ePub checkers versions (1.0.1 - 1.0.4)

[bcogney]

hello,

I am testing epub checkers by google on some epub files I currently have. The main reason I ran this test is to check if I get different errors from running the same epub file with differenct checkers versions. And I did. But I was suprised from my results. I got more errors when I check files using version 1.0.3 more than what I get when I check using version 1.0.1. Plus the errors kinf of throw me off, which something else I would need some help on and suggestions.
Version 1.0.3 complains about attribute "page-map" & "width"not allowed.
Version 1.0.1 only complains about attribute "width"?
I thought 1.0.3 would have less stricter rules than 1.0.1? Also, Will these errors prevent the epub files from DRM? If so, how can I fix them?

[bcogney]

Is this the right forum for epub questions?

[zelda_pinwheel]

hi bcogney, and welcome to the forum ! this is indeed the right forum for questions about epub, hopefully you'll get a few more responses soon. sometimes people don't have a chance to reply right away, give them some time .

i can't tell you much about the specific differences between the versions you're using, although i would expect that more recent versions of *any* tool would be more precise and more sophisticated, which could explain some of the differences you're seeing. also, since the epub format is itself still evolving, that could explain some of the differences as newer versions might take into account elements which have only recently been defined in the standard.

when you say "page-map", are you referring to Adobe's page-map ? that should be supported...

personally i use the online checker at threepress. it will give you a detailed error report which is very helpful. little tip, it sometimes has trouble with filenames containing a comma, so if in doubt, use a very simple (one word) filename for testing purposes.

as for your question about drm, i suspect that that is an entirely separate issue, however in any case, you should always make valid files for compatibility and display reasons.

can i ask why you are concerned about drm ? if you're planning to distribute books with drm, i'd like to suggest you do a bit of research around here first ; drm is generally considered to be inefficient at preventing illicit copies, expensive (a cost which gets passed along to the customer, raising prices overall), and frequently a source of frustration, inconvenience, or unfair limitation / loss of access to *legitimate customers*. and since the examples of non-drm methods are convincingly successful (see : previous transitions to digital media in the film and music industry who, after years of doggedly clinging to drm and many many stories of legitimate customers losing access to their media because of it, have finally abandoned drm, as well as examples of publishers selling drm-free ebooks like baen books, pan macmillan, or o'reilly). given all that it seems rather pointless to continue using drm !

[bcogney]

Thanks Zelda for your comments. I agree with you on the DRM part, but it was just an option.

For page-map, I meant page-map element referenced in OPF file, and I do have a page-map.xml file.

Also, the checker complaints about width refernced in XHTML file? and suggestion on these items would be great.

Also, if I want it to make changes to the XHTML file and rebuild it, what applications would you suggest using.

[zelda_pinwheel]

Originally Posted by bcogney

Thanks Zelda for your comments. I agree with you on the DRM part, but it was just an option.

For page-map, I meant page-map element referenced in OPF file, and I do have a page-map.xml file.

Also, the checker complaints about width refernced in XHTML file? and suggestion on these items would be great.

Also, if I want it to make changes to the XHTML file and rebuild it, what applications would you suggest using.

it's hard to answer your questions about the specific errors without having the code and the error report to look at ; can i suggest you check it on the threepress epubcheck and paste the error report here ? that would make it a lot easier.

the page-map.xml file should be supported ; what error are you getting exactly ?

for correcting the xhtml files, you can use any text editor or dreamweaver, etc. just unzip the epub file, edit the files, and then replace them in the epub (it's just a .zip container file). there is also an application called azardi for doing precisely that, you might want to take a look, however i've not used it myself.

[bcogney]

here's the error I get.

ERROR: ISBN_EPUB.epub/OPS/package.opf(56): attribute "page-map" not allowed at this point; ignored

I tried zipping and unzipping epub file, it's not working. am I missing anything.

I haven't tried azardi yet!

[zelda_pinwheel]

hm. can you show us line 56 from your package.opf ? sounds like a syntax error, maybe ?

what is not working about unzipping epub file ? do you mean you were not able to ? the simplest way, is to change manually the file extension to .zip, then double click. or, you can right click, "open with" then select winrar / winzip / etc.

[bcogney]

ok...after trials and error...I figured out how to re package epub files. I used winzip in windows.

[bcogney]

as far as the error, let me correct myself, this is a warning. page-map is used for refernce linking and index linking and references pagemap.xml.
For now, it's ignored.
checking these epub files, I am wondering, does the checkers check for everything? I mean, does it validate the structure? are there things that the checker will pass, and it's an error and can cause unwanted results.
I figured you are not a big fan of DRM, I am not either. But it's an option and I would like to explore it. So, if you or someone else can guide me to test some DRM that would be great.

[kovidgoyal]

The checkers are largely useless, there is a whole host of things that could be wrong with epub files that would prevent them from displaying in various readers that the checkers wont catch. On the other hand there is another host of things that the checkers will complain about that wont cause problems in any EPUB reader.

Still, fixing all the things that they complain about wont hurt, if you have the time and patience for it.

[bcogney]

Kovid,

This all sounds good. Since I am new to ePub, I am interested to know what the checkers will miss and what checkers might pass that the readers will complain about? Also, Are there tools/suggestions on anything else beside these "useless" checkers to help validate ePub files?

[Jellby]

I don't think they are useless, Kovid. I write ePUBs manually, and the checkers help me find errors like missing closing tags, duplicate ids or out-of-sync .opf files.

[bcogney]

Where can I find a detailed a list of errors that chckers can find, Jellby?

[kovidgoyal]

You can use any standard XML validater to validate OPF, NCX and XHTML files against a DTD (that's really all that epubchecker does anyway).

I just think that releasing tools called "EPUB checkers" that don't guarantee anything at all is rather counterproductive. People new to them are going to expect that if the tool passes their EPUB file, the file is guarateed to work with EPUB supporting readers. Unfortunately, that is not the case at all.

@bcogney: there is no tool that will check everything in an EPUB file. What I do is usually run hand crafted EPUB files through calibre as it does a fair bit to try to output EPUB files that will actually work on common EPUB readers. Another strategy is to manually check the files with ADE and leave it at that as most people consider it to be the "reference" implementation for EPUB readers and most ebook reader devices use it to render EPUB.

[Jellby]

Originally Posted by bcogney

Where can I find a detailed a list of errors that chckers can find, Jellby?

I don't know, but a (surely partial) list can be found here