Nice work @kidonng! Thanks for fixing that upstream.

This seems a bit hacky. Ky really has no idea what the platform is going to do if the request object is malformed. This PR bakes some assumptions into Ky that don't really seem necessary given a standards compliant Request constructor, which Deno aims to provide. In other words, this is likely an upstream bug that should be fixed in the implementation.

While I do not feel strongly about it, I'm slightly against merging this.

Our use case is in the browser, I only mentioned other environments as an example of how it could benefit the larger community. Ky targets browsers primarily. We just don't want to specifically rely on the DOM or window. So we try to avoid referencing document.baseURI or window.location. That makes it difficult for us to use new URL() because it doesn't support relative URLs, which we are sometimes given as input because we are operating in a browser and relative URLs are a common occurrence in browser land.

I'm fine with this. Though, the hooks already receive the request and response, so in many cases it wouldn't make a difference. But it is a nice convenience to have those the error.

We could, but my concern is it will then be confusing if it still applies the prefix to input URLs like ../meow.jpg, which is another case you probably don't want it. More likely, you want some type of URL resolution, and Ky currently does not provide any mechanism for that when using prefixUrl. It's just a simple prefix.

In the past, I've proposed adding a baseUrl option that would be akin to the HTML <base> tag, which would resolve the input URL against a base URL. Sindre thought it would be confusing to have both options, and it turns out a lot of people wanted the prefix behavior so that /meow.jpg and meow.jpg with a prefix would load the same URL, whereas those are very different URLs when resolving against a base. So that's how we ended up with the current system.

You're probably thinking this makes sense but is still a bit kludgy and I agree. One thought I've been kicking around is to let beforeRequest return a URL instance. Internally, t's a little tricky to update the request URL without messing up the request object, unfortunately. But it should be doable.

Open to other suggestions!

Please provide reproducible sample code.

There is already some imports like this.

In Pogo? No. I was aware of export type but not import type.

How do you run the tests ?

The test script is defined in .travis.yml.

That post mentions needing to use import type, which I didn't know was even a thing. Isn't Pogo going to run into problems with that, too?

If launched with --unstable option, for example with --watch, we got a typescript error when exporting types. This PR fix this

Strange that --unstable exposed this. I wonder why. Anyway, LGTM.

Thanks!

The short of it is that for libraries we prefer for CI to test the latest dependencies. After all, if Ky were to use a lockfile, It would only affect our CI, not end-users' apps. Thus the lockfile actually provides very little value for a library and can actually cause CI to be green even though a dependency had broken that would affect anyone installing Ky from scratch, which makes the lockfile actively harmful.

Lockfiles for apps, not libraries.

Understood. I won't have enough time until at least next week. But one of the other maintainers may be able to look at this.

There are a lot of other CDNs to test, though. I really think it would be better to "default" to ESM with main and then fix the Node.js case with exports. I'll play around with this and see how it goes, though.

Sindre and I prefer to use lockfiles in apps but not in reusable libraries. There's a lot of discussion about it on Sindre's AMA if you want to dig into that more. Anyway, thanks for fixing it.

No, that error message comes from the node-fetch dependency. They probably changed it in a minor or patch version since the last time our CI ran. I'd argue that they broke SemVer rules. Ideally, they would roll back that change and release it as SemVer major. But oh well. Since it only affects our tests, I won't bother them about it.

I believe that would break our CDN users who load Ky from https://unpkg.com/ky, for example. They would start receiving UMD instead of ESM, which I personally consider unacceptable for a browser oriented library. If unpkg.com preferred module over main, then it would be fine. But I don't think it does.

CI failed. You can run the tests locally with npm test.

I would be fine with this if it actually worked everywhere, but I believe module is not as widely supported as you might think, including by Node.js itself and its new ESM loader.

We have what I think is a more forward thinking approach in PR #244 but unfortunately that work has stalled due to some configuration problem in our testing stack that I wasn't able to track down quickly. You could help by looking into that.

Need to add value, like the other tests do.

Closes #41

This PR parses request.body and puts the result at request.payload.

The reason why Ky doesn't specify main is because we use the default, which is index.js. This is a convention for the entire Node.js platform. It has nothing to do with Ky. So your issue belongs on either the Node.js issue tracker or the Vite issue tracker, depending on how you look at it. You could try to convince the Node team to make main required or change the default value, in which case we will follow suit. But I suspect you'll have more success in convincing the Vite team to go along with how Node.js works. And it sounds like that is sort of already the case, since the message you're getting is a warning rather than an error. But what's the purpose of the warning? Why is Vite warning about it to begin with? Only the Vite team can answer that. While I'm not too familiar with Vite, as far as I can tell there is no purpose. It's just noise polluting your terminal, probably for no particular reason.

