Please recommend the best Pdf To epub Converter for mini kobo reader ...

[jackie_w]

If you're using Calibre to convert pdf to epub, another option is to try changing the "line unwrapping factor" for this pdf. You will find it under Calibre's Convert - PDFInput - Line Unwrapping Factor. It is used to try and determine how to reconstitute paragraphs from the pdf's individual lines.

I can't tell you what value to use because the 'best' value depends on the particular pdf. It is a number between zero and one. The default is 0.5, so try 0.6 and/or 0.4 and experiment until you get a better-looking epub.

[ewphoenix123]

Originally Posted by DNSB

Looking at your screen, I would agree with the comment that Higgles made. Those lines might look okay on a system with a wider page width but the Mini is showing that each line in your original is now it's own paragraph. That is where the search I suggested in my earlier post comes in -- locate all the lines ending in a lower case letter and join them to the next line.

In Sigil, I use something like:

Find: [a-z]</p> <p class="para">
Replace: single space character

Then hit find, if you want to replace, hit replace, otherwise hit find again. If you are feeling enthused, you can try hitting replace all or replace all after selecting current file instead of all files to do it one file at a time.

Repeat the find/replace looking for commas followed by the end of paragraph and repeat again for any other characters that you choose.

Regards,
David

This would replace the last letter of every found word aswell, I tend to make that mistake and have to start from scratch. :P
The proper regex for the operation would be something like:

Find: ([a-z])</span></p>\s+ <p><span class="font10">
Replace: \1 (\1single space character)

Find: ([a-z]),</span></p>\s+ <p><span class="font10">
Replace: \1, (\1,single space character)

Find: ([a-z])-</span></p>\s+ <p><span class="font10">
Replace: \1 (\1)


<p><span class="font10"> has to be replaced by the ones used in your file.

[vanminh]

[QUOTE=Higgles;2555208]

Kudos to Higgles for the brilliant diagnosis of the problem !!


I also found that "3DPageFlip PDF to ePub Converter" ( a freeware) solved the [CR] issue : it strips the [CR] in pdf file then converts it to epub .

Everything looks O.K. in my mini Kobo reader now !!

Thanks everyone for comments and advices !

[Higgles]

[QUOTE=vanminh;2555766]

Originally Posted by Higgles

Kudos to Higgles for the brilliant diagnosis of the problem !!

I must admit that I have dealt with similar issues as this when cutting and pasting text into digital signage. Since epubs are html based, the effects of a CR looked the same. At least we have been able to come up with a solution for this. I need to try some of these programs as I've got a load of PDF files that I would like to read on my Kobo. Just need to find which of them run under Linux (my OS).

[DNSB]

Originally Posted by ewphoenix123

This would replace the last letter of every found word aswell, I tend to make that mistake and have to start from scratch. :P
The proper regex for the operation would be something like:

Find: ([a-z])</span></p>\s+ <p><span class="font10">
Replace: \1 (\1single space character)

Find: ([a-z]),</span></p>\s+ <p><span class="font10">
Replace: \1, (\1,single space character)

Find: ([a-z])-</span></p>\s+ <p><span class="font10">
Replace: \1 (\1)


<p><span class="font10"> has to be replaced by the ones used in your file.

Been there and made that mistake often enough. I should have double checked instead of typing by memory.

Regards,
David

[mr_nihilism]

I congratulate those with the patience and the ability to convert PDF to EPUB in a manner that actually is readable because I certainly cannot and it frustrates me to no end. I just don't even bother trying. I see PDF and run the other way.

[vanminh]

[QUOTE=mr_nihilism;2556220]

I know what you mean ... It just a series of trials & errors and going to forums asking & sharing experiences...

Without Higgins' diagnosis, I would end up pulling hairs for weeks ...

Here are a few limited personal experiences I had with pdf To epub conversions :

1) Textbooks ( i.e. Math, Engineering with diagrams & special symbols) , just forget about the conversion . Buy a reader that can read pdf natively (if you do not want to read on a PC or laptop) .

2) Pdf from scanned books or documents ...ditto ....

3) The conversion is O.K. for texts only ( i.e. novels or non-fiction with no pictures )

4) The freeware "3DPageFlip PDF to ePub Converter" seems to produce consistently good conversions when viewed in Mini Kobo reader .

Good lucks...

[Psipherious]

Originally Posted by ewphoenix123

This would replace the last letter of every found word aswell, I tend to make that mistake and have to start from scratch. :P
The proper regex for the operation would be something like:

Find: ([a-z])</span></p>\s+ <p><span class="font10">
Replace: \1 (\1single space character)

Find: ([a-z]),</span></p>\s+ <p><span class="font10">
Replace: \1, (\1,single space character)

Find: ([a-z])-</span></p>\s+ <p><span class="font10">
Replace: \1 (\1)


<p><span class="font10"> has to be replaced by the ones used in your file.

Thanks for this guys, very helpful, I came up with my own regex expression, a little more generic like so which works fantastic for me and I just do a "replace all":

Find: ([a-zA-Z1-9/%:,])</p>\s+<p class="[a-zA-Z1-9]+">
Replace: \1

Just wanted to add to this super helpful discussion.

(My first post too, long time lurker.)

[JSWolf]

The only way to know you have a 100% error free PDF > ePub conversion is to take the ePub and the PDF and do a 100% A/B comparison and fix any found errors.

[DNSB]

Originally Posted by JSWolf

The only way to know you have a 100% error free PDF > ePub conversion is to take the ePub and the PDF and do a 100% A/B comparison and fix any found errors.

I don't ask for a 100% perfect conversion, just one that I can read without my eyeballs starting to bleed.

To quote one classic book: "Through the oily snaggles of hair stared two huge eyeballs that bulged from their sockets, so bloodshot that they appeared more like two baseballs of very lean bacon."

Regards,
David

[Josieb1]

I only convert text PDFs and I use calibre convert to RTF, then I edit the RTF removing headers, trailers, and moving up the text to close the large gaps caused by the header/trailers. I then convert RTF to ePub.

That's the only way I know how to do it, time consuming but I'm very happy with the results.

[DNSB]

Originally Posted by Josieb1

I only convert text PDFs and I use calibre convert to RTF, then I edit the RTF removing headers, trailers, and moving up the text to close the large gaps caused by the header/trailers. I then convert RTF to ePub.

That's the only way I know how to do it, time consuming but I'm very happy with the results.

The method I use is likely as time consuming but I feel it's time worth spending to get a result I will be happy with. Admittedly, at time, I wonder when I've spent more hours cleaning up the ebook that I will spend reading it but it's my time.

Regards,
David

[leftright]

Originally Posted by vanminh

4) The freeware "3DPageFlip PDF to ePub Converter" seems to produce consistently good conversions when viewed in Mini Kobo reader .

No it doesn't. The lite version recommends a upgrade to the pro versin to do a pdf conversion.

The pro version trial gave me constant "contact MS" errors when TRYING to convert a pdf.

I think that the recommendation for that software is nothing more than product placement, you said that the free version does conversions, it does nothing of the sort.

I think that you're a software troll.

[yayo]

What about just installing Koreader and reading the pdf with added reflow?

[Dr. Drib]

That program - 3DPageFlip PDF to ePub Converter - is going to cost $99 or $299, which is absurd - in my opinion.

The software is a free trial software === that means you can download it for free.