Is there a way to detect buggy pdfs without manually checking each pdf?

MarjaE · 03-27-2020, 03:24 PM

Some pdfs have corrupt text encoding to begin with. I have a pre-process pdfs for my Kindle. Some pdfs end up with corrupt text encoding after pre-processing in Ghostscript.

If I select text from these pdfs, I get either gibberish, or blank spaces punctuated with ... well, occasional punctuation.

I usually find this out by trying to search in a pdf, or by selecting text in a pdf. Is there an easy way to detect pdfs with malformed or missing text, without manually opening and selecting passages from each pdf?

Quoth · 03-27-2020, 05:20 PM

No.
This is why I avoid them and only have ones needed for documentation and read them on a 10" Tablet.

Life is too short. If I was immortal I might run OCR on the images and proof them.

Once or twice I've fixed up PDFs of rarer old books totally unavailable to be able to use them on a 7" eink.

MarjaE · 03-27-2020, 05:58 PM

I don't get to choose what formats other people publish in, so I can't avoid pdf.

Quoth · 03-28-2020, 06:18 AM

Quote:

Originally Posted by MarjaE

I don't get to choose what formats other people publish in, so I can't avoid pdf.

Then use a 10" or better tablet for those. A decent one is cheaper than many 6" eink and can be close to half the price of an 8" eink.

MarjaE · 03-28-2020, 01:47 PM

Quote:

Originally Posted by Quoth

Then use a 10" or better tablet for those. A decent one is cheaper than many 6" eink and can be close to half the price of an 8" eink.

Do you know of any 10" or better non-touchscreen tablet with an e-ink screen, button-based controls, and preferably a keyboard?

Because I have coordination problems, and can't use touch devices, as well as visual processing problems, and can't see very bright screens, and get get migraines from flashing, zooming, etc. animation.

On my computer, I pick software based on my ability to avoid too much animation, avoid blinking cursors, etc. I use Firefox and Waterfox with about:config hacks, user css, and add-ons to try to block as much problematic animation as possible, but still struggle. I use a Benq flicker-free monitor at 0% brightness, 30% contrast, 10% red, 20% green, 10% blue, because standard brightness ranges are too bright, red light is most likely to trigger seizures and migraines, and blue light is often said to be most likely to trigger eye strain.

Even compared with these extreme settings, e-ink is easier for me than glowing screens. Even with the occasional flashes during page reload, as long as I'm not flipping through things too quickly, it's still easier for me than glowing screens.

MarjaE · 03-29-2020, 11:45 PM

The old Iriver can load many of the Kindle-unreadable pdfs. Although it can't display jpx images in them. Librerator can as well. Kpv is supposed to be better, but I don't have the coordination to run it.

k2pdfopt is still a good way to convert scanned pdfs. Ghostscript lets me convert jpx images, and if it weren't for the trade-off with losing text, I'd just keep using it.

03-27-2020, 03:24 PM	#1
MarjaE Guru Posts: 929 Karma: 53902736 Join Date: Jun 2015 Device: multiple	Is there a way to detect buggy pdfs without manually checking each pdf? Some pdfs have corrupt text encoding to begin with. I have a pre-process pdfs for my Kindle. Some pdfs end up with corrupt text encoding after pre-processing in Ghostscript. If I select text from these pdfs, I get either gibberish, or blank spaces punctuated with ... well, occasional punctuation. I usually find this out by trying to search in a pdf, or by selecting text in a pdf. Is there an easy way to detect pdfs with malformed or missing text, without manually opening and selecting passages from each pdf? Last edited by MarjaE; 03-27-2020 at 04:07 PM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Easy way to check for pdfs with no text or buggy text?	MarjaE	Library Management	2	03-23-2020 04:48 PM
Adding separate PDFs into one large pdf	bradje	Calibre	5	10-13-2017 05:17 AM
Not able to detect pdf	xcuseme_1978	Kobo Reader	1	12-24-2015 04:10 AM
Combining several PDFs into one PDF with chapter breaks between each one	kerrypolka	Conversion	0	10-21-2012 08:02 AM
PDF to MOBI conversion - unable to detect any words	qwerty123456	Calibre	1	07-22-2010 07:54 AM

03-27-2020, 05:20 PM	#2
Quoth the rook, bossing Never. Posts: 12,283 Karma: 89822819 Join Date: Jun 2017 Location: Ireland Device: All 4 Kinds: epub eink, Kindle, android eink, NxtPaper11	No. This is why I avoid them and only have ones needed for documentation and read them on a 10" Tablet. Life is too short. If I was immortal I might run OCR on the images and proof them. Once or twice I've fixed up PDFs of rarer old books totally unavailable to be able to use them on a 7" eink.

03-27-2020, 05:58 PM	#3
MarjaE Guru Posts: 929 Karma: 53902736 Join Date: Jun 2015 Device: multiple	I don't get to choose what formats other people publish in, so I can't avoid pdf.

03-29-2020, 11:45 PM	#6
MarjaE Guru Posts: 929 Karma: 53902736 Join Date: Jun 2015 Device: multiple	The old Iriver can load many of the Kindle-unreadable pdfs. Although it can't display jpx images in them. Librerator can as well. Kpv is supposed to be better, but I don't have the coordination to run it. k2pdfopt is still a good way to convert scanned pdfs. Ghostscript lets me convert jpx images, and if it weren't for the trade-off with losing text, I'd just keep using it.