Pagic is a static site generator powered by Deno + React

Here are some demos:

• pagic.org
• Deno X ranking
• TypeScript 入门教程

If it is possible to use pagic to build the website, I would like to create the pull-request

Should be reported here: https://github.com/vitejs/vite

You did not answer my question.

What exactly do you want us to show? Should it include the webpack config, server framework with routes, etc? A real SSR setup can get a bit complex depending on your needs.

Generally speaking, I would suggest using something like Next.js that takes care of the rendering under the hood. Then it's really as simple as just calling ky() wherever you want to fetch the data.

Yes, scroll up in this thread. Other people have suggested various alternatives, such as using a .huskyrc file.

@TrejGun downgrade your Husky to version 3. npm uninstall husky && npm install --save-dev husky@3

Looks nice, I like it! Great name, too.

That said, the Pogo website will be built with Pogo itself or a generator / framework built on top of Pogo. It needs to be "pure" in that sense or people won't take it as seriously. I want to demonstrate that Pogo is a real, useful framework capable of being used in production projects. There are a few hurdles to overcome before I declare that to be fully solved, but building a website would go a long way towards proving it.

If there's something that would help you adopt Pogo, feel free to chime in here on the issue tracker.

For anyone who's trying to get things done, I've published a workaround in the Pogo framework and our React apps are working.

This should work, without needing to import the types yourself...

import React from 'https://jspm.dev/[email protected]';
import pogo from 'https://deno.land/x/pogo/main.ts';

const server = pogo.server({ port : 3000 });

server.router.get('/', () => {
    return <h1>Hello, React!</h1>;
});

server.start();

console.log('Server started: http://localhost:3000');

Indeed, I am able to reproduce this. Smells like a problem with node-fetch to me. For some reason, body ends up being a Node PassThrough stream, instead of a web compliant stream, even when web streams have been polyfilled.

@venikman are you still interested in working on this?

