PDFRead 1.8.2 released!

[Thorkin]

Originally Posted by nrapallo

That signifies to me that there is an error encountered during the creation of the .prc due to a number of reasons, but most likely linked to memory issues or .pdf encoding. See if you can create a .prc from the resulting .html and images using Mobipocket Creator or Calibre.

Try a range of pages, like 1 to 10, and see if that works using PDFRead's internal .prc creation.

Also, try converting the .djvu file instead. I tested it with both and there were some missing pages when I used the .pdf file (some of the first 10 pages didn't come through).

Hrm, I can get it to work with some other pyle ebooks, and I can get Creator to make ebooks out of the .pngs, but otherwise the error remains. It seems to be something related to filenames as changing the creation filename will help sometimes.

Thanks for your help though!

[malife]

Hello All,
Nick, awesome job with PDFRead absolutley loved it. Great tool. I too can confirm the "blank page" bug in Kindle 2 mentioned by AnthonyPaulO. If I generate the html and then produce the mobi either with mobipocket (in Windows) or Calibre in Mac the generated book renders correctly and no blank pages are inserted.

I have a question. Say I want to read a PDF that is mostly two columns, i.e. a scientific paper. Is there a way for me to split that document column wise even if I loose some content (like the title) that is not dobule column? For instance say I have a PDF document like the one shown in the attached sample.png, and want to split it in columns like the attached col1.png and col2.png. Can PDFRead do that? Anyone knows the unpaper setting that would make the trick?

Thanks!

[nrapallo]

Originally Posted by malife

Hello All,
Nick, awesome job with PDFRead absolutley loved it. Great tool. I too can confirm the "blank page" bug in Kindle 2 mentioned by AnthonyPaulO. If I generate the html and then produce the mobi either with mobipocket (in Windows) or Calibre in Mac the generated book renders correctly and no blank pages are inserted.

That's very interesting. Could you upload a sample .prc (and .html) of the 'with blank pages (PDFRead created)' and 'without (Mobipocket or Calibre created)' ebooks. I'll investigate it further to determine what can be done to stop that!

I have a question. Say I want to read a PDF that is mostly two columns, i.e. a scientific paper. Is there a way for me to split that document column wise even if I loose some content (like the title) that is not dobule column? For instance say I have a PDF document like the one shown in the attached sample.png, and want to split it in columns like the attached col1.png and col2.png. Can PDFRead do that? Anyone knows the unpaper setting that would make the trick?

Thanks!

PDFRead can split the page like that with the Layout mode: 'portrait-2col' (with four quadrants/pages). However, it won't detect the two columns boundaries, just split down the middle. Not too helpful, most of the time. See this post for more info.

However, for that functionality try PaperCrop - Multi-column PDF files on 6 inch display. It does this better!

[malife]

Here they are Nick. The three formats from PDFRead, Mobipocket Creator and Calibre. I confirm again the PDFRead version has blank pages every other page, at least it displays so in the Kindle 2. Interesting how the Calibre file is noticeably larger (almost 3 times) than the other two.

Thank you so much for the tip on the PaperCrop app. Really useful specially for papers. Thanks!

[nrapallo]

Originally Posted by malife

Here they are Nick. The three formats from PDFRead, Mobipocket Creator and Calibre. I confirm again the PDFRead version has blank pages every other page, at least it displays so in the Kindle 2. Interesting how the Calibre file is noticeably larger (almost 3 times) than the other two.

Thank you so much for the tip on the PaperCrop app. Really useful specially for papers. Thanks!

Thanks for these!

Nothing apparent that would cause that blank page, other than the NRhtml2mobi.exe I use.

Attached is the explosion of those .prc using my primitive .prc to .html utility called makedocN. I can't find anything that would trigger those blank pages.

More investigation needed...

[nrapallo]

Originally Posted by malife

Here they are Nick. The three formats from PDFRead, Mobipocket Creator and Calibre. I confirm again the PDFRead version has blank pages every other page, at least it displays so in the Kindle 2. Interesting how the Calibre file is noticeably larger (almost 3 times) than the other two.

Thank you so much for the tip on the PaperCrop app. Really useful specially for papers. Thanks!

malife:

OK, just to rule out the old v0.0.37 Mobiperl I previously used to create NRhtml2mobi.exe, I just updated it to the latest Mobiperl v0.0.41.

Please try again using the attached updated NRhtml2mobi.exe (up to v0.0.41) by replacing the one in the 'bin' folder of the PDFRead install directory.

Thanks!

[malife]

Sorry but no Joy. Still blank page every other page

Unfortunately I won't be able to continue reporting the Kindle 2 issues. I've decided I will return it and get a Sony Reader. There are many issues I dislike in the Kindle, being lack of PDF support the most important of them. I will still be using PDFRead though for papers and complicated PDFs.

As soon as I get my Sony Reader I will report back if the "blank page" bug is also present there, though I doubt it.

Thanks for a superb free tool!

[nrapallo]

Originally Posted by malife

Sorry but no Joy. Still blank page every other page

Unfortunately I won't be able to continue reporting the Kindle 2 issues. I've decided I will return it and get a Sony Reader. There are many issues I dislike in the Kindle, being lack of PDF support the most important of them. I will still be using PDFRead though for papers and complicated PDFs.

