Problem converting a blog

watergeus · 11-11-2011, 04:52 PM

I'm trying to convert an existing blog which has 32 html files with pics (in subfiles) and an index. On the 7th subfile it falls over with the error message:

Running file type plugin HTML to ZIP failed with traceback

and simply adds one HTML file rather than a ZIP of the whole.

I do have the debug file which I could post but that's the key line. Anyone got any ideas? Each file fails the initial parse as it's not valid XHTML (e.g. IMG tag) but carries on ok up to the failure point. I've validated the HTML of the critical file using the W3C checker but it even fails when I try to add that file alone.

The failure happens just after it's added a file with encoded spaces: Cap%20de%20la%20Hague.jpg

Chris

watergeus · 11-11-2011, 05:14 PM

As always, having posted, I've now discovered the problem. In the img statement, the filename was encoded as:

src="LeConquet_files/Cap%2520de%2520la%2520Hague.jpg"

where the spaces should of course be coded as %20.

Next question: how do I report a bug?

Chris

Starson17 · 11-13-2011, 10:52 AM

Quote:

Originally Posted by watergeus

Next question: how do I report a bug?

Here:
https://bugs.launchpad.net/calibre

DiapDealer · 11-13-2011, 11:05 AM

Isn't %2520 usually indicative of a double url-encoded space?

Starson17 · 11-13-2011, 11:38 AM

Quote:

Originally Posted by DiapDealer

Isn't %2520 usually indicative of a double url-encoded space?

Yes, it is. (For others - %20, a singly encoded space, when encoded again encodes the % symbol as %25 giving %2520.)
I thought he had some other bug to report, and was saying his source had this problem, which he recognized. However, perhaps not.

DiapDealer · 11-13-2011, 12:12 PM

Yeah, I wasn't sure if he was saying calibre was producing the %2520 in one of its intermediate conversion steps or if it was already in his source files (produced by whatever program he used to scrape the entire blog to local files on his computer).

kovidgoyal · 11-13-2011, 10:24 PM

It was already in his source files and calibre was not handling it correctly, now fixed.

11-11-2011, 04:52 PM	#1
watergeus Junior Member Posts: 2 Karma: 10 Join Date: Nov 2011 Device: Kindle, Sony	Problem converting a blog I'm trying to convert an existing blog which has 32 html files with pics (in subfiles) and an index. On the 7th subfile it falls over with the error message: Running file type plugin HTML to ZIP failed with traceback and simply adds one HTML file rather than a ZIP of the whole. I do have the debug file which I could post but that's the key line. Anyone got any ideas? Each file fails the initial parse as it's not valid XHTML (e.g. IMG tag) but carries on ok up to the failure point. I've validated the HTML of the critical file using the W3C checker but it even fails when I try to add that file alone. The failure happens just after it's added a file with encoded spaces: Cap%20de%20la%20Hague.jpg Chris

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Problem with rss blog	xfiles	Recipes	1	07-21-2011 08:58 AM
Converting problem...	gmarco	Conversion	9	02-19-2011 12:03 PM
Problem converting to EPUB in 0.7.5	ould	Calibre	6	06-27-2010 08:12 AM
Converting blog novels to epub	mr ploppy	ePub	4	04-07-2010 12:52 PM
Problem Converting	Starfish07	Calibre	3	01-07-2010 06:14 AM

11-11-2011, 05:14 PM	#2
watergeus Junior Member Posts: 2 Karma: 10 Join Date: Nov 2011 Device: Kindle, Sony	As always, having posted, I've now discovered the problem. In the img statement, the filename was encoded as: src="LeConquet_files/Cap%2520de%2520la%2520Hague.jpg" where the spaces should of course be coded as %20. Next question: how do I report a bug? Chris

11-13-2011, 11:05 AM	#4
DiapDealer Grand Sorcerer Posts: 27,546 Karma: 193191846 Join Date: Jan 2010 Device: Nexus 7, Kindle Fire HD	Isn't %2520 usually indicative of a double url-encoded space?

11-13-2011, 12:12 PM	#6
DiapDealer Grand Sorcerer Posts: 27,546 Karma: 193191846 Join Date: Jan 2010 Device: Nexus 7, Kindle Fire HD	Yeah, I wasn't sure if he was saying calibre was producing the %2520 in one of its intermediate conversion steps or if it was already in his source files (produced by whatever program he used to scrape the entire blog to local files on his computer).

11-13-2011, 10:24 PM	#7
kovidgoyal creator of calibre Posts: 43,842 Karma: 22666666 Join Date: Oct 2006 Location: Mumbai, India Device: Various	It was already in his source files and calibre was not handling it correctly, now fixed.

Advert

Advert