Bumps next from 7.0.3 to 9.3.3. <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/zeit/next.js/releases">next's releases</a>.</em></p> <blockquote> <h2>v9.3.3</h2> <h3>Patches</h3> <ul> <li>Revert "Fix assignment of props in WithApollo.getInitialProps " (<a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11384">#11384</a>): <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11236">#11236</a></li> <li>Add support for comments in tsconfig.json: e7ea27645deacb77fb3f69a830adfadbf70f4b6a</li> <li>Docs: Replace micro-cors with cors middleware: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11395">#11395</a></li> <li>Add support for comments in tsconfig.json: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11392">#11392</a></li> <li>Fixed typo: libraray -> library: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11435">#11435</a></li> <li>Minor change in README.md: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11434">#11434</a></li> <li>Fix with-aphrodite readme typo: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11436">#11436</a></li> <li>Update api-rest-example to SSG: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11362">#11362</a></li> <li>Change mdx options description: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11409">#11409</a></li> <li>Update .gitignore: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11421">#11421</a></li> <li>Update emotion dependencies in with-emotion-11: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11414">#11414</a></li> <li>Fix <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11396">#11396</a>: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11397">#11397</a></li> <li>Fix type ignore: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11479">#11479</a></li> <li>Typos fixed: overriden --> overridden and innaccurate -->inacc…: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11454">#11454</a></li> <li>Fix small punctuation issue: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11439">#11439</a></li> </ul> <h3>Credits</h3> <p>Huge thanks to <a href="https://github.com/bgoerdt">@bgoerdt</a>, <a href="https://github.com/herrstucki">@herrstucki</a>, <a href="https://github.com/MichelleLucero">@MichelleLucero</a>, <a href="https://github.com/liulanz">@liulanz</a>, <a href="https://github.com/sdhani">@sdhani</a>, <a href="https://github.com/JazibJafri">@JazibJafri</a>, <a href="https://github.com/ElForastero">@ElForastero</a>, <a href="https://github.com/moeyashi">@moeyashi</a>, <a href="https://github.com/Andarist">@Andarist</a>, <a href="https://github.com/comxd">@comxd</a>, <a href="https://github.com/chislee0708">@chislee0708</a>, and <a href="https://github.com/PullJosh">@PullJosh</a> for helping!</p> <h2>v9.3.3-canary.0</h2> <h3>Patches</h3> <ul> <li>Revert "Fix assignment of props in WithApollo.getInitialProps " (<a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11384">#11384</a>): <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11236">#11236</a></li> <li>Add support for comments in tsconfig.json: e7ea27645deacb77fb3f69a830adfadbf70f4b6a</li> <li>Docs: Replace micro-cors with cors middleware: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11395">#11395</a></li> <li>Add support for comments in tsconfig.json: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11392">#11392</a></li> <li>Fixed typo: libraray -> library: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11435">#11435</a></li> <li>Minor change in README.md: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11434">#11434</a></li> <li>Fix with-aphrodite readme typo: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11436">#11436</a></li> <li>Update api-rest-example to SSG: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11362">#11362</a></li> <li>Change mdx options description: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11409">#11409</a></li> <li>Update .gitignore: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11421">#11421</a></li> <li>Update emotion dependencies in with-emotion-11: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11414">#11414</a></li> <li>Fix <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11396">#11396</a>: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11397">#11397</a></li> <li>Fix type ignore: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11479">#11479</a></li> <li>Typos fixed: overriden --> overridden and innaccurate -->inacc…: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11454">#11454</a></li> <li>Fix small punctuation issue: <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11439">#11439</a></li> <li>Merge branch &<a href="https://github-redirect.dependabot.com/zeit/next.js/issues/39">#39</a>;canary&<a href="https://github-redirect.dependabot.com/zeit/next.js/issues/39">#39</a>; of github.com:zeit/next.js into canary: dd098ca5f6bf9bd94f7d71950560a6ad649f8b97</li> </ul> <h3>Credits</h3> <p>Huge thanks to <a href="https://github.com/bgoerdt">@bgoerdt</a>, <a href="https://github.com/herrstucki">@herrstucki</a>, <a href="https://github.com/MichelleLucero">@MichelleLucero</a>, <a href="https://github.com/liulanz">@liulanz</a>, <a href="https://github.com/sdhani">@sdhani</a>, <a href="https://github.com/JazibJafri">@JazibJafri</a>, <a href="https://github.com/ElForastero">@ElForastero</a>, <a href="https://github.com/moeyashi">@moeyashi</a>, <a href="https://github.com/Andarist">@Andarist</a>, <a href="https://github.com/comxd">@comxd</a>, <a href="https://github.com/chislee0708">@chislee0708</a>, and <a href="https://github.com/PullJosh">@PullJosh</a> for helping!</p> <h2>v9.3.2</h2> <p><strong>This upgrade is <em>completely backwards compatible and recommended for all users on versions below 9.3.2.</em> For future security related communications of our OSS projects, please <a href="https://zeit.co/security">join this mailing list</a>.</strong></p> </tr></table> ... (truncated) </blockquote> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/zeit/next.js/commit/b5d8660ef347bcdf5f178ffdaec90cd9ceea00c9"><code>b5d8660</code></a> v9.3.3</li> <li><a href="https://github.com/zeit/next.js/commit/ac223e276180611cf11e04b1edf03a1865297b31"><code>ac223e2</code></a> v9.3.3-canary.0</li> <li><a href="https://github.com/zeit/next.js/commit/dd098ca5f6bf9bd94f7d71950560a6ad649f8b97"><code>dd098ca</code></a> Merge branch 'canary' of github.com:zeit/next.js into canary</li> <li><a href="https://github.com/zeit/next.js/commit/e4bf0d4c909dcac601d540e38cfb445ec72ba0eb"><code>e4bf0d4</code></a> Fix small punctuation issue (<a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11439">#11439</a>)</li> <li><a href="https://github.com/zeit/next.js/commit/a55376851457a28e05e0ed6cd90e366661b796c2"><code>a553768</code></a> Typos fixed: overriden --> overridden and innaccurate -->inacc… (<a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11454">#11454</a>)</li> <li><a href="https://github.com/zeit/next.js/commit/d3d0b936a421a5ae1c6daba613cacf2821d44b62"><code>d3d0b93</code></a> Fix type ignore (<a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11479">#11479</a>)</li> <li><a href="https://github.com/zeit/next.js/commit/6f27e21f8aa1587e882067f02f466ce98a16f7fd"><code>6f27e21</code></a> Fix <a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11396">#11396</a> (<a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11397">#11397</a>)</li> <li><a href="https://github.com/zeit/next.js/commit/fc946053fd45fdde18912060faae20aaee9f202a"><code>fc94605</code></a> Update emotion dependencies in with-emotion-11 (<a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11414">#11414</a>)</li> <li><a href="https://github.com/zeit/next.js/commit/41764b28dc296718356e71b034328bd9f23ce4c3"><code>41764b2</code></a> Update .gitignore (<a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11421">#11421</a>)</li> <li><a href="https://github.com/zeit/next.js/commit/8d63330c15e2ff9755406fc9f407bbcc385fc72b"><code>8d63330</code></a> Change mdx options description (<a href="https://github-redirect.dependabot.com/zeit/next.js/issues/11409">#11409</a>)</li> <li>Additional commits viewable in <a href="https://github.com/zeit/next.js/compare/7.0.3...v9.3.3">compare view</a></li> </ul> </details> <br />

