You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This creates some confusion as searching the code base (e.g. to fix a typo) leads to multiple hits - but only the .md files should be edited. It also means some people edit both .md and .html files when they don't need to.
We could remove these files from the Git repo (and add the relevant .gitignore entries to ensure they are not re-uploaded).
The generated HTML files will still be generated, tested and linted as part of our GitHub action so don't think we lose anything with that.
We should also add the npm run generate to the npm run deploy script as they need to be generated and uploaded to App Engine to actually run the site.
The one downside is developers will now need to run npm run generate to generate these .html files, to be able to run and be able to develop the website. So a developer just making some CSS changes but nothing else, now has an extra step to complete. Do we see that being a problem?
Thoughts?
The text was updated successfully, but these errors were encountered:
I find it relevant to remove generated elements from the tracked code. In the development context, even for CSS only, I don't think the npm run generate is any more complicated than installing the dependencies via Python and running the web server. If you do the latter, I doni’ think s.o. will have any difficulties with the former.
One thing where storing the assets is useful is when EJS upgrades (like in #1225 ) - we can run the script and see no templates have changed and so know the upgrade does not impact us. In the past we've noticed changes to the HTML generated. Not an insurmountable problem, and it is rare, but it is something that is a little tougher if we do this.
@HTTPArchive/developers , question for you all:
We auto generate some content, including:
.html
files (from chapters.md
files)sitemap.xml
featured_chapters.html
(depending on outcome of Auto generate featured chapters #1104).html
and.pdfs
This creates some confusion as searching the code base (e.g. to fix a typo) leads to multiple hits - but only the
.md
files should be edited. It also means some people edit both.md
and.html
files when they don't need to.We could remove these files from the Git repo (and add the relevant
.gitignore
entries to ensure they are not re-uploaded).The generated HTML files will still be generated, tested and linted as part of our GitHub action so don't think we lose anything with that.
We should also add the
npm run generate
to thenpm run deploy
script as they need to be generated and uploaded to App Engine to actually run the site.The one downside is developers will now need to run
npm run generate
to generate these.html
files, to be able to run and be able to develop the website. So a developer just making some CSS changes but nothing else, now has an extra step to complete. Do we see that being a problem?Thoughts?
The text was updated successfully, but these errors were encountered: