feat(up): checkout and use helix-pages if no source scripts are present

[tripodsan]

fixes #929

[codecov]

Codecov Report

Merging #946 into master will increase coverage by 0.16%.
The diff coverage is 96.77%.

@@            Coverage Diff             @@
##           master     #946      +/-   ##
==========================================
+ Coverage   91.55%   91.71%   +0.16%     
==========================================
  Files          45       46       +1     
  Lines        1848     1908      +60     
==========================================
+ Hits         1692     1750      +58     
- Misses        156      158       +2

Impacted Files	Coverage Δ
src/up.cmd.js	87.66% <100%> (+2.62%)	⬆️
src/helix-pages.js	93.93% <93.93%> (ø)

---

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 9d95b79...3264c9b. Read the comment docs.

[kptdobe]

hummmm... why copying the whole helix-pages project as is ? I think this is weird...
Why dumping scrani.js especially if you do nothing with it...
I would simply create a default basic html.htl, html.pre.js and minimal style.css since this is the only thing you want to test for now.

[tripodsan]

hummmm... why copying the whole helix-pages project as is ? I think this is weird...

do you mean in the tests? I want to be able to run the test offline:

helix-cli/test/testUpCmdPages.js

Lines 42 to 45 in 0c09cda

	// init helix-pages fake local repo. we do this, because we don't want git checkouts in
	// the helix-cli repository, and we want to keep the tests offline
	await fse.copy(path.resolve(__dirname, 'fixtures', 'helix-pages', 'master'), helixPagesRepo);
	initGit(helixPagesRepo, 'https://github.com/adobe/helix-pages.git');

Why dumping scrani.js especially if you do nothing with it...

well, it's part of helix-pages at the moment.

I would simply create a default basic html.htl, html.pre.js and minimal style.css since this is the only thing you want to test for now.

ok. I remove it.

[kptdobe]

I think now it is worst... the code still contains scrani references in multiple places.
My point is more to the fact that we do not need so much code in this context. Your test only asserts that when there is no code in the project, some default code is provided and some rendering is executed. The "default code" provided does not need to be a copy of the current helix-pages project and could / should be something really basic to limit the amount of lines of code we introduce in the cli project. Really, this super long style.css do not bring any value here. Same for html_plain.htl, html_planing.pre.js...

[kptdobe]

Suggested change

<script src="/scrani.js"></script>

[kptdobe]

Suggested change

scrani.onload();

[tripodsan]

I think now it is worst...

ok ok ....

[tripodsan]

@kptdobe I initially thought, that we could add more and more tests, to see if the html is rendered correctly. but now I think this should be done in the example project instead. but you're right. I remove the extra files.

[adobe-bot]

🎉 This PR is included in version 4.1.0 🎉

The release is available on:

• npm package (@latest dist-tag)
• GitHub release

Your semantic-release bot 📦🚀