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

<details> <summary>Dependabot commands and options</summary> <br />

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot merge will merge this PR after your CI passes on it
• @dependabot squash and merge will squash and merge this PR after your CI passes on it
• @dependabot cancel merge will cancel a previously requested merge and block automerging
• @dependabot reopen will reopen this PR if it is closed
• @dependabot close will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually
• @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)
• @dependabot use these labels will set the current labels as the default for future PRs for this repo and language
• @dependabot use these reviewers will set the current reviewers as the default for future PRs for this repo and language
• @dependabot use these assignees will set the current assignees as the default for future PRs for this repo and language
• @dependabot use this milestone will set the current milestone as the default for future PRs for this repo and language

You can disable automated security fix PRs for this repo from the Security Alerts page.

</details>

This project has vulnerabilities that could not be fixed, or were patched when no upgrade was available. Good news, new upgrades or patches have now been published! This pull request fixes vulnerable dependencies you couldn’t previously address.

The PR includes:

• Changes to package.json to upgrade the vulnerable dependencies to a fixed version.

Vulnerabilities that will be fixed

With an upgrade:

• npm:braces:20180219

You can read more about Snyk's upgrade and patch logic in Snyk's documentation.

Note that this pull request only addresses vulnerabilities that previously had no fixes. See the Snyk test report to review and remediate the full list of vulnerable dependencies.

Check the changes in this PR to ensure they won't cause issues with your project.

Stay secure, The Snyk team

<!-- snyk:metadata:{"type":"remediation","packageManager":"npm","vulns":["npm:braces:20180219"],"patch":[],"ignore":[],"upgrade":["npm:braces:20180219"],"isBreakingChange":false,"env":"prod"} -->

Description

This PR fixes one or more vulnerable packages in the npm dependencies of this project. See the Snyk test report for more details.

Snyk Project: sholladay/build-version:package.json

Snyk Organization: sholladay

Changes included in this PR

• Changes to the following files to upgrade the vulnerable dependencies to a fixed version:
  • package.json

Vulnerabilities that will be fixed

With an upgrade:

• npm:mem:20180117 - potentially breaking change

You can read more about Snyk's upgrade and patch logic in Snyk's documentation.

Check the changes in this PR to ensure they won't cause issues with your project.

Stay secure, The Snyk team

Note: You are seeing this because you or someone else with access to this repository has authorised Snyk to open Fix PRs. To review the settings for this Snyk project please go to the project settings page.

This project has vulnerabilities that could not be fixed, or were patched when no upgrade was available. Good news, new upgrades or patches have now been published! This pull request fixes vulnerable dependencies you couldn’t previously address.

The PR includes:

• Changes to package.json to upgrade the vulnerable dependencies to a fixed version.

Vulnerabilities that will be fixed

With an upgrade:

• npm:braces:20180219

You can read more about Snyk's upgrade and patch logic in Snyk's documentation.

Note that this pull request only addresses vulnerabilities that previously had no fixes. See the Snyk test report to review and remediate the full list of vulnerable dependencies.

Check the changes in this PR to ensure they won't cause issues with your project.

Stay secure, The Snyk team

<!-- snyk:metadata:{"type":"remediation","packageManager":"npm","vulns":["npm:braces:20180219"],"patch":[],"ignore":[],"upgrade":["npm:braces:20180219"],"isBreakingChange":false,"env":"prod"} -->

Description

This PR fixes one or more vulnerable packages in the npm dependencies of this project. See the Snyk test report for more details.

Snyk Project: sholladay/first-name:package.json

Snyk Organization: sholladay

Changes included in this PR

• Changes to the following files to upgrade the vulnerable dependencies to a fixed version:
  • package.json

Vulnerabilities that will be fixed

With an upgrade:

• npm:mem:20180117 - potentially breaking change

You can read more about Snyk's upgrade and patch logic in Snyk's documentation.

Check the changes in this PR to ensure they won't cause issues with your project.

Stay secure, The Snyk team

Note: You are seeing this because you or someone else with access to this repository has authorised Snyk to open Fix PRs. To review the settings for this Snyk project please go to the project settings page.

Description

This PR fixes one or more vulnerable packages in the npm dependencies of this project. See the Snyk test report for more details.

Snyk Project: sholladay/generator-seth:package.json

Snyk Organization: sholladay

Changes included in this PR

• Changes to the following files to upgrade the vulnerable dependencies to a fixed version:
  • package.json

Vulnerabilities that will be fixed

With an upgrade:

• npm:mem:20180117 - potentially breaking change

You can read more about Snyk's upgrade and patch logic in Snyk's documentation.

Check the changes in this PR to ensure they won't cause issues with your project.

Stay secure, The Snyk team

Note: You are seeing this because you or someone else with access to this repository has authorised Snyk to open Fix PRs. To review the settings for this Snyk project please go to the project settings page.

