Is there a way to stop the ampersand symbol from being converted to html in a URL? about kramdown

[treble37]

The Amazon url's seem to redirect to the correct item even with the converted "&amp"... now I'm curious as to why Kramdown converts the ampersands in a URL instead of leaving them alone.

It is correct behavior for kramdown to encode ampersands.
ref

[treble37]

svnpenn, now I'm a bit confused...the doc you pointed to says "When writing the same URL in a plain text email message or in the location bar of your browser, you would use "&" and not "&"."....Kramdown is converting it to the HTML "&amp" in the clickable URL....

No.

$ kramdown <<< '[some link](http://amazon.com?foo=111&bar=222)'
<p><a href="http://amazon.com?foo=111&amp;bar=222">some link</a></p>

[gettalong]

@svnpenn is correct.

Values of all attributes in (X)HTML must be correctly formatted so that no problems arise. Modern browsers are able to recover from basic errors but it is better to be on the safe side.

As stated on W3C's HTML documentation the href attribute of an <a> tag must contain a %URI string which is defined as containing CDATA. There you will find that character entities must be decoded, hence a lone ampersand must only occur as &.

As a general rule: In HTML text and attribute values all <, > and & must only occur as entities (i.e. <, > and &).

[treble37]

Thanks gettalong and svnpenn I learned something new...I missed this part in svnpenn's ref: "With HTML, the browser translates "&" to "&" so the Web server would only see "&" and not "&" in the query string of the request." -- I saw the "&amp" in the Firefox browser bar but didn't realize that the webserver was seeing "&" as it should.

[gettalong]

You are welcome! It could have been a problem with kramdown, so better safe than sorry 😄