Re: [Lynx-dev] Alternative User-Agent header

[Travis Siegel]

For what it's worth, epub books are already html.  An epub file is just 
a zip file renamed to epub.

Simply extract the epub file into a directory of your choice, and 
generally (though not always) the html files are layed out in name 
order, and you can simply open each one in your browser, or if so 
inclined, just open lynx on the directory, and go down the list of html 
files reading them one at a time until you've finished the book.

With that said, there are times epub creators do stupid things with the 
html files and their names, such as making the names 40 characters long, 
and not numbering them in order (or at all), and in those cases, the 
only way to read things in order is to open the toc.ncx file (if there 
is one), or using the content.opf file for a complete reference to the 
html layout and order.  I prefer toc.ncx files when available, because 
they are much simpler than content.opf files, and it's much easier to 
pick out the files in order.  The only caveat to that is that some epub 
creation programs don't produce proper toc.ncx files, and in those 
cases, you generally get only a single entry in the table of contents 
file, which doesn't help anybody, so no clue why the programs generate 
such broken content files, but there it is.

90 percent of the time, you can simply extract the epub file, then load 
the files in order in your browser to read them as desired.  It's the 
other 10 percent of the time that requires an actual epub reader to 
extract things in the proper order.

For what it's worth, I did write an Epub reader, and you can find it at:

http://www.softcon.com/files/softconreader.exe

It doesn't handle content.opf files, so if your toc.ncx file is broken, 
you'll only be able to read the chapters in the book by typing them in 
on the file line, but as long as your toc.ncx file is properly 
formatted, the reader will work fine.

I have another version that does work with content.opf files, but 
there's still too many bugs in the opf parsing to release that version, 
so it remains in testing mode for the moment.

The only issue with that epub reader is that sometimes (for reasons 
unknown to me), the html parser I use to display the text drops 
shortcuts onto your desktop, without asking for permission to do so, nor 
telling you it did so, so sometimes you'll find hundreds of html 
shortcuts on your desktop.  The fastest way to remove these is just to 
go into your desktop folder and remove all the .htm or .html or .xhtml 
files.  Unfortunately, this is a result of the html interpreter built 
into windows, and there isn't anything I can do about this that I've 
discovered yet, but considering it's the only major bug I've found with 
it's operation, I'll take it.

There is one other bug I can't solve, though I've tried, and that is 
that if there is a vector graphics file in the epub content, (*.svg), 
and this file is referred to in one of the html files, it will hang with 
a error until you tell it no (alt-n) to trying to read/display the 
file.  Sometimes, it takes several iterations of this no answer before 
it goes away.

Otherwise, the reader works fairly well, and I personally use it for all 
of my epub reading, (sometimes generating toc.ncx files when they aren't 
present), but I plan to build a utility program that ships with the 
reader to fix broken toc.ncx files at some point in the future.

Sorry for the rambling message, but I hope it helps (someone). 
Softconreader is of course free, as are all of the programs in my files 
area.


On 6/26/2021 5:36 PM, Karen Lewellen wrote:
> While fichub.net is frankly fantastic, I get the epub files, and send 
> them  to epub2rtf@robobraille.org
> Which results in html that I convert  to text,  there is still an 
> important issue.
> First, one must have the fanfiction.net link to use fichub, which can 
> be challenging to discover.
> second, there is being able to check for story updates. I have never 
> been able to finish setting up a fanfiction.net account, but 
> bookmarked several user's m.fanfiction.net pages allowing me to check 
> for stories I was following...and read them too since the main site 
> used javascript on the next chapter link.
> That and I sometimes want to read a bit of a work before   sending 
> things to fichub.
> still, to be sure, it is far far better than nothing.
> I only browse at ao3, so browsing is less important for me then 
> keeping track of writers who either update works I love, or who have 
> not written for years then suddenly return.
> But that is me.
> Kare