EPUB output

[starrigger]

Originally Posted by Starson17

Is this an lrf problem only, or a conversion problem in general? Does the problem appear when going to epub from HTML, LIT or any other formats?

I believe the situation Phil_C described is a general problem. In my experience, Calibre seems to either insert blank lines or take them out, but never just leave the format alone. (I haven't made a close study of it, mind you, but I've seen it happen a lot.)

I was actually talking about a more subtle variation on the problem, which is when you want Calibre to remove blank lines between paragraphs because someone built an ebook that way, thinking it was desirable. (In Mobi Creator, for example, where that is one way it can come out.) But while you want those extra blank lines removed, you do not want original spaces between scenes removed, or spaces between title and text, and that kind of thing.

What I don't know is whether there's any way Calibre can distinguish between the two cases. I don't know if the markup generated to add the space between paragraphs is the same or different from the other. But it would be great if Calibre could do a "smart convert" where it identified places to leave the blank lines. One marker might be that centered text often needs to have a blank line before and after. Such as this...

***

Is that possible?

[kovidgoyal]

There is no reliable way to distinguish the cases, for example, centered text may or may not need a blank line before it (instead it could have a larger margin-top and margin-bottom), or the author may not have intended it to have spaces. While a given technique may work on some set of cases, it wont work on all, so rather than have "magic" behavior, I chose to keep it simple and predictable.

[Starson17]

Originally Posted by starrigger

I believe the situation Phil_C described is a general problem. In my experience, Calibre seems to either insert blank lines or take them out, but never just leave the format alone. (I haven't made a close study of it, mind you, but I've seen it happen a lot.)

Problem 1 description - It won't leave the blank lines alone.

I was actually talking about a more subtle variation on the problem, which is when you want Calibre to remove blank lines between paragraphs because someone built an ebook that way, thinking it was desirable. (In Mobi Creator, for example, where that is one way it can come out.) But while you want those extra blank lines removed, you do not want original spaces between scenes removed, or spaces between title and text, and that kind of thing.

Problem description 2 - You don't want Calibre to "leave the blank lines alone," but you want some left alone.

It's the difference between these two cases that confused me. You posted about what I saw as problem 2, then Phil_C said "Exactly!" and described something closer to problem 1.

To me, it seems that problem 2 is hard, while problem 1 isn't. Problem 2 needs "magic" code to detect what you want done (mind reading code?). Problem 1 just needs to spot an error in the conversion process, or at least that's what it looks like from the outside.

I was trying to figure out (if it's problem 1), is it generic or LRF only. If it's problem 1, and it's real, and it's not LRF only, then I'm wondering if it's a bug and easily fixable, or is there something deeper going on that I don't comprehend?

I don't see any way to address problem 2 in a way that will be consistent. OTOH, I'm interested in this if it fits into the "generic problem 1" box. I've been looking for something simple that interests me in the conversion code to get my feet wet in that area.

[Phil_C]

Originally Posted by Starson17

That was clear, and I appreciate it.



I still don't know if I can help. I particularly am not interested in working on lrf format (sorry, it's deprecated and I don't use it), but if your problem lies elsewhere, I might be interested. To find out, turn on debug and convert a problem file to EPUB. Then look in the debug folders and advise when the problem appears. Is it in the very first folder, or later? Is this an lrf problem only, or a conversion problem in general? Does the problem appear when going to epub from HTML, LIT or any other formats?

OK --- for lrf>epub, the Debug input and parsed folders contain exactly the same code whether or not the "Remove Spacing Between Paragraphs" option in Calibre is checked. Only the processed folder is different. Spaces are added between ALL paragraphs with the option unchecked, and removed when the option is checked.

So it appears that paragraph spaces are somehow added before the files appear in the input folder --- even though the spaces are NOT present in the lrf. (If I understand properly what I'm looking at.)

(For lit>epub, the result is more erratic. Much of the spacing is retained, but not all. For example, the blank line before a chapter title is retained, but the blank line after it is not. If there is a mid-chapter paragraph that is supposed to have a blank line before and after, those blank lines are retained. Meanwhile, page breaks after chapters are lost!)

