Alternately, the latest release has a new way to make arc files. Since I was repurposing the new bookmarks code, I added a button to extract bookmarks into a csv, which is in the right format for an arclist. So you can bookmark files manually, and export your own csv.

Also if you go the javascript injection route, there's a new Register link added to the login page.

So the latest arc format is fairly different, so hopefully the new version plays nicer. If not, I'll try to get more verbose with error messages.

Instead of "id,label", it's gotten a bit more complex:

"href","onclick","img src","label" (all from the element, not the page it's on)
With those 4 pieces, it makes a full .cellcontainer element, rather than copying it from elsewhere.

I posted my latest, see what it gives you.

Picture edit The new stuff turns this:

Into this:

Sadly no, without writing something for server-side storage, this'll remain a browser-specific feature.

Do virus scanners flag on filiename alone? I can rename it, I just didn't figure that would be an issue. :)

Edit: new copy's up, now it's entrypoint,js

So after hitting a wall with how to get more functionality out of this, I resorted to editing Ubooquity's built in tools,js. I added some to the end of it that, with a new inject.js in the theme folder, adds theme support to New/Search/Pages 2+.

This lead to the need for a redesign for how Bookmarks work, since New/Search pages don't have a unique page id that would be static related to where the comics are. So I took that in a different direction, and switched from linking the elements to copying them outright. This means that bookmarks (and later story arcs) load the pages like they would normally, rather than scraping others. Loads a lot faster/cleaner too.

I've got a new PageBuilder I'll release soon that generates the new style of story arc csv files. There's a convert.sh that can take the extracted source of an older-style arc, and save it as a newer style arc csv.

To install the code injection stuff, the tools.js in comixology.zip needs to be put into Ubooquity.jar. If you're not using a proxy prefix, the path to your theme folder will need to be edited in there first. You'll need to stop your server, replace with the new modified jar, and restart it. Once that's up again, edit the proxyPrefix in theme.js and inject.js and you're set.

I think that's another case of it looking for the hardcoded proxyPrefix. I just posted a new copy, where everything's in theme.js and it's looking for the hardcoded base hash (since apparently that's not user-specific). Should only be the one spot that should need to be edited for it to work.

Try the newest version, it parses out filenames. :)

I've actually been bouncing around an idea for that, since the filename is in the thumbnail image path. I'll see what I can do. :)

Are your comics tagged with names, or do they show in Ubooquity in a comicname number year-ish format?
Does it get the initial publisher id, or return a zero right from the beginning?

Is there any error in the javascript console? (More Tools > Developer Tools in Chrome)

I sorted it out, merged all of the .js files into theme.js and removed /ubooquity/ hardcoding for paths. Should sort out bookmarks and story arcs. New copy's posted now.