Bumps electron from 4.2.12 to 9.1.0. <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/electron/electron/releases">electron's releases</a>.</em></p> <blockquote> <h2>electron v9.1.0</h2> <h1>Release Notes for v9.1.0</h1> <h2>Features</h2> <ul> <li>Added support for <code>MessagePort</code> in the main process. <a href="https://github-redirect.dependabot.com/electron/electron/issues/24323">#24323</a></li> <li>Added support for suspend and resume events to Windows. <a href="https://github-redirect.dependabot.com/electron/electron/issues/24283">#24283</a></li> <li>Added support for suspend and resume events to macOS. <a href="https://github-redirect.dependabot.com/electron/electron/issues/24294">#24294</a></li> <li>Expose sessionId associated with a target from debugger module. <a href="https://github-redirect.dependabot.com/electron/electron/issues/24398">#24398</a></li> <li>Implemented <code>systemPreferences.getMediaAccessStatus()</code> on Windows. <a href="https://github-redirect.dependabot.com/electron/electron/issues/24312">#24312</a></li> </ul> <h2>Fixes</h2> <ul> <li>Fixed an intermittent high-CPU usage problem caused a system clock issue during sleep. <a href="https://github-redirect.dependabot.com/electron/electron/issues/24415">#24415</a></li> <li>Fixed an issue where some old notifications were not properly removed from the Notification Center on macOS. <a href="https://github-redirect.dependabot.com/electron/electron/issues/24406">#24406</a></li> <li>Fixed bug on macOS where the main window could be targeted for a focus event when it was disabled behind a modal. <a href="https://github-redirect.dependabot.com/electron/electron/issues/24354">#24354</a></li> </ul> <h2>electron v9.0.5</h2> <h1>Release Notes for v9.0.5</h1> <h2>Fixes</h2> <ul> <li>Fixed "Paste and Match Style" shortcut on macOS to match OS's "Option-Shift-Command-V". <a href="https://github-redirect.dependabot.com/electron/electron/issues/24185">#24185</a></li> <li>Fixed "null path-to-app" in test-app when Electron's path contains spaces or special characters. <a href="https://github-redirect.dependabot.com/electron/electron/issues/24232">#24232</a></li> <li>Fixed an error when calling <code>dialog.showCertificateTrustDialog</code> with no <code>BrowserWindow</code>. <a href="https://github-redirect.dependabot.com/electron/electron/issues/24121">#24121</a></li> <li>Fixed an issue where <code>shutdown</code> would be emitted both on app <em>and</em> system shutdown on macOS. <a href="https://github-redirect.dependabot.com/electron/electron/issues/24141">#24141</a></li> <li>Fixed an issue where <code>withFileTypes</code> was not supported as an option to <code>fs.readdir</code> or <code>fs.readdirSync</code> under asar. <a href="https://github-redirect.dependabot.com/electron/electron/issues/24108">#24108</a></li> <li>Fixed an issue which would cause streaming protocol responses to stall in some cases. <a href="https://github-redirect.dependabot.com/electron/electron/issues/24082">#24082</a></li> <li>Fixed an issue with click events not being emitted on macOS for Trays with context menus set. <a href="https://github-redirect.dependabot.com/electron/electron/issues/24236">#24236</a></li> <li>Fixed delayed execution of some Node.js callbacks in the main process. <a href="https://github-redirect.dependabot.com/electron/electron/issues/24178">#24178</a></li> <li>Fixed tray menu showing in taskbar on Windows. <a href="https://github-redirect.dependabot.com/electron/electron/issues/24193">#24193</a></li> <li>Fixed window titlebar not responding to pen on Windows 10. <a href="https://github-redirect.dependabot.com/electron/electron/issues/24103">#24103</a></li> </ul> <h2>Other Changes</h2> <ul> <li>Fixed issue with some IMEs on windows (for ex: Zhuyin) don't terminate after pressing shift. <a href="https://github-redirect.dependabot.com/electron/electron/issues/24059">#24059</a></li> <li>Fixed mac app store rejection notice for invalid symbolic link in bundle. <a href="https://github-redirect.dependabot.com/electron/electron/issues/24238">#24238</a></li> <li>Updated Chromium to 83.0.4103.119. <a href="https://github-redirect.dependabot.com/electron/electron/issues/24234">#24234</a></li> </ul> <h2>Documentation</h2> <ul> <li>Documentation changes: <a href="https://github-redirect.dependabot.com/electron/electron/issues/24177">#24177</a></li> </ul> <h2>electron v9.0.4</h2> <h1>Release Notes for v9.0.4</h1> <h2>Fixes</h2> <ul> <li>Added missing support for <code>isComposing</code> KeyboardEvent property. <a href="https://github-redirect.dependabot.com/electron/electron/issues/23996">#23996</a></li> <li>Enable NTLM v2 for POSIX platforms and added --disable-ntlm-v2 switch to disable it. <a href="https://github-redirect.dependabot.com/electron/electron/issues/23934">#23934</a></li> </ul> <!-- raw HTML omitted --> </blockquote> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/electron/electron/commit/a822d2639a9c9c2c670e91d73f78e921865ce38e"><code>a822d26</code></a> Bump v9.1.0</li> <li><a href="https://github.com/electron/electron/commit/9d6ac058c9ba09e5211cd703675f70901468ca5a"><code>9d6ac05</code></a> feat: expose sessionId in debugger module (<a href="https://github-redirect.dependabot.com/electron/electron/issues/24398">#24398</a>)</li> <li><a href="https://github.com/electron/electron/commit/cbe66f27e725070c7c0c590a2e485978a554be19"><code>cbe66f2</code></a> fix: intermittent 100% CPU usage on macOS (<a href="https://github-redirect.dependabot.com/electron/electron/issues/24415">#24415</a>)</li> <li><a href="https://github.com/electron/electron/commit/4f10bde056d5fe175f338b5bc7b04baf422df5bb"><code>4f10bde</code></a> feat: implement systemPreferences.getMediaAccessStatus() on Windows (<a href="https://github-redirect.dependabot.com/electron/electron/issues/24275">#24275</a>) ...</li> <li><a href="https://github.com/electron/electron/commit/473c7dbe734fd2df68fa68f3a344e3a0222dfbf8"><code>473c7db</code></a> chore: use node_bindings loop for clarity (<a href="https://github-redirect.dependabot.com/electron/electron/issues/24418">#24418</a>)</li> <li><a href="https://github.com/electron/electron/commit/4515c4db00d94175179e47b0c4d05d7d4f0074a7"><code>4515c4d</code></a> feat: add app render-process-gone event (<a href="https://github-redirect.dependabot.com/electron/electron/issues/24315">#24315</a>)</li> <li><a href="https://github.com/electron/electron/commit/6dd394a155116cec542656165973a7888fe3938f"><code>6dd394a</code></a> fix: remove same-tag notifications before showing new ones (<a href="https://github-redirect.dependabot.com/electron/electron/issues/24406">#24406</a>)</li> <li><a href="https://github.com/electron/electron/commit/4ace4995c81f1b04ab99cc1707d037e9972e1316"><code>4ace499</code></a> feat: MessagePorts in the main process (<a href="https://github-redirect.dependabot.com/electron/electron/issues/24323">#24323</a>)</li> <li><a href="https://github.com/electron/electron/commit/71e329663e24fa4773dd4bdbf8fe71844de87bb8"><code>71e3296</code></a> feat: add new render-process-gone event (<a href="https://github-redirect.dependabot.com/electron/electron/issues/24309">#24309</a>)</li> <li><a href="https://github.com/electron/electron/commit/67002fdf11b7329cdba2e70ace583cb0b853e4cc"><code>67002fd</code></a> fix: macOS modal focus (<a href="https://github-redirect.dependabot.com/electron/electron/issues/24354">#24354</a>)</li> <li>Additional commits viewable in <a href="https://github.com/electron/electron/compare/v4.2.12...v9.1.0">compare view</a></li> </ul> </details> <br />

