Convert a 2 column PDF into epub

[thorm42]

Hello,

I use Calibre and it is great, but I don't see how it is possible to convert a pdf with 2 column text into a epub with 1 column.

It is possible?


Thanks

[Worldwalker]

As I understand it (and I am not a dev) It's not possible yet, due to the limitations of the third-party PDF-handling library calibre uses. However, there is hope that may be added in the future. I certainly hope they do it!

[jackie_w]

There is another approach you could take with multi-col PDFs. You might like to try it.

If you head over to the PDF forum there is a utility called Briss (in the Sticky section).

It will allow you to convert PDF to PDF, where the output PDF is cropped to a more appropriate size for your reader. There is a GUI so you can control the cropping size.

[kovidgoyal]

Actually code to handle multicol layouts is already in calibre's new PDF engine and (mostly) works. Its just waiting for me to finish it.

[Worldwalker]

Kovid, have I told you lately that I love you? In a totally straight and platonic way, of course.

[AlbertaCowboy]

Originally Posted by kovidgoyal

Actually code to handle multicol layouts is already in calibre's new PDF engine and (mostly) works. Its just waiting for me to finish it.

I've said it before, and I'll say it again. You and Calibre are the sole reasons that eReaders work the way they should. Thanks so much for all your hard work!

[SkyDream]

Currently i'm using "stanza desktop" to convert 2-column pdf to epub.
Of course until calibre offers the same feature...

[thorm42]

Originally Posted by kovidgoyal

Actually code to handle multicol layouts is already in calibre's new PDF engine and (mostly) works. Its just waiting for me to finish it.

Thank you, it is a good news

[paolog]

Originally Posted by kovidgoyal

Actually code to handle multicol layouts is already in calibre's new PDF engine and (mostly) works. Its just waiting for me to finish it.

I bought 2 dictionaries (italian-latin) with the aim to use them on my pocketbook ereader

Beeing them pdf, each time I do a search it tooks minutes (sometimes I give up ...).

So I tried to convert them from pdf to epub but the 2 columns layout produce a mixed text of words and their definitions.

There is some parameter to set so calibre can handle this?

thanks a lot

paolo

[itimpi]

No - Calibre cannot handle 2 column PDF files.

In addition dictionaries tend to fail conversion regardless of the format. I think it is something to do with the number of cross-reference links that tend to be present in such items.

[cybmole]

Originally Posted by thorm42

Hello,

I use Calibre and it is great, but I don't see how it is possible to convert a pdf with 2 column text into a epub with 1 column.

It is possible?


Thanks

open pdf. select all - copy, paste into MSword, save as filtered HTML
that has removed the 2 column effect.
now add that word file to calibre & convert as needed.

if select+ copy all is prohibited you need to find some PDF unlock software to bypass.

[paolog]

@cybmole already tried, didn't worked as produced a pattern like this

1st row: word1 truncated def of word 1 word 2 truncated def of word2
2nd row: tail of def of word1 tail of def of word 2

@ itimpi thanks

[solarcat]

Google found this thread when I was looking for a solution to the same problem. After trying the recommendations here with little success, I found a freeware program that converts directly to epub with better results than the others. Here's the link - PDFMate Free PDF Converter.

Two things to be aware of before installing this. One, by default it will modify all your browsers to install a toolbar, default search, and other crapware. Prevent this by selecting Custom install and un-checking all the boxes. Two, it nags you to upgrade to the paid version every time you convert a file. If you can live with this, try it.

Disclaimer: I am not associated in any way, shape, or form with PDFMate, either directly or indirectly. It gave decent results converting a two-column PDF to epub. That's all.

Two more points from my experiments. One, about cybmole's technique of pasting into Word: in Adobe Reader, "Copy / Select All" gives poor results, while "Copy File to Clipboard" gives good ones. Two, when I tried using Calibre's new PDF conversion engine from the command line, I got a bunch of Python error messages.

[cybmole]

[QUOTE..about cybmole's technique of pasting into Word: in Adobe Reader, "Copy / Select All" gives poor results, while "Copy File to Clipboard" gives good ones./QUOTE]

surely those 2 methods should give identical results?

"copy" in windows-speak means put stuff on clipboard ???

but I ain't that interested in digging back into this ... just saying is all...

[solarcat]

Originally Posted by cybmole

surely those 2 methods should give identical results?

"copy" in windows-speak means put stuff on clipboard ???

but I ain't that interested in digging back into this ... just saying is all...

You'd think so, wouldn't you? I'm just reporting what happened.

The obvious implication is that Adobe Reader distinguishes between those commands in some cases. What I observed was that with "Copy / Select All" it retained the two-column format when pasted into Word. With "Copy File to Clipboard" it did not. It very nicely made the whole document into one column.