As soon as I get my Sony Reader I will report back if the "blank page" bug is also present there, though I doubt it.

Thanks for a superb free tool!

Thanks for all your effort in helping PDFRead get better! Much appreciated!!

[DaleDe]

Typically blank pages are caused by the image being slightly longer than the page itself. The page view gets cropped to the length of the page and then rest of the page shows up on the next page (which appears blank) but is actually the bottom area of the previous page. This is likely due to some extra info on the bottom of the page that causes the area of the page to be smaller than you thought.

Dale

[Student1]

Hi guys,

starting some massive conversion for pdf to lrf and wondering if these settings are the best that can be done to get good quality and more important as close as possible to full page on my prs 505. Here are the settings i ve been using (using command line as i batch it all for 300+ books so this takes weeks on 2 computers)

set LOC=C:\Program Files (x86)\PDFRead
set OPT=-p prs505-p -i pdf -c "Mythology" -f "lrf" -m "portrait-full" -r "none" --no-dilate --dpi "600" --colorspace "rgb" --colors "256" --optimize

let me know what you guys think and if i should change something, decided to keep color as new reader will be coming with color.

thanks!

[nrapallo]

Originally Posted by Student1

Hi guys,

starting some massive conversion for pdf to lrf and wondering if these settings are the best that can be done to get good quality and more important as close as possible to full page on my prs 505. Here are the settings i ve been using (using command line as i batch it all for 300+ books so this takes weeks on 2 computers)

set LOC=C:\Program Files (x86)\PDFRead
set OPT=-p prs505-p -i pdf -c "Mythology" -f "lrf" -m "portrait-full" -r "none" --no-dilate --dpi "600" --colorspace "rgb" --colors "256" --optimize

let me know what you guys think and if i should change something, decided to keep color as new reader will be coming with color.

thanks!

I think you would benefit from the dilation (i.e. don't use --no-dilate) that PDFRead does especially if you get the update in post #1 of this thread in the file 'pdfread-MinFilter5-mod-bin.zip' and use 600 DPI.

The dilation bolds the text so that any reduction in size from it's original dimensions, say, 3000x4000 down to 520x640 can still retain some thickness of the original text.

Just experiment on a small page range like --first-page "31" --last-page "31" to see which setup you prefer.

The 'optimize .PNGs' will help with the resulting filesize without too much of a processing hit.

See these sample conversions of page 31 of a Archive.org (merryadventureso00pyle2.pdf) scanned book.

[Student1]

Originally Posted by nrapallo

I think you would benefit from the dilation (i.e. don't use --no-dilate) that PDFRead does especially if you get the update in post #1 of this thread in the file 'pdfread-MinFilter5-mod-bin.zip' and use 600 DPI.

The dilation bolds the text so that any reduction in size from it's original dimensions, say, 3000x4000 down to 520x640 can still retain some thickness of the original text.

Just experiment on a small page range like --first-page "31" --last-page "31" to see which setup you prefer.

The 'optimize .PNGs' will help with the resulting filesize without too much of a processing hit.

See these sample conversions of page 31 of a Archive.org (merryadventureso00pyle2.pdf) scanned book.

Wow huge difference! You just saved me a week in processing power! Well i have to restart but atleast the result will be a lot better! Also i noticed the images did not take the whole page and most of the time i had to zoom once or twice on my device, is there an option so the images will fit the screen without loosing quality (ie stretching)? Again while using the same options without the no dilate this time, also picked up the update, missed the the first time!!

[nrapallo]

Originally Posted by Student1

Wow huge difference! You just saved me a week in processing power! Well i have to restart but atleast the result will be a lot better! Also i noticed the images did not take the whole page and most of the time i had to zoom once or twice on my device, is there an option so the images will fit the screen without loosing quality (ie stretching)? Again while using the same options without the no dilate this time, also picked up the update, missed the the first time!!

If the border is truly 'white', then it can be stripped off, but in my sample above (see attahced original page as a .jpg) , there is a black outlined box around the page so, in essence, there is 'no white border'.

A smug, tear, or dark spot may also thwart the stripping of the border as most definitely will a page number, header text or watermark. In these cases, the 'unpaper' option may serve you well with the 'pre-border' option mentioned a few pages back in this thread.

[Student1]

Yes thanks, i read about unpape, probably wont need it as my scans are very clean, well most of them anyway! I just started another batch on 2 computers... about 400 books now in all to process! Guess it will be a while !

Thanks again for your help! Great programm, thanks for the dedication!!!

I wonder are there any plans to output to epub? Seems more and more the format that might come on top!

[nrapallo]

Originally Posted by Student1

Yes thanks, i read about unpape, probably wont need it as my scans are very clean, well most of them anyway! I just started another batch on 2 computers... about 400 books now in all to process! Guess it will be a while !

Thanks again for your help! Great programm, thanks for the dedication!!!

I wonder are there any plans to output to epub? Seems more and more the format that might come on top!

Epub output is getting closer to a reality, but since the original author of PDFRead was going to add it in his v2.0 release of the program, I was holding off.

I may just go ahead and beat him to it, though. However, I will need to brush up on my python programming skills first.