PDF images upside down - a solution

[Fingerstoo]

I am a new user of Calibre and have been frustrated with the bug in conversions from PDF originals. The embedded images and logos are incorrectly rotated or inverted. The developer says it is too difficult because the images are inverted in the PDF files and there is an embedded rotation command which he has not been able to implement in the conversion engine.

I think I may have found a simple way around it. I have already installed on my Windows 7 PC the freeware CutePDF that will turn any text, picture or document into a PDF file. It appears as another printer on your PC. Instead of printing your program's output on a real printer, you simply choose the CutePDF "printer" and it will write a PDF file of the output. This works on Word and many other programs, including graphics editors.

I viewed an original PDF file (text and embedded diagrams/pictures/logos) in Adobe Reader X. Printing has to be allowed on the original PDF for this to work. Just print the whole of the document using the CutePDF "printer". Then use this newly created PDF file as the input to Calibre. Hey presto, the images are all the right way up! No editing or fancy conversions to HTML etc. Quick and easy!!!!

Now whether this works on all PDF files and all the formats you can convert to in Calibre, I don't know, but if anyone has been struggling with this, it is worth a try.

[alansplace]

....and another solution to add to it.

I use PDFCreator as a printer driver that outputs a PDF. Though I do not have the same problem as you I tried a variation of your solution using PDFCreator instead. The problem it appears to have solved is one of PDF conversion using Calibre losing the second of a double letter sequence and instead having a space in place of the second of the two letter sequence. Like 'all' being converted to 'al '.

Creating a new PDF first (using PDFCreator) then using Calibre to convert the new PDF to an ePUB seems to have solved the double letter problem.

Hooray! And thatks for the nudge.

[Fingerstoo]

Originally Posted by alansplace

....and another solution to add to it.

I use PDFCreator as a printer driver that outputs a PDF. Though I do not have the same problem as you I tried a variation of your solution using PDFCreator instead. The problem it appears to have solved is one of PDF conversion using Calibre losing the second of a double letter sequence and instead having a space in place of the second of the two letter sequence. Like 'all' being converted to 'al '.

Creating a new PDF first (using PDFCreator) then using Calibre to convert the new PDF to an ePUB seems to have solved the double letter problem.

Hooray! And thatks for the nudge.

Thanks for that! You have me worried now. Do I have the "double letter" bug?

[theducks]

Originally Posted by Fingerstoo

Thanks for that! You have me worried now. Do I have the "double letter" bug?

No such thing.
Those are called Ligatures (a single character that represents 2 or 3 small letters together)and they are known to misbehave on conversion.
ffi ffl ll

[alansplace]

Originally Posted by theducks

No such thing.
Those are called Ligatures (a single character that represents 2 or 3 small letters together)and they are known to misbehave on conversion.
ffi ffl ll

I figured "why call them ligatures when Calibre conversion with 'keep ligatures' set doesn't fix it", I thought it would be less confusing. I guess maybe not!!