Removing Soft hyphens

[phossler]

https://www.mobileread.com/forums/showthread.php?t=77992

I can see the C2 AD (194 and 173) with my hex editor, but as others have pointed out, they're invisible in Sigil

I tried copy/paste the characters, but nothing worked.

The post above is old and 7.2 is out, so maybe things have changed???

Is there any way to strip them out of Sigil? RegEx maybe?

Paul

[theducks]

Originally Posted by phossler

https://www.mobileread.com/forums/showthread.php?t=77992

I can see the C2 AD (194 and 173) with my hex editor, but as others have pointed out, they're invisible in Sigil

I tried copy/paste the characters, but nothing worked.

The post above is old and 7.2 is out, so maybe things have changed???

Is there any way to strip them out of Sigil? RegEx maybe?

Paul

When you pasted, did you try and escape the character?

[phossler]

When you pasted, did you try and escape the character?

No. I just did and it works.

That's one for my saved searchs

Still, it would be nice (IMHO) if Sigil had a 'Revel Hidden Codes' View option that S&R would work in. Poking around in hex, it looks like there's some more stuff to investigate


Paul

[Doitsu]

BTW, the Calibre Hyphenate This! plug-in can automatically remove all soft hyphens.

[mrmikel]

Originally Posted by phossler

Still, it would be nice (IMHO) if Sigil had a 'Revel Hidden Codes' View option that S&R would work in. Poking around in hex, it looks like there's some more stuff to investigate

Agreed. But it may not be available in webkit. However, perhaps they could build a search function for these types of characters which would allow us to do something with them. Hidden spaces which throw things out of line have been my issue, though mostly with imported html.

[phossler]

@Doitsu -- thanks, but a lot of time Calibre will add a lot of CSS that I don't want. I will follow up. Maybe run the html into Calibre, convert to epub, and then the plug in???

@mrmike -- The biggest problem is that F&R seems to want the character as text. This means that I have to locate and identify the troublesome character, use CharMap to copy it, paste in into a Sigil Find (remembering to escape it -- thanks 'theducks'), etc.

I knew it was 173, so I did try the \ and then alt+numpad 0173 in the Find, but didn't work. But CharMap works if I know what I'm looking for

Now that I have it as saved search it will be easier. I hope that as I find more things like this, I can just keeping addeing them to my 'Delete Bad Char' saved search

I couldn't figure out why spell check had 100+ occurances of just 'ed' and 'ing' flagged in things like 'walked' and 'walking'.


Paul

[Toxaris]

If you know a location where it is in Sigil, you can actually select and copy it. Click on the character next to it and press shift+arrow in the direction you want to select. If you have the right character, the cursor will not move although you pressed the arrow. Copy and paste in the S&R window.

[DiapDealer]

Just use the \x{FFFF} method provided by PCRE Regex to search for unicode code points. Replace FFFF with hexdecimal representation of the unicode code point you're wanting. In this case 00ad (or just ad)

Using regex, search for \x{00AD} (or \x{ad) and replace with nothing to remove soft-hyphens.

[phossler]

@Toxaris and DiapDealer -- thanks !!

Both very useful tips

The \x{00AD} is MUCH easer to see

So if I wanted to follow this theme then could I include even more characters in my SavedSearch:?
[\x{00AD}\x{2000}-\x{200D}]

where 2000 is En-Quad and 200D is Zero Width Joiner (what ever that is)

That would include Thin, Hair, and Zero Width spaces that I think mrmike mentioned

Paul