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

<details> <summary>Dependabot commands and options</summary> <br />

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot merge will merge this PR after your CI passes on it
• @dependabot squash and merge will squash and merge this PR after your CI passes on it
• @dependabot cancel merge will cancel a previously requested merge and block automerging
• @dependabot reopen will reopen this PR if it is closed
• @dependabot close will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually
• @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)
• @dependabot use these labels will set the current labels as the default for future PRs for this repo and language
• @dependabot use these reviewers will set the current reviewers as the default for future PRs for this repo and language
• @dependabot use these assignees will set the current assignees as the default for future PRs for this repo and language
• @dependabot use this milestone will set the current milestone as the default for future PRs for this repo and language

You can disable automated security fix PRs for this repo from the Security Alerts page.

</details>

Closes #49

You can validate that the pika URL was causing the issue by running deno cache -r https://cdn.pika.dev/[email protected].

This produced the thread panic error.

thread 'main' panicked at 'url redirection should provide 'location' header', cli/http_util.rs:141:29
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace

Switching to the skypack.dev dependency fixes this issue. I did not modify any other pika.dev routes, since it does not seem to be consistent for all.

I fixed this separately on master in https://github.com/sholladay/pogo/commit/64ef11371480e32eb8323ddbe3ad46781d1d75ea. Should've pushed it to this PR first... oops.