I realize that lrf is dead as far as future use. But many of us with hundreds of lrf books surely would like a way to convert them to epub one time and be done with them.

Additional information: All of my lrf books were originally formatted in RTF, then finished in Book Designer and output as LRF. They are perfect at that point.

Thanks for your input...

[Phil_C]

Originally Posted by Starson17

I don't see any way to address problem 2 in a way that will be consistent. OTOH, I'm interested in this if it fits into the "generic problem 1" box. I've been looking for something simple that interests me in the conversion code to get my feet wet in that area.

Sorry, I missed the subtlety of Problem 2 in starrigger's post. That one would appear impossible to solve.

However Problem 1 seems to be something that can be addressed. Why can't the original spacing be left as is (all of it) during conversion to epub?

[starrigger]

Originally Posted by Starson17

Problem description 2 - You don't want Calibre to "leave the blank lines alone," but you want some left alone....

I don't see any way to address problem 2 in a way that will be consistent.

Perhaps there isn't. I thought possibly a different tagging system was used by the programs that create all the extra spaces between paragraphs. Something like the difference between <p> and <br>. (Or, as I often see now, <br />, though I don't know what the difference is between that and <br>.) Or maybe something more subtle. If that were the case, I thought reasonable guesses might be made that would remove the one but not the other. I didn't really think it was possible to achieve perfection, but then, what we have now isn't perfection, either.

Personally, I would find it useful if the program could do what I said above, and select on the basis of centered text--while knowing that it wouldn't get it right all the time.

In any case, I quite agree that there ought to be an option to simply leave blank line spaces alone. And that does seem a much easier problem.

[Pol]

I'm having an odd issue. Ran multiple files through the latest version of Calibre; when doing so I picked the output option of ePub for iPad. Three of the files, after conversion, show up perfectly on my Mac. I opened and checked them and can see all of the pages.

However, when I put them onto my iPad (through iTunes as I have done hundreds of others), the page count remains "2." Just the title page and no content.

The three files are from three different places, and of varying lengths (page count). I have tried the option of converting for "Sony," and for "iPad," all with the same result on these three files.

Has anyone else encountered this problem? I'm stumped. As I said, the original html files are complete, as are the ePub's on my Mac. It is only on my iPad that the remainder of the file is not there.

Thanks.
Polly

[JSWolf]

Originally Posted by starrigger

Don't know if this has been brought up before or not. Something that's been bugging me (in a minor way) is blank line spaces. Generally when I convert something to epub, I tell it to remove blank lines between paragraphs. (Or I might run an existing epub through to do that, if it was made with spaces between paragraphs.) It generally works well.

The problem is, there are some lines you want left blank: scene breaks, spaces between chapter number and text, etc. Is there any way of telling Calibre to leave these blanks alone? I'm not sure how it would know the difference, but Calibre is a very clever program, and I wondered if there was a way.

What I do is do the conversion without removing line spaces. Then I go into the ePub and edit the CSS to remove the line spaces. That way, I control what gets removed and it works most of the time without a problem.

The problem is that most eBooks have line spaces such as Mobipocket. And these get converted to ePub as full line spaces.

[moon face]

i have an issue:

i have some HTML files wana converted to EPUB but every file takes so mach time

i have around 180 files >> i leave it to convert after 2 days i sow that only 5 file are converted
any help>>??

it's an arabic file

[.rob.]

Ok, here's my problem.... I convert files (pdf's, lit's, and rtf's) to epub to read on my jetBook (not Lite). Now with the update of the jB to its newest version, the converted epubs no longer work. Also, some more recent converts to epub (I just converted 4 new lit files) will not work either.

>help<

[DoctorOhh]

Originally Posted by .rob.

Ok, here's my problem.... I convert files (pdf's, lit's, and rtf's) to epub to read on my jetBook (not Lite). Now with the update of the jB to its newest version, the converted epubs no longer work. Also, some more recent converts to epub (I just converted 4 new lit files) will not work either.

