Thanks for the quick reply, Jon. It really needs to be something which doesn't require borrowing, because, as far as I know, the source for those will always be unavailable.
For this example, the directory of files is: https://archive.org/download/billgalactichero00harr and the OCR text (what I call the "source") is: https://archive.org/download/billgalactichero00harr/billgalactichero00harr_abbyy.gz which gives a permission violation when you attempt to download it. Ditto with the epub: https://archive.org/download/billgalactichero00harr/billgalactichero00harr.epub I really need a test case that is publicly accessible. Of course, they IA staff has access to everything, so they could fix the bug themselves, but for a community contributed fix we'll need a public test case to have much hope of finding it. Also, any patterns that you've noticed about when pages are missing and when they're not would be helpful. Tom On Thu, Nov 5, 2015 at 8:26 PM, Jon Leech <oddh...@sonic.net> wrote: > On Thu, Nov 05, 2015 at 08:07:02PM -0500, Tom Morris wrote: > > On Mon, Sep 28, 2015 at 2:28 AM, Greg Lindahl <lind...@pbm.com> wrote: > > > On Fri, Sep 25, 2015 at 05:06:13PM -0700, Jon Leech wrote: > > > > > > > Just for comparison, though, here's an example: > > > > > > > > https://openlibrary.org/books/OL7433769M/Bill_the_Galactic_Hero > > > > > > The code which does the conversion is open, and hasn't been updated > > > since 2011. See https://github.com/internetarchive/epub > > > > > > Perhaps we can dig up a Python-literate volunteer who'd like to look > > > into it? > > > > > > > If it gets added to the repo's issue tracker ( > > https://github.com/internetarchive/epub/issues), preferably with > examples > > where both the source and generated epub are available, I'll take a look > > when I have time. This example requires the epub to be borrowed and the > > source XML isn't available at all. > > I can give a few more examples of epubs I previously borrowed that I > noted were missing pages - but I don't think the public catalog says > what the source XML is, or whether it's available for a title, does it? > > Would there be any point to my filing a github issue with more > sample titles, in the hope one of them had the source available for > someone who has the internal OL access to make use of it? That seems to > be the blocking point. I'm actually willing to take a look at the epub > generation code myself, but without the input documents to refer to, > would have no way of diagnosing the problem. > > Jon > _______________________________________________ > Ol-tech mailing list > ol-tech@archive.org > http://mail.archive.org/cgi-bin/mailman/listinfo/ol-tech > Archives: http://www.mail-archive.com/ol-tech@archive.org/ > To unsubscribe from this mailing list, send email to > ol-tech-unsubscr...@archive.org >
_______________________________________________ Ol-tech mailing list ol-tech@archive.org http://mail.archive.org/cgi-bin/mailman/listinfo/ol-tech Archives: http://www.mail-archive.com/ol-tech@archive.org/ To unsubscribe from this mailing list, send email to ol-tech-unsubscr...@archive.org