Turned out to be a little tricky to fully debug and fix due to https://github.com/denoland/deno/issues/7380 and the deno_types repo recommending denopkg.com, which doesn't actually work because of that Deno bug.

Thanks for the useful bug report and getting the ball rolling on this!

Pika is no longer returning this package. If you request it, it's a 302 without a location. This causes Rust to panic.

curl the file, and you get a message that you should use this - https://cdn.skypack.dev/csstype

thread 'main' panicked at 'url redirection should provide 'location' header', cli/http_util.rs:141:29
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace

I can not tell what dependency in Pogo that is requesting this. But I have isolated it as the use of pogo that has caused this.

https://cdn.pika.dev/csstype/dist=es2019,mode=types/index.d.ts

You can see this by running a test with -r to redownload.

The snowpack url does work though. Even though I am not using this feature in pogo, it impacting me. Maybe some of them react support could be a plugin in the future. I have a feeling this may be from the inclusion of react/jsx with pogo.

Fixed in https://github.com/sholladay/pogo/commit/64ef11371480e32eb8323ddbe3ad46781d1d75ea.

I decided to migrate back to the JSPM CDN for now, because Pika is too unstable.

I also confirmed that it's not specific to denopkg.com because the same problem occurs if you use https://github.com/<owner>/repo/raw/<path>.

I wonder if it could have something to do with the redirect location pointing to a different origin?

When the URL given to @deno-types returns an HTTP redirect, Deno seems to ignore the response rather than following it.

In https://github.com/hapipal/boilerplate/issues/91#issuecomment-677774990 it was mentioned that the hapipal boilerplate's use-case for having empty .env and .env.example files seemed reasonable and that a PR would be welcome. This is that PR ☺️

To document the use-case: the boilerplate supports a single, optional PORT environment variable with a default set via runtime validation. It's desired to not force the user to fill this environment variable in manually during installation. The standard way to handle optional variables with envy is to just leave them commented-out in .env.example, but in our case that would result in an empty example file, which envy does not allow. This PR removes the check for an empty example file, which is a breaking change.

This is looking good so far, thanks Devin!

Drop node v8 support

👍

Test against windows

Seems like this PR closes #3, then. Nice. One thing I'm not sure about is whether Travis CI's Windows environment provides a useful representation of file permissions, as I believe it uses git bash, which of course has to translate *nix style permissions to Windows style permissions when our setup-tests.sh script runs, and this conversion is "lossy" as the permission models are fairly different. The end result is that, while setup-tests.sh may run without errors, I'm uncertain as to whether it's a good approximation of real (non-bash) Windows usage. On a related note, according to https://github.com/sholladay/envy/issues/4, our Windows support may be broken as the user reports that chmod 555 doesn't really work (presumably due to the lossy conversion), though I haven't gotten around to reproducing that bug report yet. Ultimately, we just need to figure out what the strictest set of permissions are (from Node's fs.statSync() perspective) that can be set on Windows that are as close as possible to 600, then update the code accordingly. (Note: I'm okay with requiring permissions that cannot be set from git bash via chmod, as long as there's some way to do it, whether it be through PowerShell or otherwise).

Fix all lint warnings ... Admittedly the 30 statement limit within functions made me write some things differently than I would have liked, but it still seems like a good result.

Ah, yes, the max statements rule, I've run into that before. Something about this codebase makes me want to write one long function with more statements than I usually would. I think because the internals are fairly domain specific and there's not a whole lot of isolated or reusable logic.

I'd actually prefer for you to write the code as clearly / elegantly as possible and then disable that rule if necessary. I'll either refactor it myself or loosen up the linter configuration in package.json to clean up the noisy comments after merging.

Support env option to override implicit process.env input

I like it, generally. How do you see the interplay between this and https://github.com/sholladay/envy/issues/6? Should envy.assign() assign the result to process.env directly or assign to the env option object, which merely defaults to process.env? The latter seems reasonable, but what if a user wants to use both a custom env object as input and assign the output to process.env? I suppose we could have an assign option that takes either a boolean or an object to assign to, but envy({ assign : someObject }) seems confusingly named for that behavior. Thoughts?

Support allowUnknown option to allow specific vars not listed in .env files

Did you see the CLI tip? It sort of covers this use case and that same pattern could be extended to mix in environment variables, but I'll grant that it may not be as simple as desired. I'm open to allowUnknown (should probably be named stripUnknown, though).

Sounds like they don't detect the try. Might be worth asking the Yarn folks whether this is intentional. I would assume not, because if an optional dependency can't be required(), then what's the point of having it? They might be using a different definition of "optional" or something, though.

I don't see the problem here. We mark it as optional because if the dependency is not installed, we catch the load error and don't do anything.

