Template talk:Roundy

`{{roundytl}}`	`{{roundytop}}`	`{{roundytr}}`
`{{roundyleft}}`	`{{roundy}}`	`{{roundyright}}`
`{{roundybl}}`	`{{roundybottom}}`	`{{roundybr}}`

Addition

border-{{#switch: {{{2|}}}|tl=top-left-|tr=top-right-|bl=bottom-left-|br=bottom-right-}}radius: {{{1|10px}}}; -moz-border-radius{{#switch: {{{2|}}}|tl=-topleft|tr=-topright|bl=-bottomleft|br=-bottomright}}: {{{1|10px}}}; -webkit-border-{{#switch: {{{2|}}}|tl=top-left-|tr=top-right-|bl=bottom-left-|br=bottom-right-}}radius: {{{1|10px}}}; -khtml-border-{{#switch: {{{2|}}}|tl=top-left-|tr=top-right-|bl=bottom-left-|br=bottom-right-}}radius: {{{1|10px}}}; -icab-border-{{#switch: {{{2|}}}|tl=top-left-|tr=top-right-|bl=bottom-left-|br=bottom-right-}}radius: {{{1|10px}}}; -o-border-{{#switch: {{{2|}}}|tl=top-left-|tr=top-right-|bl=bottom-left-|br=bottom-right-}}radius: {{{1|10px}}};<noinclude>[[Category:Table templates]]</noinclude> This will add KHTML, iCab, and Opera support. I can prove Opera for sure.--immewnitythe mew 19:06, 6 May 2010 (UTC)

There's also the issue of bloat, too many CSS workarounds slow load times slightly. iCab is barely used, for example, so it's not really worth it to slow down load for everyone just so a small fraction of users see rounded corners.

KHTML is slightly more widespread, because Safari also uses it, not just Konqueror. I'll need to test it a bit more to be certain, but we can switch that one out.

I'm not sure about the Opera code either; I'll test it on the older version of Opera I have on my laptop, but for future reference, Opera, as of recently, uses the plain border-radius code, and doesn't support the -o-border-radius or -opera-border-radius. -- Jïörüjï Ðērākō.>.cнаt^ 22:30, 6 May 2010 (UTC)

Safari uses -webkit
I've tested just -o on Opera 10 alpha and it worked.--immewnitythe mew 22:22, 3 June 2010 (UTC)

And put the standard border-radius first!--immewnitythe mew 22:23, 3 June 2010 (UTC)

There's an unneeded > after noinclude...--immewnitythe mew 18:00, 6 June 2010 (UTC)

I believe different versions of Safari use different engines, but it's a moot point since we have KHTML and webkit both covered.

I don't believe -o is disabled entirely on newer versions of Opera, but I know for a fact that the newest versions of Opera use plain border-radius already (getting a head-start on CSS3 specs).

The extra > isn't hurting anything, just a mistake when I was copy-pasting.

And lastly, do you have a particular reason why you say border-radius should be first? From what I know about CSS in general, it should usually be last in line for this sort of code. -- Jïörüjï Ðērākō.>.cнаt^ 05:48, 8 June 2010 (UTC)

Opera never supported -o-border-radius. Also generic border-radius property should go *after* vendor specific ones. — Tenno Seremel 08:52, 21 July 2010 (UTC)

Roundycorners are dead.

Hope you guys are happy. TTEchidna 01:52, 16 June 2010 (UTC)

Oh stop pouting. They're not necessarily dead yet. That this template is partially to blame for the issues (never said it's the whole thing) is still a theory that's worth testing. If the roundy's were to blame....then we should try and find a way to do them without them causing issues, because I personally do like them roundy. --Archaic 04:24, 16 June 2010 (UTC)

No, I'm not happy. >:{ Although I don't believe we need to include the -webkit- prefix anymore as the newest versions of Safari and Chrome now support regular old border-radius (which I just discovered the other day), as does the latest version of Opera (no more need for -o-border-radius). Perhaps we could uncomment the regular and -moz- versions, leaving the rest hidden, and see what happens? Quagbert 14:03, 16 June 2010 (UTC)

Roundycorners are back!

Sorta.

We've split roundy into nine pieces. Currently, we have to fix a lot of the templates and pages whose coding is broken by this change. Should be done ASAP, if the servers cooperate. TTEchidna 08:27, 25 June 2010 (UTC)

Internet Explorer

Is there a way to make roundy work for IE? Pikiwyn ^talk 00:56, 6 December 2010 (UTC)

Not that I know of. IE currently does not support this. —♥ Jello ^talk 01:02, 6 December 2010 (UTC)

It works with IE 9 yessss! Pikiwyn ^talk 18:57, 17 January 2011 (UTC)

It's time to remove the prefixes.

Please remove the prefixes (and the extra spaces); none of the prefixes are needed:

iCab is dead. The versions now all use Safari's engine, and the really old versions that used their own engine have SSL errors.
I am pretty sure Konqueror uses WebKit now.
Opera never needed a prefix, and Opera Mini doesn't have support for rounded corners at all.
Both versions of Safari and Chrome that needed the prefix have had no version cut off since then.
Firefox 3.6 can't seem to load Bulbapedia properly, it whites out.

Why?

The data usage increase is ridiculous. Some pages have their size reduced by more than 70% by removing prefixes. Every time the template is used, 135 bytes (365 in Roundytop) of unnecessary code is used, and the template is used a lot. Even with gzipped compression (the website appears to be gzipped with no compression), these large pages take up RAM, parsing time, and cache once decompressed.
Bulbapedia doesn't look half bad with square corners.
Bulbapedia isn't even consistent with the prefixes: MediaWiki only uses -webkit and -moz prefixes (which you could also use if you were insistent on those browsers), and the homepage CSS has only the unprefixed version
This hasn't been updated in 8 years. Browsers have changed.
No modern websites use more than -moz or -webkit, and frankly, very few that I've seen use those prefixes either. Not even Google.
Most of the users viewing this page are tech savvy enough to keep their browser up to date.

I mentioned this a few times on Discord a while back, yet still nothing has been done about it. Easyaspi314 (talk) 02:29, 16 April 2018 (UTC)

Such an edit would cause significant stress to the server as well as potentially lengthy downtime. Ultimately when the page loads an additional 10 MB of images, scripts, styles, some thousand bytes make no significant impact. At the moment, the benefits (pages being a fraction of a percent smaller overall) don't outweigh the negative impacts of major technical issues. --Abcboy (talk) 03:24, 16 April 2018 (UTC)