search

lizmat / App-Raku-Log

Cro application for presenting Raku IRC logs
Artistic License 2.0
3 stars 2 forks source link

Numerous validation failures #6

Open raydiak opened 2 years ago

raydiak commented 2 years ago

https://validator.w3.org/nu/?doc=https%3A%2F%2Flogs.liz.nl%2Fhome.html https://validator.w3.org/nu/?doc=https%3A%2F%2Flogs.liz.nl%2Fraku%2F2021-10-15.html

While some validator complaints are often rather pedantic, it's wise to check anyway as it can also highlight real problems. This output points to a variety of actual issues including repeated use of IDs (which will cause problems when trying to access elements from javascript or anchor links), malformed tags and structure (which might look fine in some browsers but have different failure modes in others, and may confuse screen readers, search engine crawlers, etc), and lack of escaping for many codepoints (most of which probably shouldn't be in the document at all).

lizmat commented 2 years ago

Thanks for the report!

lizmat commented 2 years ago

I've fixed some of the obvious ones. With regards to using duplicate ID's, those ID's are not to be used externally, and are intentionally the same to allow for similar functions (such as the search form) in multiple views.

@Altai-man any comments?

Altai-man commented 2 years ago

Hi, @raydiak ! Do you experience some visual and/or behavior issues with the website? If yes, please report them.

As for the warnings, we will address the bogus ones, but please do not expect it to become 0 warnings, as the checker itself sometimes suggests bogus things. For example, if you check some very popular web pages like https://validator.w3.org/nu/?doc=https%3A%2F%2Fwww.github.com https://validator.w3.org/nu/?doc=https%3A%2F%2Fwww.pixiv.net https://validator.w3.org/nu/?doc=https%3A%2F%2Fwww.amazon.com https://validator.w3.org/nu/?doc=https%3A%2F%2Fwww.youtube.com https://validator.w3.org/nu/?doc=https%3A%2F%2Fwww.wikipedia.org you will see they have plenty of warnings.

raydiak commented 2 years ago

That's why the very first thing I wrote in this issue was "While some validator complaints are often rather pedantic", and then went on to highlight a few of the specifically concerning points. I'm not trying to put down your work, and I appreciate that you've done it at all. But as a professional web developer of nearly two decades who knows precisely what he's looking at, I thought you should take a look at these, more the errors than the warnings.

Duplicate IDs are and always have been invalid and likely to cause problems unless they're entirely unused. Maybe that's why the search form throws a JS error on sumbit, idk didn't take the time to look closer. There are stray closing tags and other potential structural issues, maybe that's why someone on IRC was reporting elements stacking in the wrong Z order under certain circumstances.

The metavalue of my point is simply: "don't forget to check the error console and the validator," not "you should expect absolutely no output from the validator." I have nothing else to add and am going to unsubscribe from this issue, close it if you like. Best of luck.

Altai-man commented 2 years ago

I'm not trying to put down your work

No worries, no offense taken. I understandably have zero knowledge about your expertise, so I explicitly wrote "we will address the bogus ones" and appended a small note to prevent a possible misunderstanding (it still has happened, alas).

I am sorry if my attempt to write "Do you see issues? Please report, they are welcome" and "Sure, we will fix the errors" was taken as something hostile, I did not intend anything of this sort and fail to see from my perspective why shouldn't you have more confidence in yourself and your skills.

If you know what I wrote about warnings and agree with it, then we share some knowledge and both agree on it, it's a good thing then, is it not? I don't think I implied "Oh, you are an amateur, you don't even know this thing, let me explain it to you, you cinnamon roll, now go away", nothing of this sort.

Altai-man commented 2 years ago

I've addressed some of the simpler ones. The rest go about as follows:

lizmat commented 2 years ago

I think either Cro::WebApp should handle the code point ones, or we should ignore those.