Adventures in discovering the K4PC PID.

[soalla]

first of all I must thank some_updates for all his hard work on this!!

I tried the pack of scripts, first using the old version of convert2xml and getting similar errors to those gotten by Stewball, and then using the new version of the script (created with help from Stew - thanks, man!! ) with success, creating the xml folder. Then I used the genhtml script and got a nice html file with some minor errors as when starting a chapter with an image or due (I believe) to some typos on the original topaz file and a not so small error as the original italics are gone!
but it's an excellent effort, nonetheless and I have now a perfectly readable file – I'll try to use it later on Calibre and convert it to epub for my prs-505

also, I got this errors when converting from xml to html:

Processing …
metadata0000.dat
other0000.dat
page0000.dat
Unknown region type synth_fcvr.center
Warning: skipping this region
Unknown region type synth_fcvr.center
Warning: skipping this region
Unknown region type synth_fcvr.center
Warning: skipping this region
page0001.dat
page0002.dat
page0003.dat

anyone got it and know what it means?

[stewball]

Originally Posted by soalla

first of all I must thank some_updates for all his hard work on this!!

I tried the pack of scripts, first using the old version of convert2xml and getting similar errors to those gotten by Stewball, and then using the new version of the script (created with help from Stew - thanks, man!! ) with success, creating the xml folder. Then I used the genhtml script and got a nice html file with some minor errors as when starting a chapter with an image or due (I believe) to some typos on the original topaz file and a not so small error as the original italics are gone!
but it's an excellent effort, nonetheless and I have now a perfectly readable file – I'll try to use it later on Calibre and convert it to epub for my prs-505

also, I got this errors when converting from xml to html:



anyone got it and know what it means?

Soalla,
I get the same thing. I was wondering if came from the empty pages at the beginning of the book, or open space. Doesn't seem to affect the conversion though.


Stew

[Liviu_5]

I guess we can officially declare Topaz as done, so no major/current ebook drm is standing so far; it is amazing btw to look at the structure of the Topaz files as dissected with the scripts (or in the svg image) to see how Amazon just took the book and scanned it, ran an ocr and some corrections/images and put together a huge slow package and called it a salable ebook...

well at least we should be happy the books are available as e's and from now on they can be read anywhere

[soalla]

Originally Posted by soalla

(...)

also, I got this errors when converting from xml to html:

anyone got it and know what it means?

well, that was solved with the last version (v1.3), and also the errors at the start of chapters when you have an image as the first letter...

now it's time for cleaning the html (reaplying the lost bold and italics formatting) and converting to epub

a very nice job was done (and I believe they're going to improve ir still a little more) by a great group of people, so here is again a BIG THANKS for all of them!!

[Terisa de morgan]

Originally Posted by delphidb96

Care to share it? I"d love one.

Derek

The bat to decrypt? I've only followed the instructions calling to the three scripts? If you want I post it here but it's very simple (I'm very lazy so I always make scripts )

Only one comment. If you try to convert a Topaz book downloaded to Kindle for PC you don't need PID.

[Terisa de morgan]

Originally Posted by soalla

well, that was solved with the last version (v1.3), and also the errors at the start of chapters when you have an image as the first letter...

How do you know the version?

Originally Posted by soalla

now it's time for cleaning the html (reaplying the lost bold and italics formatting) and converting to epub

I don't understand, do you have to look all the books?

[Pardoz]

Originally Posted by Terisa de morgan

I don't understand, do you have to look all the books?

Yes - the conversion process loses italics, boldface, em-dashes, etc. Of course you can also manually correct any OCR errors in the text while you're at it.

[Terisa de morgan]

Originally Posted by Pardoz

Yes - the conversion process loses italics, boldface, em-dashes, etc. Of course you can also manually correct any OCR errors in the text while you're at it.

And I've have begun with a book full of italics

[phenomshel]

Actually, version 1.3 kept dashes in the book I converted. But that's why they say to generate the svg version of the book, for proofreading purposes.

[stewball]

I have been using a free program called Sigil that can edit html and epub files. It can also create epub files after making corrections to the html. You can add in chapter breaks, correct typos, etc.

One caveat, it sometimes stops responding and can be slow to use. A new version is due out soon.

My first two topaz have been pretty good in quality, and definitely better than some scans I have seen.


Stew

[clarknova]

Originally Posted by phenomshel

Actually, version 1.3 kept dashes in the book I converted. But that's why they say to generate the svg version of the book, for proofreading purposes.

The dashes should be there, but the Topaz creation software tends to OCR both en dashes (-) and em dashes (—) as hyphens (en dashes).

Granted, I've read so many darknet books with glaringly bad OCR errors, that the few mistakes that seem to be in Topaz books are pretty minor to me.

[phenomshel]

Originally Posted by clarknova

The dashes should be there, but the Topaz creation software tends to OCR both en dashes (-) and em dashes (—) as hyphens (en dashes).

Granted, I've read so many darknet books with glaringly bad OCR errors, that the few mistakes that seem to be in Topaz books are pretty minor to me.

Yeah, exactly. A purchased book that I'm reading right now has so many glaring OCR errors that it's painful to read. Not to mention, it's a LOT worse than the output I'm getting with these scripts. So I can definitely handle mixups with en dashes and em dashes.

[orwell2k]

Originally Posted by phenomshel

Yeah, exactly. A purchased book that I'm reading right now has so many glaring OCR errors that it's painful to read. Not to mention, it's a LOT worse than the output I'm getting with these scripts. So I can definitely handle mixups with en dashes and em dashes.

Make sure you submit a complaint to the publisher/distributor where you bought the book.

I have had numerous instances where there have been terrible OCR errors throughout the book. I have complained, submitted various screenshots of the errors, and demanded a corrected copy or a refund (e.g. open the ePub in ADE and take a few screen caps - I use ePub here because they seem to be the main culprits, as I don't recall so many badly formatted Mobi books). Naturally, a refund is always forthcoming (rather than a better copy - 'cos there aint one!). I also point out that if I bought a paper book with these kind of errors I would return it to the publisher immediately.

The only distributor so far who is dragging their feet is Waterstones (surprise, surprise) - but so far the others seem to be playing the game (Fictionwise, BooksOnBoard, Diesel, Penguin UK, WH Smith (UK)).

Seriously, this kind of shoddy product is just not on, especially given the still-inflated eBook prices of most places... we should be getting pristine eBook editions for those prices, but instead we're getting a lot of crappy scan-and-OCR versions full of typos and recognition errors. The only way to get these people in line is to exercise what limited consumer power we have, and return every purchased book that is not up to scratch for a refund... maybe then they'll get the message?

[omk3]

Orwell2k, I had exactly the same experience! I am now in my third email to waterstones (about the same book), and only getting generic "we care about you" responses. I bought a book with a lot of french names and phrases inside, and they were all completely unreadable. I sent them a very characteristic screenshot - I actually would have given me double the book price back if I was in their place and saw the state that book was in...
Anyway, I regret not having complained before, because of course it was not the first book with errors, just the worst.
If it was a paper book we of course could have returned it at once, but it seems that returning an ebook is another of those rights that are not self-evident in the digital world

[nothatkind]

Thanks to all of you!!!
Specials thanks for phenomshel that help me a lot to understand how the scripts worked...
Now I can read my topaz file on my sony!!!
Since I'm outside US I can't buy ebooks from sony... Amazon is having a new costumer...