Ugh. That's so annoying, thanks for the PR. I think the best solution may be to revert https://github.com/sholladay/pogo/commit/e5cc45c0a1fb945d336ebb2ea2c9f5e70d1271b8. Multiple times now I've changed Pogo to use Pika when I see it's working well and then some new issue comes up. It's so unstable.

Timeout means to give up on the request, e.g. to avoid having users stare at a loading spinner forever. I can see how this could be a little confusing since if the request seems to be hanging, then one could argue it is in a failure state and we retry other failure states. However, the difference is that in the other cases we have a definitive result, whereas a timeout is more like a judgement or prediction without any contextual information. Retrying a request just because it's taking a long time is probably not advisable in most cases. The server might be overloaded and retries would exacerbate the problem. The internet connection might be slow and retries would probably not improve the overall throughput (given that the request would be restarted from scratch).

What I would recommend is setting a relatively long timeout like 30 seconds (or even disable timeouts altogether) and rely on the browser to report failures which Ky will retry. If you're still experiencing hanging requests, it would be good to investigate a little further to determine what is causing it.

This seems like some kind of proprietary environment issue. I see nothing that justifies why this logic belongs in Ky specifically. If anything, it sounds like it affects all node-fetch users and should be fixed there. But ultimately, it's probably a wider issue with the module loader.

At the time I was thinking that an empty example file would likely be a mistake. But your use case seems valid. Let's remove that assertion. PR welcome.

I'm happy to help work on this if you run into any issues, but I'm currently in the process of selling my house so fairly busy for a few weeks.

If anyone does want to pursue making Got work on Deno, the best course of action would be to contribute to Deno's std/node library, which is a compatibility layer designed to provide the Node APIs that Got needs to run. It already has some of the required Node APIs, including fs and events, but it's missing others such as net, http, tls, and stream, for example. Most of the missing APIs should be relatively easy to add and also more approachable than modifying Got itself.

Pogo v0.5.0 has been published with these changes. Great work, @afaur!

Might be helpful to link out to the code in readme examples. Links would use a tag or sha to avoid taking the user to the latest changes.

• Update: This might introduce a lot of maintenance though, and I am not sure about it.

server.router.get(String, RouteHandler<ServerRequest, Toolkit>)

// Javascript Example
server.router.get('/', (request, { file }) => {
    return file('./foo.jpg');
});

// Typescript Example
server.router.get('/', (request : ServerRequest, { file } : Toolkit) => {
    return file('./foo.jpg');
});

Thanks, good idea!

Done: https://github.com/sholladay/pogo/commit/fef521c0b065bb5da627d4e40780a0187312b441

I added TSDoc comments to further improve the deno doc output in commit https://github.com/sholladay/pogo/commit/8fa9c0c77be12cffc8a52cc8f51ad50c13366ab4. I also updated the language of the code examples in the documentation to enable GitHub's link-to-definition feature in commit https://github.com/sholladay/pogo/commit/4d24945730f5c4892bed90103be625e6189dd5d3.

With your PR, plus these changes, I think we can probably close this. What do you think?

I'm open to any other suggestions for improving the documentation. The next logical step I can think of would be to have a website for the docs.

If you're comfortable with that, then great, that's another option. IMO, though, duplicating dotfiles on every machine is gross. Among other problems, you'll have to keep those files in sync, which you may easily forget to do.

Can't this issue be closed now that PR #558 has been merged? That PR closed issue #466, which seems like a duplicate of this issue.

The solution is to downgrade to Husky v3.

@Alex0007 maybe you could edit the original issue comment to mention the solution so it's more visible to people?

• Make sure names show up in deno doc command.
• Add mod entry points
  • Don't have to have these but..
    • mod.ts in lib/util simplifies main.ts and mod.ts by making them not change if files get added to util.
    • mod.ts in lib allows main.ts and mod.ts to have less duplication.
  • Maybe main.ts should be removed from root directory to use mod.ts as other Deno projects seem to do?
  • main.ts and mod.ts in the root directory are exactly the same right now.
    • If both are kept, maybe one should only export the Main object that contains server and router?
      • Doing this might provide a slimmer entry point (less imports), but I am not really sure how important that is.

Update:

• Possible alterations (https://github.com/afaur/pogo/commit/248d692c1bcca8ea1b11986ab6465130dce17eca).
  • Rename lib/mod.ts to lib/main.ts
    • This file does not really function as a mod here.
    • This might be a good change to make.
  • Remove main.ts and fix tests
    • Probably not a good decision since users of the library would have to change imports.
  • Change mod.ts in the root to a one line export
    • deno doc stops showing the main entry point from the output with this change.
    • This might be okay.
      • The object/class that main provides, contains server and router functions.
        • Outside of the object/class they have their own entries in deno doc.