Detect duplicate HTML attributes

[winniehell]

Current behavior is that duplicate attributes in HTML tags overwrite previous values:

$ kramdown -ihtml -okramdown <<<'<a href="a" href="b">c</a>'
[c](b)
$ kramdown -ihtml -okramdown <<<'<img src="a" src="b" alt="c" alt="d"/>'
![d](b)

I would at least expect a warning in such a case.

[gettalong]

Duplicate HTML tags have to be dealt somehow and using the last found one seems to be good comprise for bad input. However, a warning really might be useful.

[winniehell]

I would state that a tool like kramdown may assume valid HTML when the input type is set to HTML—and duplicate attributes are not valid HTML.

Nevertheless, when processing large HTML files or a large amount of HTML files (as in my case 😉 🐠) it is very helpful to know every discomfort of the parser.

So maybe, I would label this issue rather a feature request than a bug report.

[gettalong]

When an HTML attribute is overwritten, a warning will now be generated.