w3c validator doesn't like dynamic url's?

Dj_Jestar

Dj_Jestar

Dj_Jestar

Caporegime
Joined
18 Oct 2002
Posts
29,493
Location
Back in East London
Just checked over my page output on the validator, to find numerous errors such as:

Warning Line 58 column 50: cannot generate system identifier for general entity "file".
...><td><a href="index.php?download=yes&file=addproducts.php">addproducts.php</a
Click to expand...

now.. is that an actual 'fail' or the validator's parser is unable to depict links properly?

(and before anyone get's nitpicky - I will be using the full address in links, this is still just a test page :p)
 
Dj_Jestar said:
Just checked over my page output on the validator, to find numerous errors such as:



now.. is that an actual 'fail' or the validator's parser is unable to depict links properly?

(and before anyone get's nitpicky - I will be using the full address in links, this is still just a test page :p)
Click to expand...

Why would you want to use full address links? You should use relative.
 
not really, it's a question of which standards will actually affect someone's useability, and which are crap and can be completely ignored. the standards have become so convoluted that they're destroying any chance of having uniformity in all browsers - which is their point.

the truth of the matter is when you work in the web development business 99% of companies aren't going to care about accesibilty etc. etc. when it only covers a tiny % of their visitors. What they care about is having a fully functioning, pretty site with CMS in x days, and if that deadline means you can't dick around worrying about standards then they get left out.

Sad as it may be, that's how it is.
 
Last edited:
Spunkey said:
not really, it's a question of which standards will actually affect someone's useability, and which are crap and can be completely ignored. the standards have become so convoluted that they're destroying any chance of having uniformity in all browsers - which is their point.

the truth of the matter is when you work in the web development business 99% of companies aren't going to care about accesibilty etc. etc. when it only covers a tiny % of their visitors. What they care about is having a fully functioning, pretty site with CMS in x days, and if that deadline means you can't dick around worrying about standards then they get left out.

Sad as it may be, that's how it is.
Click to expand...

While I agree with you that some standards are a pain and somewhat subjective that does not mean that they are all difficult to implement.

I don't know who your clients but I find with mine that accessibility is starting to play a big part in aspects of site design and usability. Add to that the legal framework starting to form around accessibility of websites and it should be something that's on your radar. When building a site applying good coding practices and using standards compliant code should not slow down the development of the site in question.
 
Spunkey said:
standards have become so convoluted
Click to expand...
Huh? :confused:

I can knock out a compliant site in minutes with Notepad....

Besides, correctly encoding ampersands is nothing to do with 'standards', it's more to do with 'basic computer science'....
 
Spunkey said:
pfff. and people wonder why standards aren't always adhered to.
Click to expand...
Do you even know why ampersands aren't allowed in unescaped form or are you just being reactionary for the sake of it?

Ampersands are the opening delimiters for entity references, just as left angle-brackets (<) are opening delimiters for tags. Yeah, stupid bloody standards, what are those people thinking!?

Code: 
<div><p>My friend has a great new cd-player&amplifier that cost &pound;3000; but, to be honest with you, he's a bit of a <<insert expletive here>> </div>
:cool:
 
Dj_Jestar said:
In my defence.. I was fully aware of the need to use HTML entity 'codes' for output.. just not in href's :o
Click to expand...
They are completely legal in URIs, so it's quite common to see them unescaped within markup. Most browsers likely have to employ some of their convoluted error correction code to recover from it, which is why it's not normally considered a major problem.

Spunkey said:
like i said before, some standards are worthwhile, others... not so much :)

it's down to the developer to realise which can be negated.
Click to expand...
And do you consider having a defined set of delimiters to be a worthwhile standard to adhere to? Out of curiosity, how would you parse the code snippet I posted?
 
KingAdora said:
Why would you want to use full address links? You should use relative.
Click to expand...
There are a very small number of browsers that do not support relative URL's, and given all browsers support full URL's, the choice is a no brainer :)

The links will be generated dynamically, so there won't be any problems with cross-domain issues etc.
 
Spunkey said:
like i said before, some standards are worthwhile, others... not so much :)

it's down to the developer to realise which can be negated.
Click to expand...

Out of interest do you develop with standards in mind or produce what your client requires no matter how you do it? For me it's more of a pride thing that makes me want to produce good code no matter what the client asks for.
 
You must log in or register to reply here.
Back
Top Bottom