Edit Book: How to remove formatting (without deleting)?

[chaot]

formatted characters

I convert a .pdf to .epub. But some words are still formatted, e.g. ''Anhang''.
I can not search for the word by writing ''Anhang'' in the Search(Find)-Box.


''Anhang'' formatted
C&P the formatted ''Anhang'' I can search for it.

Both cases allow to remove with S&R, but the formatted version only in whole code-expression:

<p class="calibre1">Anhang <i class="calibre3"> </i></p>

or

<p class="calibre1">A </p>

Isn't that strange!?

[jhowell]

It looks to me from your first image that the "nhang" characters in that word may actually be unicode small cap glyphs. Move the cursor there and check the bottom right status line in the editor.

[chaot]

Yes, it looks like small cap glyphs, and what you may not be able to recognize clearly: the characters are somewhat blurred/indistinct.


''no name letter''


LATIN SMALL LETTER B (Just to make the difference recognizable, a ''normal'' character.

[pdurrant]

So the answer is that this isn't formatting. Those small caps are actually special Unicode characters.

Wikipedia:

Although small caps are not usually "semantically important", the Unicode standard does define a number of "small capital" characters in the IPA extensions, Phonetic Extensions and Latin Extended-D ranges (0250–02AF, 1D00–1D7F, A720–A7FF). These characters, with official names such as latin letter small capital a, are meant for use in phonetic representations. For example, U+0280 ʀ (HTML ʀ) represents a semi-voiced uvular trill.

As of Unicode 5.1, the only characters missing from the ISO basic Latin alphabet are small-capital versions of Q and X.

In short, someone's done small caps the wrong way. You'll have to search and replace the actual characters.

[chaot]

Continuation!

[chaot]

Another remarkable thing: I can search for Der Neue Tag, (incl. comma & empty space) and find, but not for the date . .  (!!! even here it keeps the formatting (or whatever that is !!!). The date is only to handle per C&P.


Der neue Tag, incl. comma & empty space

In practical handling that creates some problems!

[pdurrant]

If this is the same publication that used Unicode small caps, they're probably also messing with the numbers.

Having check, yes, the person who made that book decided to use a private use area in Unicode (specifically 0xF731-0xF739), that apparently is used by the font in the book for old-style numbers.

That's really messed up and non-standards compliant.

[chaot]

If this is the same publication that used Unicode small caps, they're probably also messing with the numbers.

It's the same publication (book) - and it apparently only affects the numbers (possibly exclusively, I do not know yet).

It's out of a rather new book (my version is Werke – Ausgabe 1989) with 'older' texts (not so much private).

In order to get the problem under control, I exchanged now all numbers  to  with 0 to 9 by S&R.

I have to say, however, that I do not edit this archive.org-version (used only for control purposes), but a PDF converted ePub (origin unknown).

Noticeable is also, that marking these abnomal patterns it looks a little bit like bar code.