remove all a href....

[DiapDealer]

Earlier I mentioned that I was worried that:

</?a ?([^>]+)?>

might find other tags that started with 'a'--and indeed it does. The addr, abbr, and area tags are probably able to be dismissed, but the <aside> tag is one that I'm sure we're only going to see more and more of. And my regex will include it.

So for the paranoid/pedantic type (like myself), it's probably best to use:

</?a\b([^>]+)?>

instead (should work in pretty-much all regex flavors).

The \b just matches a "word" boundary so that no other tags that start with 'a' will be caught up in the match.

[cybmole]

Id never heard of <aside> i had to google it. it doesnt seem like anything I'd ever expect to encounter in an epub though - would ADE even have a clue what to do with it?

So I think I'll just go on happily nuking everything beginning with <a !

I have an HTML5 cuick guide on my tablet so lets see what else there is in its index for a:
acronym - though its says that is now unsupported
address
article - new for html 5

[DiapDealer]

Fair enough. As I said; my change is more of a "stickler for details" thing.

But if you're cleaning up the cruft from retail epubs, you're going to start running into the <aside> tag sooner rather than later. Both Kobo and B&N (and even Kindle) books are exhibiting more and more html5/epub3 features.

Deleting the <aside></aside> tags and leaving their contents intact could make for some very confusing reading (footnotes and other ancillary details stuffed into the middle of sentences and the like).

[theducks]

Originally Posted by cybmole

thanks - now I'll have to remember where I might have saved a test case

does that code strip the nested tags from inner to outer, as usually the outermost one is the best candidate for keeping ?

No, it is pretty stupid, when I can, I usually wrap that in a
(<body.*>)\s*
and
\s+</body>

including putting those back in the replace


I do frequent saves, NOT auto fix, and I stop and repair before getting a deep pile. Use the Preview Message to help locate the exception (usually there was a mid document close BQ and another Open BQ.
A tool that could count the additional inner Opening BQ tags and only remove the close after counting back down)