>help<

Calibre's device interface plugin was designed to put a # between author and title. The newest Jetbook update causes these books to no longer be readable.

You might be better off connecting to the Jetbook via the connect to folder option and adjusting the Save template under send to device to configure how you want the files to look.

Bottom line is removing the # symbol will allow you to read the books again.

[bayside]

Hello, I am a newbie to the world of epubs, metadata, and onix, etc. I just finishing exporting my book to an epub file with Adobe Indesign and submitted it to apple for ibook.
Now, I am attempting to send my ebook out to borders with kobo. Kobo is asking so much that must be in my metadata or it will be rejected. I am using Calibre and just updated the latest download.

These are the field that must be in the metadata
1.eISBN:
2.Contributor(s):
3.Description: Full description of eBook as it should appear on site.
4.Publisher:
5.Language:
6.Account Holder ID:
7.Monetization
8.Number of Free Chapters: The number of chapters (after front matter, known as "Openers") that you would like to offer as a free sample. "1" chapter is the default offering.
9.Licensing: A one-digit number (0, 1, 2 or 3) which indicates licensing for the title. Please refer to the appendix section of Kobo’s METADATA template for information.
10.Territorial Rights (Country Code): Leave blank to indicate world rights or enter 2-letter codes for territories in which you have the rights to sell your content
11. SRP: Suggested List Price (SLP). Kobo requires, at the very least, one of the provided currency fields to be filled out. This data needs to be formatted as a number with two decimals and no dollar sign.
12. Publication Date: Publication date in format YYYY-MM-DD
13. BISAC Code: At least one BISAC code is required. Enter the primary BISAC code for the title here. Use columns S and T to enter up to 2 additional BISAC codes if desired. You can find a list of BISAC codes

Now, not all of these are in the are Calibre. I can't add the editor, the illustrator, etc.
Nor is there a section for BISAC code.
How does one add these to the xml file or the onix file that Needs to be sent with the epub file to kobo? Can this be done in excel. If so, who?

[mwbiggs]

I'm am new to Calibre and attempting to convert a Word doc file to epub for ipad. Everything seems to work fine, but when I run it through epubcheck, I get like 1400 errors. It is mostly the same errors over and over again. Like below. I saved as a filtered web page from Word 2007. Then imported in Calibre. I could edit the epub files manually to remove the errors, but some errors are vague and simply say "needed resource missing". What am I doing wrong? I am using the latest version of Calibre.

5. ERROR: The Case for Lilith 1st ed v021 epub10 - Mark Biggs.epub/The Case for Lilith_1st_ed_v021_epub10_split_001.htm(207): attribute "link" not allowed at this point; ignored
6. ERROR: The Case for Lilith 1st ed v021 epub10 - Mark Biggs.epub/The Case for Lilith_1st_ed_v021_epub10_split_001.htm(207): attribute "vlink" not allowed at this point; ignored

[mtravellerh]

Originally Posted by mwbiggs

I'm am new to Calibre and attempting to convert a Word doc file to epub for ipad. Everything seems to work fine, but when I run it through epubcheck, I get like 1400 errors. It is mostly the same errors over and over again. Like below. I saved as a filtered web page from Word 2007. Then imported in Calibre. I could edit the epub files manually to remove the errors, but some errors are vague and simply say "needed resource missing". What am I doing wrong? I am using the latest version of Calibre.

5. ERROR: The Case for Lilith 1st ed v021 epub10 - Mark Biggs.epub/The Case for Lilith_1st_ed_v021_epub10_split_001.htm(207): attribute "link" not allowed at this point; ignored
6. ERROR: The Case for Lilith 1st ed v021 epub10 - Mark Biggs.epub/The Case for Lilith_1st_ed_v021_epub10_split_001.htm(207): attribute "vlink" not allowed at this point; ignored

Just ignore epubcheck, I'd say. As long as the epubs work, no sweat.

[kovidgoyal]

calibre's output will only pass epubcheck if the input html would also pass epubcheck. Basically you need to clean up your input html to make it XHTML 1.1 compliant. I believe tools exist to help you do that (though I haven't used any personally), html tidy comes to mind.