PDFRead 1.8.2 released!

[carld]

Originally Posted by nrapallo

Thanks, Carl! I haven't had too much work to do lately (improving PDFRead) as the software has matured very nicely indeed!

For the WOWIO books you convert, you may be interested in trying the .html output option as this leaves the .html used to create the .prc along with it's (page) images. Then you can delete/edit out the "ad" images and manually convert it, if you would like.

This can also be accomplished by creating an empty file called 'debug' in the main PDFRead install directory.

Thanks for the tip, I'll give that a try. For the first book, I just started the conversion on page 3 to bypass the first couple of advertising pages. I'll see how well going to html first does.

I did try it on another document, a game guide with lots of text per page. It converted well, but the resulting type was too small for my tired old eyes to read without effort.

Carl

[nrapallo]

Originally Posted by carld

Thanks for the tip, I'll give that a try. For the first book, I just started the conversion on page 3 to bypass the first couple of advertising pages. I'll see how well going to html first does.

Try using the 'debug' empty file method and choosing .prc as an output format. This way the .opf file will be left behind, which then can easily be loaded into Mobipocket Creator. There you can can re-build it (without those ad pages ).

I did try it on another document, a game guide with lots of text per page. It converted well, but the resulting type was too small for my tired old eyes to read without effort.

Carl

Make sure you choose a landscape Profile (ones without the '-p' suffix) as this will magnify the text a bit. If you're using portrait mode, then the text will usually come out too small to read comfortably. Experimentation is the key!

[jossos]

Hey,

Is there any way to do batch conversions?

Thanks.

[nrapallo]

Originally Posted by jossos

Hey,

Is there any way to do batch conversions?

Thanks.

I have not done any batch processing because the Title, Author (and Category to a lesser degree) need to be specified for each .pdf converted and it's not easy to automatically obtain that info from the source file or filename.

That said, there are ways to do batch processing using the GUI as a starting point. Another software package I've written, Mobi2IMP, uses the below approach and it works well primarily because no Title or Author needs to be entered. See Mobi2IMP method of batch processing all files in the directory and recursivey down into its subdirectories.

The only real issue/problem doing this in PDFRead is that the Title and Author need to be entered separately. See PDFRead method of batch processing for a previous version which can be adapted for your purpose.

Between the two methods, you can probably get batch processing working, but since the setup time is lengthy, it is hard to automate this.

Hope this helps.

[jossos]

Thanks. I figured it'd be a hard feature to implement.
I'm going to give this a try!

[dedo]

I'm trying to create a .prc for iLiad from a .pdf, but at the end of the conversion PDFRead just opens a temporary folder containing images and .html, no .prc is generated.

[nrapallo]

Originally Posted by dedo

I'm trying to create a .prc for iLiad from a .pdf, but at the end of the conversion PDFRead just opens a temporary folder containing images and .html, no .prc is generated.

If you double-click that .html, does it show the ebook in your browser?

If so, then you did convert the ebook and the output is in the same directory that your input file is located in. Otherwise, ensure your output format is '.prc' since this behaviour is also typical of the .html output format.

If not, then the .pdf source file has 'issues' being rendered into images i.e. protected .pdf, encrypted .pdf or corrupt .pdf.

Try it with another .pdf to see if the software works "as advertised"

[dedo]

Originally Posted by nrapallo

If you double-click that .html, does it show the ebook in your browser?

If so, then you did convert the ebook and the output is in the same directory that your input file is located in. Otherwise, ensure your output format is '.prc' since this behaviour is also typical of the .html output format.

If not, then the .pdf source file has 'issues' being rendered into images i.e. protected .pdf, encrypted .pdf or corrupt .pdf.

Try it with another .pdf to see if the software works "as advertised"

The html created is fine and images too, but they are in a temporary directory and no .prc is created. With another .pdf it worked fine.

[nrapallo]

Originally Posted by dedo

The html created is fine and images too, but they are in a temporary directory and no .prc is created. With another .pdf it worked fine.

If you don't use the empty file 'debug' to retain the temp directory, then the 'popping up' of that temp directory at the end most definitely signifies the .html to .prc creation has encountered a problem. The build of the .prc was aborted due to some error/inconsistency with that file.

Try completing the .prc build using Mobipocket Creator or mobigen using the .opf file (if it was created by PDFRead in the source directory). Internally, I use a modified version of html2mobi written by tompe to convert the .html/images to .prc.

Hope this helps.

