Remove rhombus question marks from newly created epub

[Joseph The Grave]

I created an ebook (epub) from a Project Gutenberg book that was in html format after importing it into calibre but the result has rhombus question marks. Otherwise it came out pretty good. How do I remove these rhombus marks from the newly created epub? If I have to redo it that's okay but I just want an epub free of these marks.

https://i.imgur.com/EafI07i.jpg

[Martinoptic]

I think they may be unknown glyphs in the font you are using to read the epub. You could try embedding a font within your epub that contains the glyph. Chareink6 might do the job amongst others, and it’s available on MR.

https://www.mobileread.com/forums/sh...ight=Chareink6

Or maybe your .opf file in the epub is not encoding to utf8?

"Modify epub" in Calibre will let you do this painlessly.

The first line in mine is: <?xml version="1.0" encoding="utf-8"?>

[theducks]

I concur. Those are missing Glyph symbols.

Either your current font does not have those
or
the book used the wrong character set during conversion (from text) somewhere along the line.

[Joseph The Grave]

Update: I did discover a fix. Instead of just saving the html directly from Firefox's Save Page As, I used an FF extension I've been using for years called SingleFile (which saves the complete page). Once I imported that html/zip file into Calibre then converted it to epub, it created a perfect epub ebook of the entire novel with no strange rhombas marks. Now if only I could figure out how to create OCR html files from those PDF ebooks I got off internet archive to convert into epubs...

[Sarmat89]

Probably the file is encoded in Windows-1252, but the declaration says it is UTF-8.

[Karellen]

Originally Posted by Joseph The Grave

Now if only I could figure out how to create OCR html files from those PDF ebooks I got off internet archive to convert into epubs...

Maybe try this... https://www.mobileread.com/forums/sh...3&postcount=23

I've been able to create epubs relatively quickly using that workflow.
Maybe read the whole thread for context.

[JSWolf]

Originally Posted by Joseph The Grave

I created an ebook (epub) from a Project Gutenberg book that was in html format after importing it into calibre but the result has rhombus question marks. Otherwise it came out pretty good. How do I remove these rhombus marks from the newly created epub? If I have to redo it that's okay but I just want an epub free of these marks.

https://i.imgur.com/EafI07i.jpg

Please attach your image so it can be seen. imgur is over capacity,

[DNSB]

Originally Posted by JSWolf

Please attach your image so it can be seen. imgur is over capacity,

@Joseph The Grave: What does the epub version of the book look like when you download it?

As for the image, here you go, Jon:

[Quoth]

I've been downloading Gutenberg since before I used Calibre, maybe about 25 years. No issues. It must be wrong character set or something. But I never use the HTML files. Started with plain text

[JSWolf]

Originally Posted by DNSB

@Joseph The Grave: What does the epub version of the book look like when you download it?

As for the image, here you go, Jon:

Thank you for the image.

What is the character that's not displaying? Is it supposed to be an em-dash or en-dash? If it is such a charcter, it's quite easy to load it into the calibre editor and perform a search/replace to fix this.

[Joseph The Grave]

I appreciate additional posters here continuing to try to help but did you guys miss my update above where I found a fix? Maybe it's not posting right and can't be seen for some reason. It was the way I was saving the ebook off the Gutenberg site that was the problem. Once I used my SingleFile save web page extension in my FF browser to save and import it into calibre, it worked fine. I now have a perfectly formatted epub ebook with working Table of Contents.

As for the OCR PDF to HTML thing I mentioned above, I think I had unrealistic expectations in what that would entail. I thought you could just use OCR software with no need of a scanner to convert a PDF ebook into HTML to create in Calibre an epub ebook, but I see now that it's much much more involved and intricate than that. Guess I'll just have to stick with those massive 400MB PDF ebooks I got off IA and perhaps find a way to reduce the size.

[DNSB]

The books from IA tend to be scanned images of a page with a very poorly OCRred text page to allow for searching. Unless you are heavily into masochism, I would avoid trying to do anything with them.

And again, since there is an epub version of the book on Gutenberg, why not simply download that instead of saving the web page version and converting to epub? This link should point to Before the Dawn (epub) which I find much easier than messing with conversions.

[Joseph The Grave]

Originally Posted by DNSB

The books from IA tend to be scanned images of a page with a very poorly OCRred text page to allow for searching. Unless you are heavily into masochism, I would avoid trying to do anything with them.

And again, since there is an epub version of the book on Gutenberg, why not simply download that instead of saving the web page version and converting to epub? This link should point to Before the Dawn (epub) which I find much easier than messing with conversions.

Yeah, I'm giving up on the OCR PDF to HTML thing. Too much tedium and pain. I would like to find a way to substantially reduce the size of those PDFs though without too much loss of quality but when I attempt it I notice the new file size is barely smaller than the original so it hardly seems worth it.

I have problems finding stuff on PG with their search engine. If I type in "John Taine" his name won't even come up though I know he has books on there. I also have trouble finding the epub versions there but maybe I didn't look hard enough. It's otherwise been a valuable learning experience doing it the hard way with creating my own epubs so I don't regret it too much. Thanks for the help, everyone. Peace out.

[DNSB]

Originally Posted by Joseph The Grave

I have problems finding stuff on PG with their search engine. If I type in "John Taine" his name won't even come up though I know he has books on there. I also have trouble finding the epub versions there but maybe I didn't look hard enough. It's otherwise been a valuable learning experience doing it the hard way with creating my own epubs so I don't regret it too much. Thanks for the help, everyone. Peace out.

And I thought Gutenberg Canada's search engine was crud. OTOH, Canada is a life+50 country and since John Taine passed away in 1960, his works would still be under copyright in life+70 countries. Likely why his works show up on Gutenberg Canada and not on Gutenberg USA or Gutenberg Australia.

[jmurphy]

Originally Posted by DNSB

OTOH, Canada is a life+50 country

Canada went to life+70 back in 2022.

https://parl.ca/DocumentViewer/en/44...9/royal-assent

--jmurphy