Load third-party libraries using Cdnjs

[stephengroat]

Include semantic ui from cdn.

still need a tiny bit of help, my font-size override trick works for everything except the icon in the header

[mxxcon]

Let's implement SRI tags. You can use https://www.srihash.org/ to speed things up.

[stephengroat]

OK. Let's get the sizing down first

…

On Sun, Mar 5, 2017, 21:27 mxxcon ***@***.***> wrote: Let's implement SRI tags. You can use https://www.srihash.org/ to speed things up. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#2427 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABGv4hhdY2HngB-g9t7ldSa8ZJaDMCklks5ri5lIgaJpZM4MTuwG> .

[fpigerre]

Awesome stuff @stephengroat! 👍

At the moment, the Semantic-UI library sets the icon font-size to 2em through the following classes:
.ui.icon.header .circular.icon, .ui.icon.header .square.icon {
It's quite likely that somewhere, either in our own CSS or in Semantic-UI's CSS, there is a line of code that sets a base font-size lower than 18px specifically for a class which affects the logo icon. I'll have a dig around over the next day or two and see what I can find.
You can also see exactly which classes have what effects on particular elements by right-clicking on an element and selecting "Inspect". From there, specific classes provided for by the CSS appear in the Styles tab. The "Computed" tab is also really useful, as it shows the whole collection of styles applied to that specific element, regardless of how they were applied.

Would you be able to re-factor the style="font-size:18px" snippet into the css/base.scss file by any chance? This helps keep all the CSS in one place.

Thanks,
psgs 🌴

[fpigerre]

• Refactor style tag into base.scss
• Fix logo icon sizing
• Implement SRI hashes

[stephengroat]

@PSGS thanks for the help, i don't really know what i'm doing with css (learning curve is steep).

i moved the font-size to the css and it seems to fix the icon issue, could you look the size over and see if you can tell any major differences? i tried it on chrome and firefox and it looks the same (to me)

[fpigerre]

Awesome work @stephengroat! 👍

The only difference with the sizing of elements that I can see is a slight difference in the width of table columns. It's probably caused by several elements gradually inheriting a larger font size. Visually, the difference is negligible, however, so I think we can go ahead and merge the pull request.

@Carlgo11, @mxxcon: Should we specify one particular type of hash to use for SRI hashes (eg. sha-256), or allow different hash types to be used for each resource?

[mxxcon]

Should we specify one particular type of hash to use for SRI hashes (eg. sha-256), or allow different hash types to be used for each resource?

I think it doesn't really matter as long as they are secure hashes(not sha1/md5). We could stick to one type just for consistency sake.

[jamcat22]

I think if we decide to stick to one type, we should use sha-384, since that's what https://www.srihash.org uses when generating hashes.

[stephengroat]

The srihash.org generated hashes weren't working for the Google fonts, so I switched to Chrome generated 256 that worked

…

On Tue, Mar 7, 2017, 06:07 Jacob ***@***.***> wrote: I think if we decide to stick to one type, we should use sha-384, since that's what https://www.srihash.org uses when generating hashes. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#2427 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABGv4oJLDAXRnJ83Zk3UYEVCZ_cyz4_4ks5rjWSmgaJpZM4MTuwG> .

[stephengroat]

@PSGS i'm still seeing a problem for the backup entry pogoplug

do you see anything?

[fpigerre]

@stephengroat, I can't seem to see any problems with Pogoplug.
Do the screenshots below seem right to you?

[stephengroat]

@PSGS you're 100% right. i was comparing against an unmerged live copy that still had the in progress badge on it and was getting VERY confused!

[stephengroat]

@2factorauth/collaborators can some more people look this over? i think this one needs more than 2 reviews

[mxxcon]

I think we should NOT use SRI for Google Fonts. Apparently CSS they serve is intentionally unversioned and it changes depending on which browser requests it.
See google/fonts#473

[jamcat22]

Is it possible to change the hashes of our jquery source to be sha384?

[fpigerre]

👍