Why is the pdf made by converting with Calibre no longer searchable and the text "copied" from it gibberish?

Question

I'm trying to use Calibre to convert mobi and epub formats to pdf and it is working for the purposes of reading, but there are problems.

Namely, the character encoding seems messed up. Searching the text never matches anything, even thought the searched text is "in" the document. Also the text copied from the document looks like this:

ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@ˆ@

Actually when I copied it in here it was "invisible" characters, but the above is what it looks like when copied into VIM.

How can I make Calibre not have these issues when converting to PDF?

Answer 1

As I found in this link, method 1:

1) Convert with calibre to HTMLZ.

2) Replace the .htmlz extension of the file with .zip

3) Unzip the file

4) Create the PDF with Adobe Acrobat Pro using "Create from webpage"

Method 2:

If you are looking for a fast solution and don't mind sacrificing the internal links, simply open the Calibre-produced PDF in Mac OS Preview, go to the Print dialog, and from there "Save as PDF". This will produce a PDF that is searchable, with copyable content, but which lacks the internal links.

There are more solutions in this link, but I do not figure out them myself.