My prs-350 does not go to chapter start in oreilly epubs

[rkomar]

Originally Posted by bluerfoot

I did speak with oreilly and they will look at the issue, I actually have a bit of faith that if it is something they can fix that they will. I see I have a 21 say return policy with dell so I will hang on a bit longer.

once you try touch clicking is a crutch...

I was going to try hacking an epub by hand today to see if I could make it work (by adding the additional TOC entries). If it does work, it wouldn't take long to write a python script to do it automatically. I'll report on my results later in the day.

[rkomar]

Originally Posted by rkomar

I was going to try hacking an epub by hand today to see if I could make it work (by adding the additional TOC entries). If it does work, it wouldn't take long to write a python script to do it automatically. I'll report on my results later in the day.

I did a quick test on one chapter, and my idea of inserting an extra sub-item to point to the start of the chapter or section does work on my PRS-505. It will take me a couple of days to figure out how to do it in python, but a script to automate it should be available sometime soon. Even if O'Reilly come up with something better, this could be useful for unencrypted epubs from other sources. So, bluerfoot, hang onto that 650.

[bluerfoot]

Originally Posted by rkomar

I did a quick test on one chapter, and my idea of inserting an extra sub-item to point to the start of the chapter or section does work on my PRS-505. It will take me a couple of days to figure out how to do it in python, but a script to automate it should be available sometime soon. Even if O'Reilly come up with something better, this could be useful for unencrypted epubs from other sources. So, bluerfoot, hang onto that 650.

That is nice news. I have an oreilly python book perhaps if the toc worked I would be writing the script. Seriously thanks I really want the portability and touch of the prs-350

[bluerfoot]

Insult to injury, I did not get a case due to possibly returning it, and I just put a massive scratch across both the silver paint and the screen.

[rkomar]

Originally Posted by bluerfoot

Insult to injury, I did not get a case due to possibly returning it, and I just put a massive scratch across both the silver paint and the screen.

Oh well, I guess it's a keeper, then. On the plus side, I got that script for 'Sony'fying the TOC of epubs done. I've really only tested it on O'Reilly epubs, but it worked on all 14 of mine. I've also only tested it on my Linux system, perhaps it needs some tweaking for use on a Windows machine. Anyways, I run it like this:

python sonyfy.py Erlang_Programming.epub

and it produces a file called Erlang_Programming__PRS.epub with a new TOC.

Hmm, the attachment software doesn't accept .py files, so I renamed it to sonyfy.txt. Just change it back to .py after you download it.

[bluerfoot]

For sure be trying this today, thanks for helping me make lemonade ...

[bluerfoot]

It works perfectly, boy on a separate note I really need to learn python, I started learning perl but perhaps that is not the way to go...one last question...would it be possible to
a) make it open the table of contents expanded? as it is now it opens up just book title and you then have to open the book. kind of silly since you are already reading that book its like duh...why should I have to click it again?

Anyway great thanks...

[rkomar]

Originally Posted by bluerfoot

It works perfectly, boy on a separate note I really need to learn python, I started learning perl but perhaps that is not the way to go...one last question...would it be possible to
a) make it open the table of contents expanded? as it is now it opens up just book title and you then have to open the book. kind of silly since you are already reading that book its like duh...why should I have to click it again?

Anyway great thanks...

I used to use perl for some heavy-duty text processing, and it's very good for that kind of work. Python, however, seems cleaner as a language, and I find it much easier to understand the code I wrote when I look at it again six months down the road.

As for your request, how about I tell you how to go about it, and you use it as an exercise in learning python? The extra button push doesn't bother me enough to make me want to put in the extra effort . The place to add your code is in the method 'sonyfyTOC', just before the loop that starts inserting additional items. You want to check if the first 'navPoint' element has any other 'navPoint' siblings, and if it doesn't, you want to move all of its 'navPoint' children (and the '#text' nodes around them) up to be its siblings. To move each one in turn, use 'removeChild()' based on its current navPoint parent, and then 'insertChild()' based on the navMap parent. That probably sounds confusing, but if you unzip the original epub file and look at the 'OEBPS/toc.ncx' file, you'll see how the XML structure is laid out and it'll make more sense. The indenting will be wrong in the new TOC file, but that won't affect how it works in the e-reader. Go to http://docs.python.org/library/xml.dom.html and look at the 'Node Objects' section to see how to manipulate the XML elements. I expect this would take about ten to twenty lines of code. (It's funny that the net result of all this will be to move one '</ncx:navPoint>' line in the file )

[bluerfoot]

Thanks, I will look at it. Btw when I say a little perl I mean I have never programmed before and I am at chapter 5 in oreilly's intro (learning perl) book. So this will be quite a challenge to me

I do plan on fully reviewing your script and documenting it as a start.