epub to epub: removing hyphens without other changes

[Thomas_Georg]

Hi! I need to remove unnecessary hyphens in an epub3, and have tried the heuristic processing option when converting from epub to epub. It seems that Calibre also edits file structure and contents, leaving the resulting epub3 unusable to my needs.

I have tried to tweak the settings to make Calibre edit just the hyphens, but to no avail. Is there a way to just remove the unnecessary hyphens without editing the entire epub?

[JSWolf]

Originally Posted by Thomas_Georg

Hi! I need to remove unnecessary hyphens in an epub3, and have tried the heuristic processing option when converting from epub to epub. It seems that Calibre also edits file structure and contents, leaving the resulting epub3 unusable to my needs.

I have tried to tweak the settings to make Calibre edit just the hyphens, but to no avail. Is there a way to just remove the unnecessary hyphens without editing the entire epub?

The hyphens are not in the eBook.The hyphens are from the program being used to display the eBook. In the CSS, you can turn off hyphenation using the following CSS code. This is the body style I use with the code for turning off hyphenation. This will turn off hyphenation for the entire eBook.

body {
  widows: 1;
  orphans: 1;
  margin-top: 0;
  margin-right: 0;
  margin-bottom: 0;
  margin-left: 0;
  text-align: justify;
  -epub-hyphens: none;
  -webkit-hyphens: none;
  hyphens: none;
}

[Thomas_Georg]

Originally Posted by JSWolf

The hyphens are not in the eBook.The hyphens are from the program being used to display the eBook. In the CSS, you can turn off hyphenation using the following CSS code. This is the body style I use with the code for turning off hyphenation. This will turn off hyphenation for the entire eBook.

Thanks for the tip, but I use these epubs for further conversions into accessible formats for sight impaired. Hyphens in the middle of words exist within the body text, and are kept across different formats, so I am looking for a method to erase these hypens early in the conversion process without deleting every single one manually.

[theducks]

Originally Posted by Thomas_Georg

Thanks for the tip, but I use these epubs for further conversions into accessible formats for sight impaired. Hyphens in the middle of words exist within the body text, and are kept across different formats, so I am looking for a method to erase these hypens early in the conversion process without deleting every single one manually.

So what does the sight impaired person get for phrases that do NEED hyphens, like : double-barreled ? doublebarreled? double barreled?

Seems to me, that program needs to adjust its interpretation based upon context. double<very short new word pause>barreled

I believe the CSS code JUST instructs the render engine (viewer) on how to BREAK words for visual line fit. Those places might not always coded in the book (soft).

[retiredbiker]

Hyphens in the middle of words exist within the body text, and are kept across different formats,

If your book came from an OCR process, it is very possible to have real hyphens scattered through the book, where the print copy had them. Some OCR is smart enough to remove them from the ends of lines, but not all, especially older ones.

If that is your case, you could remove them with search and replace, but like theducks said, what do you do where they are valid, like "forty-two"? If you just have a bad book to start with, your only option is to proofread it and manually remove the bad ones. Very tedious indeed.

One thing you might try--examine them closely in the editor. I've seen the odd book where the end-of-line hyphens came through as "- " (hyphen space), but the valid ones had no space, and I could clean it up by searching out the "- " instances. But never perfect: proofing still needed.

[JSWolf]

Spell check the book. It should find words broken with hyphens.

[BetterRed]

I would use the Editor's Tools->Reports->Words list with a '-' in the filter, it should be easy to spot any questionable hyphenation, and fix in the code editor, viz:



BR

[Phil MTL]

brilliant !