[fmeneuhe]

Hi,
I have been testing PDFRead 1.8 and worked succesfull except I could not find the next feature: when I convert a PDF file to sony LRF, the font size is too small. Is there any functionality to choose the destination font size?
Thanks

[quantamix]

Thank you Nick. I'm so grateful for your software! You made it possible for me to keep my kindle because I was ready to throw it in the garbage after disappointing experience with amazon pdf conversion service and mobireaderpublisher results. The converted ebooks are so small and so light that it's not readable at all.

For PDF ebooks from scanned books.
Here's my settings for my Kindle:
Iliad profile
Layout: Landscape -third
Unpaper: -v
Rotation: Right
Dilation DPI:250
Cropping%:2.0
Colorspace:Gray
Color:4
EdgeLevel: 5
Size: H=520 V=640
Optimize PNGs

Any suggestions? on how to improve the output? I wish I can read the scanned pdf portrait with the text and graphics still big enough to read.

[nrapallo]

Originally Posted by fmeneuhe

Hi,
I have been testing PDFRead 1.8 and worked succesfull except I could not find the next feature: when I convert a PDF file to sony LRF, the font size is too small. Is there any functionality to choose the destination font size?
Thanks

No, you cannot 'choose' the final font size; it is a product of the cropping and rotating (for landscape modes) conversion steps.

PDFRead renders each .pdf page into an image and then tries to 'expand' the image to its maximum width (without white borders). This allows the font size to 'appear' larger but it cannot be changed unless that page is 'split' over several images.

Layout mode 'Landscape-third' potentially offers the most magnifaction of the font, but requires three page turns for each pdf page. Whereas layout mode 'Portrait-full' only requires one page turn, but with a smaller font! Usually somewhere in between is a good trade-off.

[nrapallo]

Originally Posted by quantamix

Thank you Nick. I'm so grateful for your software! You made it possible for me to keep my kindle because I was ready to throw it in the garbage after disappointing experience with amazon pdf conversion service and mobireaderpublisher results. The converted ebooks are so small and so light that it's not readable at all.

For PDF ebooks from scanned books.
Here's my settings for my Kindle:
Iliad profile
Layout: Landscape -third
Unpaper: -v
Rotation: Right
Dilation DPI:250
Cropping%:2.0
Colorspace:Gray
Color:4
EdgeLevel: 5
Size: H=520 V=640
Optimize PNGs

Any suggestions? on how to improve the output? I wish I can read the scanned pdf portrait with the text and graphics still big enough to read.

Thanks!

I usually resort to 'Landscape-half' as my preferred layout mode, but if I can get it to work in 'Portrait-full', then I go that route.

If by "improve the output" you mean file size, then for sure using the gray colorspace and reduced number of colors helps keep the file size down. And so should 'Optimized .PNGs'.

If you mean image legibility, then increase the edge level by one or two for sharper text, or increase the DPI for less dilation/blurring of the text or a combination thereof.

Experiment but remember when doing so to limited the page range 'Pages: __ to: __' or you could be wasting a lot of time when things don't work out!

[quantamix]

I appreciate everything you shared with us. Thank you!

What I meant about output is actually image legibility which you answered. Thank you.

>If you mean image legibility, then increase the edge level by one or two for >sharper text, or increase the DPI for less dilation/blurring of the text or a >combination thereof.

I will try.

[nrapallo]

Recently, I was trying to convert some manga .cbz and PDFRead failed (miserably). The .cbz was just a .zip file with .jpg images, but it was done so as to break the simplistic .cbz module (I wrote) within PDFRead.

Originally the .cbr/.cbz files I tried when writing the code worked for me.

The sample .cbz I used downloaded as a .zip and when renamed to .cbz would not convert. The reason being that a directory was used within to hold all the images and to make matters worse, another directory was also present with links to .jpg images (MACOSX remmanents!).

I found a simple workaround to allow PDFRead to convert the .cbz, namely:

1. unzip the .zip (.cbz) and search for the directory with the main pictures.
2. start PDFRead and from the Input box, browse to that main pictures directory above and double-click any image therein.
3. select the 'imgdir' In Format and insert the title, author, and category.
4. select your preferred profile and layout mode and click the Convert button.
5. your resulting file will be in that main pictures directory.

Below I provide some sample conversations for the EBW1150 done using this workaround method. Please use this method until PDFRead can be improved to better handle the ever increasing use and differing electronic storage of manga/comic books.

Thanks to Nekokami for pointing me in this direction...

p.s. sorry but I forgot to change the ebook author description for the landscape versions; my bad!