, we need to introduce a separate distribution file since ESM export syntax isn't allowed in scripts. I expect numerous projects to involve mixed ESM imports and CJS require usage, especially when integration layers and plugins are involved. To avoid a split-brain problem, this means we need to somehow have both export the same one. The way I intend to do this is by detecting if QUnit was already defined, and if so, discard our local definition and use that one instead. This means we need to only init the browser runner only if we're the first one (e.g. window.onerror, QUnit.begin for fixture, QUnit.testDone for HTML reporting, etc). To accomplish that, start by making html.js side-effect free, deferring its init code to a (for now) unconditional call at the end of qunit.js.
=== What === Turn html.js into an HtmlReporter class where the HtmlReporter.init static method contains (most) of the browser runner logic that we would run even if the HtmlReporter were turned off. This patch primarily deals with making html.js itself free of side-effects by delaying any browser setup logic to the init() function. In a future patch we can take this further and move the logic to a "browser runner" function of sorts, such that what is left in the reporter can be made conditional on the presence of `div[id=qunit]` and/or `QUnit.config.reporters.html`. This is tracked in qunitjs#1118 and qunitjs#1711. Ref qunitjs#1486. === Why === We need html.js to be side-effect free in order to add native ESM support, tracked at qunitjs#1551. To support native ESM exports, we need to introduce a separate distribution file since ESM export syntax isn't allowed in scripts. I expect numerous projects to involve mixed ESM imports and CJS require usage, especially when integration layers and plugins are involved. To avoid a split-brain problem, this means we need to somehow have both export the same one. The way I intend to do this is by detecting if QUnit was already defined, and if so, discard our local definition and use that one instead. This means we need to only init the browser runner only if we're the first one (e.g. window.onerror, QUnit.begin for fixture, QUnit.testDone for HTML reporting, etc). To accomplish that, start by making html.js side-effect free, deferring its init code to a (for now) unconditional call at the end of qunit.js.
|
I'm going to consider this done. The HTML Reporter has to use plugin callback and cannot use purely the js-reporters
I think this is fine since Instead I'll prioritise: decoupling from internals, decoupling from browser running logic, and making it truly optional and able to disable entirely. These are tracked as: |
=== What === Turn html.js into an HtmlReporter class where the HtmlReporter.init static method contains (most) of the browser runner logic that we would run even if the HtmlReporter were turned off. This patch primarily deals with making html.js itself free of side-effects by delaying any browser setup logic to the init() function. In a future patch we can take this further and move the logic to a "browser runner" function of sorts, such that what is left in the reporter can be made conditional on the presence of `div[id=qunit]` and/or `QUnit.config.reporters.html`. This is tracked in #1118 and #1711. Ref #1486. Intentional changes: * Use of previousFailure is no longer conditional on QUnit.config.reorder in order to make it more stateless. The message already did not state that it relates to config.reorder, and stating that it previously failed is accurate either way. * Calling ev.preventDefault() is no longer conditional. This has been supported since IE9. QUnit 2 already required IE9+, and QUnit 3.0 will require IE11. jQuery removed similar check in jQuery 2.2.0, with commit jquery/jquery@a873558436. === Why === We need html.js to be side-effect free in order to add native ESM support, tracked at #1551. To support native ESM exports, we need to introduce a separate distribution file since ESM export syntax isn't allowed in scripts. I expect numerous projects to involve mixed ESM imports and CJS require usage, especially when integration layers and plugins are involved. To avoid a split-brain problem, this means we need to somehow have both export the same one. The way I intend to do this is by detecting if QUnit was already defined, and if so, discard our local definition and use that one instead. This means we need to only init the browser runner only if we're the first one (e.g. window.onerror, QUnit.begin for fixture, QUnit.testDone for HTML reporting, etc). To accomplish that, start by making html.js side-effect free, deferring its init code to a (for now) unconditional call at the end of qunit.js.
Would help identify any gaps or inconveniences which we can then feed back into the CRI standard at https://github.com/js-reporters/js-reporters.
The text was updated successfully, but these errors were encountered: